Yield prediction program
The yield prediction program enhances yield forecasting accuracy by calculating harvestable dates and using solar radiation and dry matter production, allowing precise yield prediction for agricultural products.
Patent Information
- Application Number
- JP2020168458
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2020-10-05
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2040-10-05
AI Technical Summary
Existing methods for predicting agricultural product shipment volumes and yields lack accuracy, especially over short periods, making it difficult for wholesale markets to adjust prices and distribution effectively.
A yield prediction program that calculates the cumulative temperature after flowering, determines harvestable dates, and uses solar radiation and dry matter production to estimate the number of fruits, then predicts yield based on these factors.
Accurately predicts agricultural product yields daily and weekly, enabling wholesale markets to adjust prices and distribution strategies with high precision.
Smart Images

Figure 0007580748000004 
Figure 0007580748000005 
Figure 0007580748000006
Abstract
Description
[Technical field]
[0001] The present invention relates to a yield prediction program. [Background technology]
[0002] Fruit and vegetable production may be concentrated in one area, and the selling price may depend on the shipping volume of the production area. Therefore, wholesale markets and other places grasp the increase or decrease (trend) in the shipping volume of fruit and vegetable production areas as a whole, and limit the shipping volume to adjust the transaction price, and make early inquiries to secure the distribution volume of other production areas.
[0003] Conventionally, methods described in Patent Documents 1 and 2, etc., are known as methods for predicting shipping volumes for an entire production area. However, at present, there are issues with the accuracy of the predictions, and it is difficult to say that the trends in shipping volumes are accurately captured.
[0004] Methods for improving forecast accuracy include using forecasting tools that utilize AI and machine learning based on production characteristics and weather data for the entire production area (called "direct forecasting"), and predicting shipping volumes for individual farms or greenhouses, and aggregating shipping volume information for the entire production area (called "indirect forecasting"). Direct forecasting has low implementation and operating costs, but it is difficult to adjust shipping volumes for the entire production area. On the other hand, indirect forecasting has the advantage of making it easier to adjust shipping volumes for individual farms. [Prior art documents] [Patent documents]
[0005] [Patent Document 1] JP 2018-201417 A [Patent Document 2] JP 2019-170359 A [Patent Document 3] JP 2020-053039 A Summary of the Invention [Problem to be solved by the invention]
[0006] In indirect prediction, it is known that the yield can be predicted by using a material production model, for example, as in Patent Document 3. The use of a material production model allows for stable yield prediction, but from the perspective of improving the accuracy of short-term (daily, weekly, etc.) yield predictions, there is room for improvement in the parameters and prediction methods used for prediction.
[0007] The present invention aims to provide a yield prediction program capable of predicting agricultural crop yields with high accuracy. [Means for solving the problem]
[0008] The yield prediction program of the present invention is a yield prediction program for predicting the yield of an agricultural crop, When flowers of an inflorescence included in a stalk of a crop plant bloom, information for identifying the bloomed inflorescence and the flower and information on the flowering date are acquired and stored in a memory unit, and a day on which an accumulated temperature after the flowering date becomes equal to or greater than a predetermined value is stored in the memory unit as a harvestable date for the fruit of the bloomed flower, and by referring to the memory unit, a harvested fruit number which is the number of fruits that can be harvested on an i-th day which is any day after the planting date of the stalk of the plant, and a total fruit number which is the number of fruits that have bloomed and are attached to the stalk by the i-th day in all the inflorescences of the plant, are acquired, and a total dry matter production of the plant during the fruit growth period is calculated from an equation showing the relationship between the amount of solar radiation and the total dry matter production of the plant and the amount of solar radiation during the fruit growth period from the flowering date of the fruit that can be harvested on the i-th day to the i-th day, and a harvest ratio which is the ratio of the number of harvested fruits to the total number of fruits is calculated from a product of the total dry matter production of the plant during the fruit growth period and the fruit distribution rate divided by the fruit dry matter rate and a harvest ratio which is the ratio of the number of harvested fruits to the total number of fruits. This is a program that causes a computer to execute processing to predict yield. Effect of the Invention
[0009] The yield prediction program of the present invention has an effect of being able to predict the yield of agricultural crops with high accuracy. [Brief description of the drawings]
[0010] [Figure 1] FIG. 1 is a diagram illustrating a configuration of an agricultural system according to an embodiment. [Diagram 2] FIG. 2 is a diagram illustrating a hardware configuration of the information processing device in FIG. [Diagram 3] FIG. 2 is a functional block diagram of an information processing device. [Figure 4] 13 is a flowchart showing the process of a fruit condition prediction unit. [Diagram 5] FIG. 13 is a diagram showing an example of a management table (part 1). [Figure 6] FIG. 2 is a diagram showing an example of the management table (part 2). [Figure 7] 1 is a table showing the average number of fruits (flowers) per inflorescence. [Figure 8]FIG. 8(a) is a graph plotting the number of inflorescence tiers that can be harvested on each day, and FIG. 8(b) is a graph plotting the number of harvested inflorescences on each day. [Figure 9] 9(a) and 9(b) are diagrams for explaining a method of calculating daily light use efficiency in a comparative example. [Figure 10] 10(a) and 10(b) are diagrams for explaining a method for calculating daily light use efficiency in one embodiment. [Figure 11] This is a graph showing that the rate of leaf and stem production has a strong positive correlation with the amount of solar radiation. [Figure 12] FIG. 12(a) is a graph showing the results of calculating daily light use efficiency from daily average CO2 concentration using a method of one embodiment, and FIG. 12(b) is a graph showing the results of calculating daily total dry matter production using the daily light use efficiency calculated as in FIG. 12(a). [Figure 13] FIG. 13(a) is a graph showing the results of calculating daily leaf DMi using the method of one embodiment, and FIG. 13(b) is a graph showing the results of calculating daily stem DMi using the method of one embodiment. [Figure 14] 13(a) and 13(b). FIG. 13(b) is a graph showing the calculation results of daily fruit distribution rates calculated based on the results of FIG. 12(b), FIG. 13(a), and FIG. 13(b). [Figure 15] 9 is a graph showing the results of calculation of the harvest rate calculated from the data in FIG. 8(b). [Figure 16] FIG. 16 is a graph illustrating prediction error in one embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0011] An embodiment of an agricultural system will be described in detail below with reference to Fig. 1 to Fig. 16. Fig. 1 shows a schematic configuration of an agricultural system 100 according to an embodiment. The agricultural system 100 of this embodiment is a system for estimating yields in farms that cultivate crops (such as tomatoes) that produce fruit.
[0012] 1, the agricultural system 100 includes an information processing device 10, an outdoor sensor 12, and an indoor greenhouse sensor 14. The information processing device 10, the outdoor sensor 12, and the indoor greenhouse sensor 14 are connected via a network such as the Internet, enabling information to be exchanged between the devices.
[0013] The information processing device 10 is a device installed in a farm that cultivates agricultural products (eg, tomatoes) in a greenhouse 18, and estimates the number of tomatoes that can be harvested for each period (for each day or week in this embodiment) based on environmental information acquired by the outdoor sensor 12 and the greenhouse sensor 14, and information input by an operator. The information processing device 10 also predicts and displays the yield of tomatoes for each period (for each day or week) based on the number of tomatoes that can be harvested. Furthermore, the information processing device 10 is connected to a server operated by a wholesale market or the like via the Internet or the like, and transmits the yield prediction result of one farm to the server. The server aggregates the yield prediction results of the farms in the production area and analyzes the increase or decrease (trend) in the shipping volume of the entire production area. The configuration and processing of the information processing device 10 will be described in detail later.
[0014] The outdoor sensor 12 includes a temperature sensor that detects the air temperature outside the greenhouse 18 and a solar radiation sensor that detects solar radiation, and inputs the detection results to the information processing device 10.
[0015] The greenhouse sensors 14 include a temperature sensor that detects the air temperature inside the greenhouse 18, a solar radiation sensor that detects the solar radiation inside the greenhouse 18, and a CO2 concentration sensor that detects the CO2 concentration inside the greenhouse 18, and input the detection results to the information processing device 10.
[0016] Here, the configuration and processing of the information processing device 10 will be described in detail. FIG. 2 shows a schematic hardware configuration of the information processing device 10. As shown in FIG. 2, the information processing device 10 includes a CPU (Central Processing Unit) 90, a ROM (Read Only Memory) 92, a RAM (Random Access Memory) 94, a storage unit (e.g., a HDD (Hard Disk Drive)) 96, a network interface 97, a display unit 93, an input unit 95, and a drive for a portable storage medium 99. The display unit 93 includes a liquid crystal display, and the input unit 95 includes a keyboard, a mouse, a touch panel, and the like. These components of the information processing device 10 are connected to a bus 98. In the information processing device 10, the CPU 90 executes a program (including a yield prediction program) stored in the ROM 92 or the HDD 96, or a program (including a yield prediction program) read by the drive for a portable storage medium 99 from the portable storage medium 91, thereby realizing the functions of the components shown in FIG. 3. The functions of the various units in FIG. 3 may be implemented by an integrated circuit such as an Application Specific Integrated Circuit (ASIC) or a Field Programmable Gate Array (FPGA).
[0017] Fig. 3 shows a functional block diagram of the information processing device 10. In the information processing device 10, the CPU 90 executes a program to realize functions as an input receiving unit 30, an environmental information acquiring unit 32, a fruit condition predicting unit 34, a parameter setting unit 35, a yield predicting unit 36, and an information output unit 38, as shown in Fig. 3.
[0018] The input reception unit 30 acquires information input by the worker through the input unit 95 and transmits it to the fruit condition prediction unit 34 and the parameter setting unit 35. The information input by the worker includes information on which of the tomato plants cultivated in the greenhouse 18, which are predetermined survey targets, has bloomed and information on the number of bloomed inflorescence stages. The information on the bloomed plants and the number of inflorescence stages does not have to be input by the worker. For example, if a robot patrolling the greenhouse 18 can acquire information on the bloomed plants and the number of inflorescence stages using a camera or the like, the input reception unit 30 may acquire information input from the robot. The information input by the worker also includes information on the results of a destructive investigation of the tomato plants cultivated in the greenhouse 18. The destructive investigation is performed by the worker at a predetermined timing.
[0019] The environmental information acquisition unit 32 acquires the detection results of the outdoor sensor 12 and the greenhouse sensor 14, and transmits them to the fruit condition prediction unit 34 and the parameter setting unit 35. In addition, the environmental information acquisition unit 32 acquires past data (average values, etc.) and forecast data (weather forecast mesh, etc.) from an external server, etc., as future environmental information, and transmits them to the fruit condition prediction unit 34.
[0020] The fruit condition prediction unit 34 predicts how many fruits of each inflorescence of each plant can be harvested each day and each week, based on the information on the number of inflorescence stages received from the input reception unit 30 and the information received from the environmental information acquisition unit 32. Here, the fruit condition prediction unit 34 estimates that each fruit can be harvested when the accumulated temperature of each fruit (the accumulated value of the average temperature of each day since flowering) exceeds a predetermined threshold value (e.g., 800°C). The fruit condition prediction unit 34 transmits the predicted harvestable number to the yield prediction unit 36.
[0021] The parameter setting unit 35 calculates parameters (daily light use efficiency and daily fruit distribution rate, which will be described later) based on the destruction investigation results received from the input receiving unit 30 and information received from the environmental information acquiring unit 32. The parameters calculated by the parameter setting unit 35 are used in the yield predicting unit 36.
[0022] The yield prediction unit 36 uses the prediction result by the fruit condition prediction unit 34 and the parameters calculated by the parameter setting unit 35 to predict the yield for each plant and also predict the yield for one farm.
[0023] The information output unit 38 transmits to the server the prediction result of the yield prediction unit 36. The information output unit 38 may generate a screen for displaying the prediction result of the yield prediction unit 36 and display it on the display unit 93.
[0024] In addition, equipment for controlling the environment inside the greenhouse 18 is provided inside the greenhouse 18. Examples of such equipment include a heat pump, a ventilation window, a heater, a CO2 applicator, and a shading / heat-retaining curtain. The heat pump is equipment for lowering the temperature inside the greenhouse 18, and the ventilation window is a window that takes in outside air into the greenhouse 18. The heater is equipment for raising the temperature inside the greenhouse 18, and the CO2 applicator is equipment for adjusting the CO2 concentration inside the greenhouse 18. In addition, the shading / heat-retaining curtain is a curtain that adjusts the sunlight and temperature inside the greenhouse 18.
[0025] (Fruit condition prediction processing) Next, the fruit condition prediction process of the information processing device 10 will be described in detail with reference to the flowchart of FIG. 4 and other drawings as appropriate.
[0026] As a premise for the processing in Figure 4, it is assumed that the environmental information acquisition unit 32 acquires current environmental information inside and outside the greenhouse 18, past data (average values, etc.), and forecast data (weather forecast mesh, etc.), and transmits them to the fruit condition prediction unit 34.
[0027] In the process of FIG. 4, first, in step S10, the fruit condition prediction unit 34 waits until the flowering plant of the survey subject and the number of flowering inflorescence stages are input from the input reception unit 30. When the operator inputs the flowering plant and the number of flowering inflorescence stages, the input reception unit 30 acquires the input information and transmits it to the fruit condition prediction unit 34. When the fruit condition prediction unit 34 receives information from the input reception unit 30, it proceeds to step S12. In this embodiment, when the operator confirms flowering, it is assumed that the operator inputs the information on the day of flowering. Therefore, it can be said that the operator inputs the information that the plant has flowered and the date of flowering. In addition, when the operator inputs the flowering later, the information on the date of flowering can be input together with the information on the number of flowering inflorescence stages.
[0028] When the process proceeds to step S12, the fruit condition prediction unit 34 inputs information on the bloomed inflorescence into a management table prepared for each plant to be surveyed as shown in FIG. 5. Here, the management table in FIG. 5 is a table for managing the state of the fruit for the plant with Plant No.=1 in Area 1, as an example. The fruit condition prediction unit 34 predicts the daily average temperature based on past data and predicted data on the environmental information acquired from the environmental information acquisition unit 32, and inputs the predicted daily average temperature into the management table in FIG. 5. Therefore, the fruit condition prediction unit 34 predicts the temperature in the greenhouse 18 from the predicted temperature value, calculates the daily average temperature based on the predicted temperature, and inputs the predicted daily average temperature into the management table. The operator may determine in advance whether to set the temperature higher or lower.
[0029] In addition, in the management table of FIG. 5, information on "flowering confirmation" and "accumulated temperature" is managed for each flower of each inflorescence (first inflorescence-1, first inflorescence-2, ..., second inflorescence-1, second inflorescence-2, ...). For example, it is assumed that on the first day, it is input that one flowering occurred in the first inflorescence of the plant with Plant No. = 1 in Area 1 (step S10). In this case, in step S12, the fruit condition prediction unit 34 inputs "○" in the "flowering confirmation" column of "first inflorescence-1" in the row of the first day in the management table of FIG. 5, and inputs "0" in the "accumulated temperature" column of the first inflorescence. Note that, if multiple flowers of the first inflorescence are confirmed on the first day, the fruit condition prediction unit 34 inputs flowering confirmation information and accumulated temperature information to the management table in the same manner as described above.
[0030] Next, in step S14, the fruit condition prediction unit 34 predicts the condition of the input fruit based on the daily average temperature. Specifically, the fruit condition prediction unit 34 updates the management table as shown in FIG. 6. In this case, the fruit condition prediction unit 34 accumulates the daily average temperature for each day for the first fruit of the first inflorescence. Then, the fruit condition prediction unit 34 displays the value of the day when the accumulated temperature reaches, for example, 800°C or higher in gray (see the 50th day in FIG. 6). This grayed-out day indicates the day when the fruit can be harvested.
[0031] After that, steps S10 to S14 are repeatedly executed. This makes it possible to know the harvestable date of the fruits of each inflorescence of each plant. Also, from the table of Fig. 6, it is possible to know how much of the fruits of the whole plant can be harvested on each day and each week for each plant.
[0032] FIG. 7 is a table showing the average number of fruits (number of flowers) per inflorescence (average for plants surveyed by one farmer). FIG. 8(a) is a graph plotting the number of inflorescence tiers that can be harvested on each day, obtained by repeating the processes of steps S10 to S14 in FIG. 4. FIG. 8(b) is a graph plotting the number of fruits harvested from one plant on each day. By combining FIG. 8(a) and FIG. 8(b), it is possible to know how many fruits can be harvested from which tier of inflorescences on each day. Note that, unlike Dutch tomato varieties, common Japanese tomato varieties are designed to bloom in sequence within each inflorescence.
[0033] (Regarding the processing of the parameter setting unit 35) Next, a detailed description will be given of a calculation process of parameters (daily light use efficiency, daily fruit distribution rate) used for yield prediction, which is executed by the parameter setting unit 35.
[0034] (1) Calculating daily light efficiency The total dry matter production, which is the amount of photosynthesis (the sum of the dry matter production in leaves, stems, fruits, and roots), can be expressed by the following equation (1). Total dry matter production (g m -2 ) = Light reception rate (%) x solar radiation (MJ m -2 ) × Light utilization efficiency (g MJ -1 ) …(1)
[0035] The light interception rate is the proportion of light energy absorbed by the crops and depends on the state of vegetation. The light interception rate can be expressed by the following formula (2) using the light absorption coefficient and leaf area (LAI). Light receiving rate (%) = (1-exp(-1 × extinction coefficient × LAI)) ... (2)
[0036] where LAI is the leaf area of the crop per unit area, 4 (m 2 m -2) is considered to be a desirable value. Since the light absorption coefficient and LAI can be determined by non-destructive growth surveys, the light interception rate can be calculated daily. Note that light interception rate x solar radiation means the amount of light interception. Since the light interception rate can be calculated daily, the amount of light interception can also be calculated daily.
[0037] On the other hand, light use efficiency, which means the total dry matter production generated per 1 MJ of light received, was time-consuming to calculate. Specifically, many destructive surveys were conducted continuously during the cultivation period, and environmental information (e.g., CO2 concentration (CO2 j ) and daytime temperature) and the total dry matter production that increased during each destructive investigation. In addition, it was necessary to plot the data obtained during each destructive investigation period on a graph with the average environmental information for each destructive investigation period on the horizontal axis and the light utilization efficiency (= total dry matter production / average light receiving rate × accumulated solar radiation) calculated from the total dry matter production for each destructive investigation period on the vertical axis, to create a scatter diagram as shown in Figure 9(a). For example, to plot nine points as in Figure 9(a), ten destructive investigations were required. Furthermore, as shown in Figure 9(b), the relationship between environmental information and light utilization efficiency (LUE) was obtained by fitting a function using each point in Figure 9(a). s = a × ln(CO2 s )-b). Note that a and b are coefficients, and s means the number of days. And the relational expression in Figure 9(b): LUE s = a × ln(CO2 s The daily light utilization efficiency was calculated by substituting the daily average CO2 concentration into (a)-b. Thus, the method shown in Figure 9(a) and (b) required many destructive investigations to calculate the daily light utilization efficiency.
[0038] Therefore, in this embodiment, in order to calculate the daily light use efficiency, the light use efficiency LUE on the i-th day is calculated. i The following equation (3) was defined as an equation (first equation) including the above. Note that the maximum value of i, n, is the number of days from the previous destructive survey to the current destructive survey.
[0039]
number
[0040] where TDM is the total dry matter production (g m) accumulated within the destructive survey period (n days). -2 ) and the total dry matter production TDM obtained by the destructive survey performed at the end and start of the destructive survey period, as shown in Figure 10(a). t+1 , TDM t Difference (TDM t+1 -TDM t ) LAI i is the total leaf area of the crop per unit area on the i-th day obtained from the growth survey (m 2 m -2 ) PAR i is the solar radiation on the i-th day (MJ m -2 ), L.U.E. i is the light utilization efficiency on the i-th day (g MJ -1 ).
[0041] In addition, the light utilization efficiency (LUE) on the i-th day i For example, if the CO2 concentration is used as the environmental information, it can be expressed by the following formula (4) (second formula). i is the average CO2 concentration (ppm) during the day (e.g., 8:00-16:00), and a and b are unknown coefficients. LUE i = a × ln(CO2 i )-b …(4)
[0042] In this embodiment, the parameter setting unit 35 calculates the LAI for each day during the destructive investigation period. i , PAR i , CO2 i By using the above, the parameter setting unit 35 calculates optimal a and b so as to satisfy the above expressions (3) and (4). As a result, the parameter setting unit 35 calculates the LUE as shown in FIG. iand the daily average daytime CO2 concentration is obtained. Then, the parameter setting unit 35 calculates the light use efficiency of each day by substituting the average CO2 concentration of each day into the relational expression in FIG. 10(b) (the above formula (4) in which a and b are determined). As described above, in the case of this embodiment, the number of destructive inspections is only required to be twice, at the start timing and the end timing of the destructive inspection period (n days), so the number of destructive inspections can be reduced compared to the cases of FIG. 9(a) and FIG. 9(b).
[0043] In the above, the case where the average CO2 concentration is used as the environmental information has been described, but the present invention is not limited to this, and other environmental information, such as temperature, may be used. Also, an equation between a plurality of environmental information and light use efficiency may be obtained, and the daily light use efficiency may be calculated from the equation.
[0044] (2) Calculation of fruit distribution rate Next, a method for calculating the fruit distribution rate per unit time (for example, per day) will be described.
[0045] In this embodiment, in order to calculate the fruit distribution rate per unit time, (a) Estimate the amount of total dry matter production per unit time that is distributed to fruit per unit time (fruit dry matter production per unit time), (b) Fruit dry matter production per unit time divided by total dry matter production per unit time; The following process is executed.
[0046] In order to predict the yield of fruit vegetables such as tomatoes, the fruit partitioning ratio, which indicates the proportion of the total dry matter production that is allocated to the edible part of the fruit, is an important parameter. This fruit partitioning ratio fluctuates significantly from day to day depending on the plant vigor, etc.
[0047] In this embodiment, attention is focused on the fact that the growth amount of the fruit changes depending on the growth amount of the leaves and stems, and the fruit dry matter production amount is estimated taking into account the balance between the leaves, stems, and fruit.
[0048] Daily total dry matter production TDMi can be expressed as the following equation (5). TDM i = leaf DM i +Stem DM i + Fruit DM i +root DM i …(5)
[0049] In addition, leaf DM i is the daily leaf dry matter production and stem DM i is the daily stem dry matter production, and fruit DM i is the daily fruit dry matter production, and root DM i is the daily root dry matter production. i For total dry matter production TDM i It is assumed that the ratio is a certain percentage (for example, 5%) of
[0050] Here, the leaf and stem production rates have a strong positive correlation with the amount of solar radiation, as shown in Fig. 11(a) and Fig. 11(b). Similarly, the leaf and stem production rates also have a strong positive correlation with temperature. Therefore, in this embodiment, the daily dry matter production of leaves and stems per 1 MJ is estimated based on, for example, the amount of solar radiation. The calculated leaf dry matter production and stem dry matter production per 1 MJ are then multiplied by the daily solar radiation (MJ m -2 ) to reduce the daily leaf DM i (g m -2 ) and Daily Stem DM i (g m -2 ) can be obtained.
[0051] Therefore, the parameter setting unit 35 sets the daily TDM frequency obtained from the above formula (1). i and daily leaf DM calculated based on daily solar radiation. i and daily stem DM i And, NeDM i By substituting and into the above equation (5), the fruit DM per unit time is calculated as i The parameter setting unit 35 also calculates the fruit DM per unit time. i TDM per unit time i The daily fruit distribution rate is calculated by dividing by .
[0052] In this way, the parameter setting unit 35 can calculate the daily fruit distribution rate taking into account the cultivation environment and plant vigor.
[0053] (Yield prediction processing) Next, the yield prediction process performed by the yield prediction unit 36 will be described in detail.
[0054] In this embodiment, the daily yield is predicted using the "total dry matter production from flowering to harvest" and the "number (proportion) of fruits at harvest" of the fruits to be harvested.
[0055] The total dry matter production of the fruit can be expressed by the following equation (6). Total dry matter production of fruit (g m -2 ) = total dry matter production (g m -2 ) × Fruit distribution rate (%) ÷ Fruit dry matter rate (%) … (6)
[0056] Incidentally, the total dry matter production of fruits in the above formula (6) is the daily fruit weight accumulated in the entire crop plant (all fruits) on which multiple fruits are borne, and is not the fruit weight accumulated by the fruits that are actually harvested. In reality, the weight of the harvested fruits is the energy stored from when the flowers of a certain inflorescence bloom (including flower bud differentiation) until harvest. Therefore, in this embodiment, the number of flowers (number of fruits) of each inflorescence is grasped, and the daily yield is predicted by using the above-mentioned daily light use efficiency and fruit distribution rate.
[0057] As described above, in this embodiment, the harvest timing and harvest number of each inflorescence are known through processing by the fruit condition prediction unit 34 (see FIG. 8). In addition, the period from flowering to harvest of each fruit is also known (see FIG. 6). Therefore, the percentage (called the harvest percentage) of the number of fruits that can be harvested to the total number of flowers (total number of fruits) on one plant is calculated, and the daily yield is predicted by multiplying the total fruit yield by the harvest percentage.
[0058] The daily yield can be expressed by the following equation (7).
[0059]
number
[0060] In the above formula (7), the number of days from planting is i (days), the number of inflorescence stages is j, and the flowering date of the jth inflorescence is t j , the first harvest date of the jth inflorescence is T j , the number of days to harvest the jth inflorescence on harvest date i is d j (The first day is 1). Also, the fruit distribution rate is FR i (Fruit distribution rate), fruit dry matter rate FDR i (Fruit Dry Matter Rate), harvest rate CR i (Cultivation Rate), promoting product success rate i (Product Rate), Y i In the brackets in the above formula (7), the total dry matter production from flowering to harvest is multiplied by the fruit distribution rate and divided by the fruit dry matter ratio, so the amount of fruit during the period is Total It can be said to be production volume.
[0061] As mentioned above, Japanese tomato varieties flower and are harvested in sequence within the inflorescence. Therefore, the total fruit yield period starts at the beginning (t j In other words, the period from the day the harvest stage number changed to the flowering inflorescence period is fixed, and the starting point is shifted back by one day on the next day.
[0062] Here, the harvest rate CR i can be expressed by the following equation (8).
[0063]
number
[0064] In the above equation (8), P i is the number of pieces harvested on that day (pieces m -2 ), F jis the total number of flowers (total number of fruits) harvested on that day (pieces m -2 ), c means the number of inflorescence tiers on one plant.
[0065] The yield prediction unit 36 calculates the yield Y i At this time, the yield prediction unit 36 uses the daily light use efficiency calculated by the parameter setting unit 35 when calculating the daily total dry matter production TDM (see the above formula (1)). In addition, the yield prediction unit 36 uses the fruit distribution ratio FR i The daily fruit distribution rate calculated by the parameter setting unit 35 is used as the daily fruit distribution rate.
[0066] The yield prediction unit 36 also calculates the yield Y per plant obtained from the above formulas (7) and (8). i Based on this, the yield on the i-th day for one vinyl greenhouse and the yield on the i-th day for one farm are calculated. Then, the yield prediction unit 36 transmits the calculated information on the i-th day yield for one farm via the information output unit 38 to a server used by an organization such as a wholesale market. The server predicts the i-th day yield for the entire region by aggregating the yields of the farmers in the region. In addition, the server can predict the weekly yield for the entire region by aggregating the yields of the farmers in the region for, for example, one week. By displaying the prediction result, the wholesale market can impose restrictions on shipping volume to adjust the transaction price and make early inquiries on the farmers to ensure distribution volume from other production areas. The server may automatically control the equipment for controlling the environment in the greenhouse 18 of each farm based on the increase or decrease (trend) of the yield to adjust the yield of each farm.
[0067] In this embodiment, the information processing device 10 can calculate parameters (daily light use efficiency and daily fruit distribution rate) used for daily yield prediction by inputting the flowering information of each inflorescence of each plant, environmental information such as temperature, solar radiation, and carbon dioxide concentration, and the results of a minimum dissection investigation. In addition, the information processing device 10 uses the calculated daily light use efficiency and daily fruit distribution rate, and calculates the harvest ratio CR i Since the daily yield is predicted using the above formula, the daily yield can be predicted with high accuracy.
[0068] FIG. 12(a) shows the results of calculating the daily light utilization efficiency from the daily average CO2 concentration using the method described above. In FIG. 12(a), there are two destructive investigation periods, and the daily light utilization efficiency LUE was calculated using the relational equation (above equation (4)) obtained in each destructive investigation period. i The results of calculating the daily total dry matter production (TDM) are shown in Fig. 12(b). i The calculated results are shown.
[0069] Furthermore, Fig. 13(a) and Fig. 13(b) show the daily leaf DM obtained by the above-mentioned method. i , Daily Stem DM i The results of calculating the daily fruit distribution ratio based on the results of Figs. 12(b), 13(a), and 13(b) are shown in Fig. 14. Also, Fig. 15 shows the results of calculating the harvest ratio based on the data of Fig. 8(b).
[0070] As a result of performing weekly yield prediction using these data, the prediction error of the yield is small and prediction can be performed with high accuracy, as shown in FIG. 16. The prediction error means {(actual yield-predicted yield) / actual yield}×100(%). Furthermore, for agricultural crops such as tomatoes, the size and weight of the fruit change depending on the harvest time, even for the same plant. However, by performing yield prediction as in this embodiment, it is possible to predict daily and weekly yields with high accuracy, as shown in FIG. 16.
[0071] As described above in detail, according to this embodiment, the yield prediction unit 36 receives from the fruit condition prediction unit 34 the number of inflorescence stages (j) and the number of fruits (P i ) and the total number of fruits (total number of flowers) (F j The yield prediction unit 36 also obtains the total dry matter production amount for the period calculated based on the environmental information from the flowering date of the harvestable fruit to the i-th day, and the fruit distribution rate (FR i ) and fruit dry matter ratio (FDRi The yield prediction unit 36 then calculates the total dry matter production of the fruit from the total number of inflorescence tiers (c), the total number of fruits in the inflorescence tiers (j) (F j ), the number of harvestable fruits (P i ) and the harvest rate (CR i ) and the total dry matter production of the fruit, the yield on the i-th day is predicted (see formula (7)). In this manner, in this embodiment, the amount of photosynthetic products allocated to the fruit that can be harvested on the i-th day out of the photosynthetic products produced in the crop can be estimated as the yield, so that the yield can be predicted with high accuracy.
[0072] In the present embodiment, the parameter setting unit 35 calculates the total dry matter production amount obtained from the results of the dissection and inspection performed at the beginning and end (start and end) of the dissection and inspection period, the light receiving rate value and the amount of solar radiation value obtained for each unit time during the dissection and inspection period, and the light use efficiency value for each unit time by calculating the above formula (3) and the carbon dioxide concentration (CO2 i The relationship between the LAI and the light utilization efficiency is expressed by the above formula (4) using the coefficients a and b. i , PAR i , CO2 i Based on this, the coefficients a and b of the above formula (4) are determined. Then, the parameter setting unit 35 calculates the daily light use efficiency value by substituting the daily carbon dioxide concentration value into the above formula (4) for which the coefficient has been determined. This makes it possible to obtain an equation for calculating the daily light use efficiency from the results of two dismantling surveys conducted at the beginning and end of the dismantling survey period, thereby reducing the labor of workers and the cost required for the dismantling survey compared to deriving an equation using the results of multiple dismantling surveys.
[0073] In this embodiment, the parameter setting unit 35 calculates the leaf dry matter production (leaf DM) on the i-th day based on the environmental information (solar radiation amount and temperature). i ) and stem dry matter production (stem DM i) and subtracting the dry matter production of leaves, stems, and roots on the i-th day from the total dry matter production including leaves, stems, roots, and fruits on the i-th day to calculate the dry matter production of fruits on the i-th day. Then, the parameter setting unit 35 calculates the fruit distribution rate on the i-th day by dividing the dry matter production of fruits on the i-th day by the total dry matter production on the i-th day. In this way, in this embodiment, the fruit distribution rate can be accurately determined taking into account the plant vigor, etc.
[0074] In the above embodiment, the information processing device 10 executes all of the yield predictions considering the harvest rate, the yield prediction using the daily light use efficiency, and the yield prediction using the daily fruit distribution rate, but is not limited to this. That is, the information processing device 10 only needs to execute at least one of the above three yield predictions.
[0075] In addition, the daily fruit distribution rate FR i It can be said that the strength of the plant is expressed. For example, the daily fruit distribution rate FR i If the value is small, it means that the distribution of dry matter production to stems and leaves is increasing, and plant vigor is increasing. Therefore, the information processing device 10 may grasp the daily fruit distribution rate and perform environmental adjustment (for example, adjustment of CO2 concentration) so that the fruit distribution rate becomes a predetermined value. This makes it possible to stabilize the growth of agricultural crops and improve productivity.
[0076] The above processing functions can be realized by a computer. In that case, a program is provided that describes the processing contents of the functions that the processing device should have. The above processing functions are realized on the computer by executing the program on the computer. The program that describes the processing contents can be recorded on a computer-readable storage medium (excluding carrier waves).
[0077] When a program is distributed, it is sold in the form of a portable storage medium on which the program is recorded, such as a DVD (Digital Versatile Disc), a CD-ROM (Compact Disc Read Only Memory), etc. The program can also be stored in a storage device of a server computer and transferred from the server computer to other computers via a network.
[0078] A computer that executes a program stores, for example, a program recorded on a portable storage medium or a program transferred from a server computer in its own storage device. The computer then reads the program from its own storage device and executes processing in accordance with the program. The computer can also read the program directly from a portable storage medium and execute processing in accordance with that program. The computer can also execute processing in accordance with the received program each time a program is transferred from the server computer.
[0079] The above-described embodiment is a preferred example of the present invention, but the present invention is not limited to this embodiment and can be modified in various ways without departing from the scope of the present invention. [Explanation of symbols]
[0080] 34 Fruit Condition Prediction Section 35 Parameter setting section 36 Yield Forecasting Department 90 CPU (Computer)
Claims
1. A yield prediction program for predicting a yield of an agricultural crop, When the flowers of the inflorescence included in the stalk of the agricultural crop bloom, information for identifying the bloomed inflorescence and the flowers and information on the blooming date are acquired and stored in a storage unit; A day when the accumulated temperature after the flowering date becomes equal to or greater than a predetermined value is stored in the storage unit as a harvestable date of the fruit of the flower that has bloomed. With reference to the storage unit, a harvested fruit number is obtained, which is the number of fruits that can be harvested on an i-th day, which is any day after the planting date of the plant, and a total fruit number is obtained, which is the number of fruits that have bloomed and are attached to the plant by the i-th day in all inflorescences of the plant; Calculating the total dry matter production of the plant during the fruit growth period from an equation showing the relationship between the amount of solar radiation and the total dry matter production of the plant and the amount of solar radiation during the fruit growth period from the flowering date of the fruit that can be harvested on the i-th day to the i-th day; The yield of the plant on the i-th day is predicted from the product of the total dry matter production of the plant during the fruit growth period and the fruit distribution rate divided by the fruit dry matter rate and the harvest rate, which is the ratio of the number of harvested fruits to the total number of fruits. A yield prediction program that causes a computer to carry out the processing.
2. The computer is further caused to execute a process of calculating a value of light use efficiency for each predetermined time period, which is included in the formula showing the relationship between the amount of solar radiation and the total dry matter production amount of the plant; In the process of calculating the value of the light use efficiency for each predetermined time, determining the unknown coefficient based on: a first equation indicating that the total dry matter production amount during a predetermined first period obtained from the results of a destructive survey of the agricultural crops conducted at the beginning and end of the first period is equal to an integrated value during the first period of a product of the light receiving rate value per unit time obtained during the first period, the solar radiation value per unit time obtained during the first period, and the light use efficiency value per unit time during the first period; and a third equation expressing the light use efficiency per unit time during the first period with the unknown coefficient, which is obtained by substituting the carbon dioxide concentration value per unit time obtained during the first period into a second equation expressing the light use efficiency using the carbon dioxide concentration and an unknown coefficient; calculating a value of the light use efficiency for each predetermined time period by substituting the value of the carbon dioxide concentration for each predetermined time period into the second equation in which the coefficient has been determined; 2. The yield prediction program according to claim 1 .
3. The computer further executes a process of calculating the value of the fruit distribution ratio, In the process of calculating the value of the fruit distribution rate, calculating the leaf dry matter production amount of the plant per unit time from the product of the leaf dry matter production amount per amount of solar radiation and the solar radiation per unit time; calculating the stem dry matter production amount of the plant per unit time from the product of the stem dry matter production amount per amount of solar radiation and the solar radiation per unit time; and calculating the root dry matter production amount per unit time from the product of the total dry matter production amount of the plant per unit time and a predetermined ratio; The ratio of the total dry matter production of the plant per unit time minus the dry matter production of the leaves, stems, and roots of the plant per unit time to the total dry matter production of the plant per unit time is calculated as the value of the fruit distribution rate.
3. The yield prediction program according to claim 1 or 2.
4. A yield prediction program for predicting a yield of an agricultural crop, determining the unknown coefficient based on: a first equation indicating that the total dry matter production amount during a predetermined first period obtained from the results of a destructive survey of the agricultural crops conducted at the beginning and end of the first period is equal to an integrated value during the first period of a product of the light receiving rate value per unit time obtained during the first period, the solar radiation value per unit time obtained during the first period, and the light use efficiency value per unit time during the first period; and a third equation expressing the light use efficiency per unit time during the first period with the unknown coefficient, which is obtained by substituting the carbon dioxide concentration value per unit time obtained during the first period into a second equation expressing the light use efficiency using the carbon dioxide concentration and an unknown coefficient; Calculating a value of the light use efficiency for each predetermined time period by substituting the value of the carbon dioxide concentration for each predetermined time period into the second equation in which the coefficient has been determined; predicting a yield of the crop using the light use efficiency for each predetermined time period; A yield prediction program that causes a computer to carry out the processing.
5. A yield prediction program for predicting a yield of an agricultural crop, calculating a dry matter production amount of leaves in the plant per unit time from the product of the dry matter production amount of leaves contained in the plant per amount of solar radiation and the amount of solar radiation per unit time; calculating a dry matter production amount of stems in the plant per unit time from the product of the dry matter production amount of stems per amount of solar radiation and the amount of solar radiation per unit time; and calculating a dry matter production amount of roots per unit time from the product of the total dry matter production amount of the plant per unit time and a predetermined ratio; The ratio of the value obtained by subtracting the dry matter production amounts of the leaves, stems, and roots of the plant per unit time from the total dry matter production amount of the plant per unit time to the total dry matter production amount of the plant per unit time is calculated as the value of the fruit distribution rate per unit time; A yield of the agricultural crop is predicted using the fruit distribution rate per unit time. A yield prediction program that causes a computer to carry out the processing.
Citation Information
Patent Citations
Yield prediction system
JP2018201417A
Plant cultivation result prediction system
JP2019170359A
Harvest information output program, harvest information output method, and harvest information output device
JP2020053039A
