JP2015176339A - Device, method and program for prediction of crop harvest time - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a harvest time prediction device that improves the accuracy of prediction of the harvesting season and the harvest time by considering variations of data for use in prediction.SOLUTION: A harvest time prediction device 100 includes: a past temperature memory unit 10 for storing temperature data every day of the past multiple years; and a past temperature check unit 20 for reading stored temperature data for the N years, checking whether or not the temperature data falls in a range of the latitude of the Lindeberg condition, and excluding the temperature data that does not fall in the range of the latitude as a missing value. The past temperature check unit 20 supplements an average value of the temperature data of the same day of the same month of the year except the year when the missing value is present as the substitute value of the missing value when the missing value is present in the temperature data. The past average temperature is calculated in moving average to the temperature data after the missing value supplement. Based on the past average temperature, and the growth minimum temperature and the growth maximum temperature of the crops, an effective temperature is accumulated from the cultivation start date, and the day when the effective accumulated temperature exceeds a predetermined target effective accumulated temperature is output as a predictive harvest day.

Description

  The present invention relates to a method for predicting crop harvest time, and more particularly to a crop harvest time prediction device, a prediction method, and a prediction program using an effective integrated temperature method.

  Conventionally, “effective integrated temperature method” (also referred to as “effective integrated temperature method”) has been used as a crop harvesting time prediction technique (for example, Patent Documents 1 and 2 below). In this method, the lower limit of the temperature (temperature) required for growth is determined for each variety of crops, and the difference (effective temperature) between the lower limit and the daily temperature is integrated for the number of days of crop growth, and the integrated effective temperature is calculated. When the temperature reaches a certain temperature, it is determined that it is the harvest time.

JP 2007-310463 A JP 2013-191107 A

  However, the methods described in Patent Documents 1 and 2 do not sufficiently consider errors related to measurement, such as observation equipment and observation methods, and errors related to fluctuations in the temperature itself, with respect to weather data used for prediction. When predicting the harvest time of crops using the effective accumulated temperature method, the average temperature until the predicted date of harvest is unknown, so the daily average temperature of each future day is calculated using the daily average temperature of the same month in the past. The effective integrated temperature is calculated. However, the past daily average temperature data varies, the daily average temperature on the same day of the same month is missing due to equipment malfunction, etc., the past temperature data is too large or abnormal weather, etc. It may be a special value and cannot be used as it is.

  Therefore, in the present invention, in view of the above-described problems, a method for measuring the degree of variation in past data and determining the availability of past data, and when past data is missing, consider the degree of variation in past data. It is an object of the present invention to provide a harvesting time prediction device, a prediction method, and a prediction program with high crop prediction accuracy by estimating missing values.

  In order to solve the above problems, the harvest time prediction apparatus, the harvest time prediction method, and the harvest time prediction program of the present invention provide the following solution means.

(1) A harvest time prediction device for predicting the harvest time of a crop, the past temperature storage unit storing daily temperature data for a plurality of past years in the cultivation place of the crop,
When the temperature data stored in the past temperature storage unit is read out for a predetermined year, the case where the read temperature data is partially missing is stored as a missing value, and the temperature data is not missing Checks whether or not the temperature data is within a predetermined tolerance range of Lindeberg condition, and excludes the temperature data that does not fall within the tolerance range as a missing value, and the past temperature check unit If there is the missing value in the temperature data after the check by the temperature check unit, the missing value that complements the average value of the temperature data of the same month of the year excluding the year in which the missing value exists as an alternative value of the missing value A value complementing unit, a past average temperature calculating unit for calculating a past average temperature with respect to the temperature data after supplementing the missing value, the past average temperature and the minimum growth of the crop The effective temperature of the crop is integrated every day from the cultivation start date based on the temperature and the maximum growth temperature, and the day when the integrated effective integrated temperature exceeds a predetermined target effective integrated temperature And an effective temperature integrating unit that outputs as the day.

  (2) In the invention of (1), the past average temperature may be a moving average of a predetermined section.

(3) A method for predicting a harvest time for predicting the harvest time of a crop, wherein the computer stores the temperature data for each day of the past multiple years at the crop cultivation place, and stored in the storing step. When the temperature data is read out for a predetermined year and the read out temperature data is partially missing, the temperature data is stored as missing values. A step of checking whether or not it falls within a predetermined tolerance range of the Lindeberg condition; a step of excluding temperature data not falling within the tolerance range as a missing value; the step of checking and the step of excluding If there is the missing value in the temperature data after checking by, the same day of the same month in the year excluding the year in which the missing value exists Complementing an average value of temperature data as an alternative value of the missing value, calculating a past average temperature for the temperature data after the missing value is supplemented, and the past average temperature and the growth of the crop Based on the minimum temperature and the maximum growth temperature, the effective temperature of the crop is accumulated every day from the cultivation start date, and the day when the accumulated effective accumulated temperature exceeds a predetermined target effective accumulated temperature, And a step of outputting as a predicted harvest date.

(4) A harvest time prediction program for predicting the harvest time of crops, which is stored in a step of storing temperature data for each day in the past for a plurality of years in the cultivation place of the crop in the computer, and in the storing step. When the temperature data is read out for a predetermined year and the read out temperature data is partially missing, the temperature data is stored as missing values. A step of checking whether or not it falls within a predetermined tolerance range of the Lindeberg condition; a step of excluding temperature data not falling within the tolerance range as a missing value; the step of checking and the step of excluding In the same month of the year excluding the year in which the missing value exists when the missing value exists for the temperature data after checking by Complementing an average value of daily temperature data as a substitute value for the missing value, calculating a past average temperature for the temperature data after the missing value is supplemented, the past average temperature and the crop Based on the lowest growth temperature and the highest growth temperature, the step of integrating the effective temperature of the crops every day from the cultivation start date, and the day when the integrated effective integrated temperature exceeds a predetermined target effective integrated temperature, And outputting the predicted harvest date of the crop.

  According to the present invention, the accuracy of harvest time prediction by the effective integrated temperature method can be increased by increasing the accuracy of the effective temperature used for prediction.

It is a figure which shows the function structure of the harvest time prediction apparatus 100 which concerns on embodiment of this invention. It is a figure which shows the specific example of the past temperature data stored in the past temperature memory | storage part 10 which concerns on embodiment of this invention. It is a figure which shows the processing flow which the past temperature check part 20 which concerns on embodiment of this invention performs. It is a figure which shows the processing flow which the missing value complementation part 30 which concerns on embodiment of this invention performs. It is a figure which shows the processing flow which the past average temperature calculation part 40 which concerns on embodiment of this invention performs. It is a figure which shows the processing flow which the effective temperature integrating | accumulating part 60 which concerns on embodiment of this invention performs. It is a figure which shows the specific example of the summary data of the effective integrated temperature which the harvest time prediction apparatus 100 which concerns on embodiment of this invention outputs.

  DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments for carrying out the present invention (hereinafter referred to as embodiments) will be described in detail with reference to the accompanying drawings. In the subsequent drawings, the same numbers or symbols are assigned to the same elements throughout the description of the embodiments. In the functional configuration diagram, an arrow between functional blocks represents a data flow direction or a process flow direction.

(Functional configuration of harvest time prediction device)
FIG. 1 is a diagram showing an outline of a functional configuration of a harvest time prediction apparatus 100 (hereinafter referred to as “this apparatus”) according to an embodiment of the present invention. This device is a device for predicting the harvest time of agricultural products, and typically, as shown in the figure, a past temperature storage unit 10, a past temperature check unit 20, a missing value complement unit 30, and a past average temperature calculation unit 40. The past average temperature storage unit 50, the effective temperature integration unit 60, the predicted harvest date storage unit 70, the prediction condition input unit 80, and the display / output unit 90 are mainly configured. Hereinafter, each functional unit will be described in order.

  The past temperature storage unit 10 is a database that stores temperature data for a plurality of years in the cultivation place of the crop every year and every day. FIG. 2 shows a specific example of data stored in this database. “Nd” in the figure indicates that the temperature data for that day is missing for some reason. The daily temperature data may use the AMeDAS database of the Japan Meteorological Agency, but the data released by the Japan Meteorological Agency are the minimum and maximum temperatures of the day, and the average temperature from 0 to 24 hours (daily average). Therefore, in order to improve the accuracy of the daily average temperature, a database that measures the temperature of the day for each time zone is built, and the average temperature in the sunshine hours, which is the time zone in which plants grow, is calculated. You may make it employ | adopt.

  The past temperature check unit 20 reads past temperature data stored in the past temperature storage unit 10 and checks the degree of variation in the temperature data for a predetermined past N years. Specifically, an average and a deviation (standard deviation) are calculated for each temperature data of N years of the past temperature, and it is checked whether the temperature data satisfies the “Lindberg condition”. Here, the Lindeberg condition is a condition represented by the following formula (1), although a detailed description is omitted.

が成り立つとき、この条件のことを、リンデベルグ条件（Lindeberg's condition）と呼ぶ。上記の条件が所定値（＝１）に収束するかどうかは、Ｘ_ｎの２次モーメント量（偏差は２次モーメント量である）によって決まる。したがって、簡単に言えば、リンデベルグ条件とは、値のばらつき度合いをチェックする条件である。過去気温チェック部２０は、このリンデベルグ条件の所定の許容範囲を満たしていないデータは除外する（欠損値として扱う）。 Let random variables be X ₁ , X ₂ ,. . . X _n , the probability density function of the distribution followed by each random variable is f (x), mean E (X ₁ ) = 0, and variance V (X _i ) is finite.
Here, Z _n = X ₁ + X ₂ +... + X _n , V _n = V (Z _n ), and a condition for an arbitrary real number ε> 0

When this holds, this condition is called a Lindeberg's condition. Whether or not the above condition converges to a predetermined value (= 1) is determined by the secondary moment amount of _Xn (the deviation is the secondary moment amount). Therefore, simply speaking, the Lindeberg condition is a condition for checking the degree of variation of values. The past temperature check unit 20 excludes data that does not satisfy the predetermined allowable range of the Lindeberg condition (treats it as a missing value).

  The past temperature check unit 20 next checks whether there is a missing value in the past temperature data. If there is no missing value, the past temperature check unit 20 checks the past temperature storage unit 10 to deliver the temperature data to the subsequent processing unit. Store in a different area from the previous past temperature. Of course, the temperature data for which the checked / missing value has been complemented may be stored in a database different from the past temperature storage unit 10. If there is at least one missing value in the past temperature data, the missing value complementing unit 30 described later is called, and the missing value is complemented by other data, and then stored in the above area of the past temperature storage unit 10. . Missing values are excluded from the data for prediction as a result of checking the tolerance of data variation due to Lindeberg conditions as described above when the data was not originally obtained due to failure of the observation equipment, etc. It also occurs if That is, if there is a missing value in the original temperature data, it is stored as a missing value, and even if the original temperature data is not missing, if the variation in temperature data exceeds the tolerance, that temperature Treat the data in the same way as the missing values stored above.

  The missing value complementing unit 30 is called from the past temperature checking unit 20 when there is a missing value in the temperature data, and calculates the average temperature of each day for the past N−1 years excluding the year with the missing value. As an alternative value, the past average temperature of the same month of the year with a missing value. That is, the missing value is complemented with the average value of the data other than the year in which the missing occurred.

  The past average temperature calculation unit 40 does not simply calculate an average of past data for a predetermined year (N years), but calculates a moving average. The moving average is obtained by shifting the average value for each certain interval in the time series data while shifting the interval. That is, with respect to the past temperature data, if the number of moving average sections is n, each date has a section between (date− [n / 2]) days and (date + [n / 2]) days. Calculate the moving average and use it as the average temperature value for each date. Here, n is assumed to be an even number, but in the case of an odd number, [n / 2] represents an integer part of n / 2. The calculated moving average temperature value is stored in the past average temperature storage unit 50. The past average temperature storage unit 50 may be a database or a temporary storage area.

  The effective temperature integrating unit 60 reads the data of the past average temperature stored in the past average temperature storage unit 50 and integrates the effective temperature. Specifically, the effective temperature from the cultivation start date of the target crop to the prediction implementation date is obtained from the daily average temperature data, and the target date is integrated. In addition, the effective temperature of the future date from the forecast execution date is calculated from the average value of the past temperature, added up to the integrated value before the forecast execution date obtained above, and the total value is used as a parameter of the prediction condition When the input target effective integrated temperature is exceeded, the date is output as a predicted harvest date (harvest date) that is a harvest time.

  The predicted harvest date storage unit 70 stores the predicted harvest date output by the effective temperature integrating unit 60. Since the predicted harvest date varies depending on the predicted execution date and the conditions of each input parameter, it is desirable to store them as a database together with summary data of those conditions.

  Of the processing blocks described above, the prediction condition input unit 80 provides the operator with parameters necessary for each process for the past temperature check unit 20, the past average temperature calculation unit 40, and the effective temperature integration unit 60. Provide a means for input. The parameters input here are the prediction conditions required for the prediction together with the past temperature data stored in the database. Specifically, the number of years (N) of the past temperature data to be used and the variation of the temperature data. The tolerance of Lindeberg conditions for judging the tolerance of the degree, the cultivation start date, the lowest growth temperature and the highest growth temperature of the target crop, the target effective integrated temperature for judging the predicted harvest date, and the movement of the past temperature It is the number of moving average sections (n) for calculating the average. Based on these parameters (prediction conditions) and past temperature data, a predicted harvest date is calculated.

  The display / output unit 90 serves to display the predicted harvest date obtained by the effective temperature integrating unit 60 on a display unit (not shown) of the present apparatus or to output it to another apparatus. At this time, not only the predicted harvest date but also the summary of the data used for the prediction and the input parameters may be output together.

The operation of the present apparatus can be briefly summarized as follows. The N-year portion of the past temperature data used for estimation is typically regarded as a probability value according to a normal distribution with mean μ and variance σ ^{2 on} the same day of the same month (not necessarily limited to the normal distribution). In addition, when the variation in the temperature data is within the allowable range of the Lindeberg condition, the average value for N years is used for the prediction. If the temperature data is outside the allowable range, the average temperature for N-1 years excluding the data is used. If there is missing data (missing value), the average temperature on the date of the missing date is obtained from the past data on the day when the missing date is used as a substitute value for the missing data. In addition, the average temperature is not simply the average of past data for N years, but a moving average of a certain number of days is taken, and this value is used as the daily average temperature on the same day of the same month. ) Is smoothed.

  In addition, the functional configuration of the above-described apparatus is merely an example, and one functional block (a storage unit and a processing unit) may be divided or a plurality of functional blocks may be configured as one functional block. Good. In each processing unit, a CPU (Central Processing Unit) built in the device reads a computer program stored in a storage device such as a ROM (Read Only Memory), a flash memory, an SSD (Solid State Drive), a hard disk, It is realized by a computer program executed by the CPU. That is, each processing unit reads and writes necessary data such as a table from a database (DB; Data Base) stored in the storage unit or a storage area on a memory, and in some cases, related computer programs It is realized by controlling hardware (for example, input / output device, display device, communication interface device). Further, the database (DB) in the embodiment of the present invention may be a commercial database, but it simply means a collection of tables and files, and the internal structure of the database itself is not questioned.

(Processing flow of past temperature check unit 20)
Hereinafter, the processing steps performed by each functional block of the present apparatus will be described in more detail. First, FIG. 3 is a diagram illustrating a processing flow performed by the past temperature check unit 20 according to the embodiment of the present invention. In the subsequent processing flow diagrams (flowcharts), the processing order of the steps may be changed as long as the relationship between the input and output between the steps is not impaired.

  The past temperature check unit 20 first acquires the number of years (N) of the past temperature data to be used from the prediction condition input unit 80 in step S10. Next, in step S11, N years worth of temperature data to be used is read from the past temperature storage unit 10 and acquired.

  Subsequently, the past temperature check unit 20 acquires the Lindeberg condition tolerance from the prediction condition input unit 80 in step S12. The tolerance of the Lindeberg condition is a value for determining that the data is within the tolerance if the data is within the predetermined range, and that the tolerance is exceeded if the data is not within the predetermined range. For example, assuming that the probability distribution follows a normal distribution, the probability that the data is distributed within the range of ± 2σ (area under the probability density curve) is 95.45% with respect to the standard deviation σ, and is within the range of ± 3σ. Since the probability is 99.73%, this percentage value itself may be the tolerance.

  When the past temperature check unit 20 acquires the input parameters (N and the Lindeberg condition tolerance) and the temperature data for the past N years, subsequently, in step S13, the average μ for each of the past N years of data. (Simple average) and standard deviation σ are calculated. In step S14, it is determined whether each temperature data is within the tolerance of the given Lindeberg condition. If it is within the tolerance, the process proceeds to step S16. If the tolerance condition is not satisfied, the data is excluded and treated as a missing value (step S15). Here, the tolerance of the Lindeberg condition is, for example, when the data distribution follows a normal distribution, and if certain data is within the range of μ ± 3σ under the probability density curve, it is within the tolerance of variation. Otherwise, it is a value for determining the degree of variation of each data, such as out of the range of variation tolerance.

  Next, in step S16, the past temperature check unit 20 determines whether or not each data has a missing value after determining the Lindeberg condition for the temperature data acquired in step S11. If there is a missing value (if there is a missing value even for one day among the past temperature data for N years), the missing value complementing unit 30 is called in step S17 to perform missing value complementing processing. If there is no missing value (step S16: N), or after the missing value complementing process in step S17 is completed, in step S18, the past temperature data after the variation check and missing value compensation is converted into a subsequent past average temperature calculating unit. 40 and stored in the past temperature storage unit 10 for delivery to the effective temperature integrating unit 60, and the process is terminated. In step S18, the past temperature data after the variation check and missing value compensation is stored in another area of the database in order to be distinguished from the original past temperature data before the fluctuation check and missing value compensation.

(Processing flow of missing value complementing unit 30)
FIG. 4 is a diagram illustrating a processing flow performed by the missing value complementing unit 30 according to the embodiment of the present invention. It is assumed that leap years are not included in subsequent date processing. In step S30, the missing value complementing unit 30 first substitutes the missing occurrence date delivered from the past temperature checking unit 20 into the variables D0 and D1. In step S31, it is checked whether or not the temperature data of (D1 + 1 day) is in the past temperature storage unit 10. If there is no temperature data, since (D1 + 1 day) is also a missing day, D1 is increased by 1 day in step S32, and the process returns to step S31.

  If there is temperature data for the day of (D1 + 1 day) (step S31: Y), the process proceeds to step S33, and the average temperature of the same month of the year excluding the year in which the missing value exists is replaced with the missing value of the day of D0. Complement as a value. In step S34, D0 and D1 are compared. If D0 is smaller than D1, D0 is incremented by 1 in step S35, and the process returns to step S33. In step S34, if D0 is equal to or greater than D1, the process proceeds to step S36.

  In step S40, the missing value complementing unit 30 finally checks whether or not the complementing process has been performed for all the missing occurrence dates every year. If there is an unprocessed missing occurrence date, the missing missing date is newly set. After setting the date of occurrence of a deficiency, the process returns to step S30, and the processing steps of steps S31 to S36 are repeated. Then, if there is no unprocessed missing date (step S36: Y), the process is terminated and the process returns to the past temperature check unit 20.

(Processing flow of past average temperature calculation unit 40)
FIG. 5 is a diagram illustrating a processing flow performed by the past average temperature calculation unit 40 according to the embodiment of the present invention. In step S50, the past average temperature calculating unit 40 first acquires the number n of moving average sections for obtaining a moving average of past temperatures from the prediction condition input unit 80. Next, in step S51, the past temperature data for N years after the variation check and missing value compensation are read from the storage area of the past temperature storage unit 10 output by the past temperature check unit 20.

  Next, in step S52, the past average temperature calculating unit 40 substitutes a value corresponding to January 1 for the variable Date. Subsequently, in step S53, Date- [n / 2] is substituted into the variable Date1 ([n / 2] represents an integer part of n / 2). In step S54, Date + [n / 2] is substituted for the variable Date2. Then, in Step S55, it is checked whether or not Date1 is before December 31, and if it is not before December 31 (Step S55: N), Date1 is set to December 31 as a process that crosses the year. Substitute-([n / 2] +1). If it is after December 31, the process of step S56 is skipped.

  Similarly, in step S57, it is checked whether or not Date2 is after January 1st. If it is not after January 1st, Date2 is set to January 1 + ([n / 2] ] -1) is substituted. If it is before January 1, the process of step S58 is skipped.

  Lastly, in step S59, the past average temperature calculating unit 40 substitutes an average value between the average temperature of Date1 and the average temperature of Date2 in the past average temperature of Date. In step S59 in the figure, the notation “Past average temperature (Date)” represents the past average temperature on the Date day stored in the storage unit. The same applies to the average temperature (Date1) and the average temperature (Date2). In this way, the moving average between the moving average section number n is obtained, and the past average calculating process is ended.

(Processing flow of effective temperature integrating unit 60)
FIG. 6 is a diagram illustrating a processing flow performed by the effective temperature integrating unit 60 according to the embodiment of the present invention. First, in step S70, the effective temperature integration unit 60 acquires the cultivation start date, the target effective integration temperature, the highest growth temperature, and the lowest growth temperature from the prediction condition input unit 80, and in step S71, the acquired highest growth temperature is T1, The minimum growth temperature is T2.

  In step S72, the effective temperature integrating unit 60 initializes the variable Date on the cultivation start date acquired above, and in step S73, initializes the variable “effective integrated temperature” to zero.

  In step S74, it is checked whether Date is after the predicted execution date. If Date is later than the prediction implementation date, the past average temperature -T2 of Date is substituted for the effective temperature (step S75), and the past average temperature of Date is subtracted from T1 and substituted into variable TD. (Step S76). In step S74, if Date is earlier than the prediction implementation date, the past temperature -T2 on the date of date is substituted into the effective temperature (step S77). Subsequently, the past temperature on the date of date is subtracted from T1, and variable TD is set. (Step S78).

  In step S79, it is checked whether or not the effective temperature calculated in step S75 or step S77 is zero or more. If the effective temperature is a negative value, the effective temperature is set to zero (step S80). Next, in step S81, it is checked whether TD is equal to or greater than zero. If the value is negative, TD is set to zero (step S82). This means that the effective temperature or TD is negative means that the temperature exceeds the maximum growth temperature, and such data does not contribute to the growth of the crops, so this is excluded. is there.

  Next, in step S83, the effective integrated temperature ← (effective integrated temperature + (effective temperature−TD)) is calculated. In step S84, it is checked whether the effective integrated temperature is equal to or higher than the target effective integrated temperature. If the effective integrated temperature does not reach the target value, after increasing Date by 1 in step S85, the process returns to step S74, and the processes of steps S75 to S84 are repeated.

  Finally, in step S86, the effective temperature integrating unit 60 outputs the day after the Date as the arrival date of the target integrated effective temperature. In this way, the effective temperature integration process ends.

(Summary data of effective accumulated temperature)
FIG. 7 is a diagram illustrating a specific example of the effective integrated temperature summary data output by the harvest time prediction device 100 according to the embodiment of the present invention. In this example, since the target effective integrated temperature is set to 1000 degrees, the day when the effective integrated temperature integrated from the cultivation start date (September 1) exceeds 1000 degrees (October 26) can be harvested (predicted) Harvest date). As already described with reference to FIG. 6, the effective accumulated temperature is calculated using the data in the “Past Temperature” column before the prediction date, and the data in the “Past Average Temperature” column after the prediction date. Is used. However, if the data of the prediction execution date (September 11) is obtained at the time of the prediction start, the data of the prediction execution date is included before the prediction execution date; otherwise, it is included after the prediction execution date. It is.

(Effect of embodiment)
As described above, by increasing the accuracy of the effective integrated temperature used for the prediction, it is possible to increase the accuracy of the harvest season harvest time prediction by the effective integrated temperature method. In other words, the meteorological data used for prediction includes errors related to measurement, such as observation equipment and observation methods, and errors related to fluctuations in the temperature itself. However, in the harvest time prediction apparatus of this embodiment, errors related to measurement (data loss) In this case, the Lindeberg condition can be checked and the missing value can be corrected. In addition, fluctuations in the weather data itself can be smoothed by using a moving average.

  As mentioned above, although this invention was demonstrated using embodiment, it cannot be overemphasized that the technical scope of this invention is not limited to the range as described in the said embodiment. It will be apparent to those skilled in the art that various modifications or improvements can be added to the above-described embodiments. Further, it is apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention. In the above embodiment, the present invention is described as the product invention in the harvest time prediction apparatus 100. However, the present invention is a method invention (harvest time prediction method) or a computer program invention (harvest time prediction program). Can also be taken as.

DESCRIPTION OF SYMBOLS 10 Past temperature storage part 20 Past temperature check part 30 Missing value complement part 40 Past average temperature calculation part 50 Past average temperature storage part 60 Effective temperature integration part 70 Predicted harvest date storage part 80 Prediction condition input part 90 Display / output part 100 Harvest Time prediction device 200 Effective integrated temperature summary data 201 Data portion before prediction execution date 202 Data portion after prediction execution date

Claims (4)

A harvest time prediction device for predicting the harvest time of agricultural products,
A past temperature storage unit for storing daily temperature data for a plurality of past years in the crop cultivation place;
When the temperature data stored in the past temperature storage unit is read out for a predetermined year, the case where the read temperature data is partially missing is stored as a missing value, and the temperature data is not missing Is a past temperature check unit that checks whether or not the temperature data falls within a predetermined tolerance range of Lindeberg conditions, and excludes temperature data that does not fall within the tolerance range as a missing value;
When there is the missing value in the temperature data after the check by the past temperature check unit, the average value of the temperature data on the same day of the same month except the year in which the missing value exists is complemented as an alternative value of the missing value A missing value complement to
For the temperature data after complementing the missing value, a past average temperature calculating unit that calculates a past average temperature,
Based on the past average temperature and the minimum growth temperature and maximum growth temperature of the crop, the effective temperature of the crop is integrated every day from the cultivation start date, and the integrated effective integrated temperature is a predetermined target effective integrated temperature. An effective temperature accumulating unit that outputs the exceeded day as a predicted harvest date of the crop;
A harvesting time predicting device comprising:   The said past average temperature is a moving average of a predetermined area, The harvest time prediction apparatus of Claim 1 characterized by the above-mentioned. A method for predicting the harvest time for predicting the harvest time of agricultural products,
Computer
Storing temperature data for each day of the past multiple years at the crop cultivation place;
When the temperature data stored in the storing step is read out for a predetermined year, the case where the read out temperature data is partially missing is stored as a missing value, and the temperature data is not missing Checking whether the temperature data is within a predetermined tolerance range of Lindeberg conditions;
Excluding temperature data not within the tolerance range as missing values;
In the case where there is the missing value in the temperature data after the checking in the checking step and the excluding step, the average value of the temperature data on the same month of the year excluding the year in which the missing value exists is calculated as the missing value. A step to complement as an alternative value,
A step of calculating a past average temperature for the temperature data after complementing the missing value;
Based on the past average temperature and the minimum growth temperature and maximum growth temperature of the crop, the effective temperature of the crop is accumulated daily from the cultivation start date,
Outputting the day when the integrated effective integrated temperature exceeds a predetermined target effective integrated temperature as a predicted harvest date of the crop;
A method for predicting harvest time, characterized in that A harvest time prediction program for predicting the harvest time of crops,
On the computer,
Storing temperature data for each day of the past multiple years at the crop cultivation place;
When the temperature data stored in the storing step is read out for a predetermined year, the case where the read out temperature data is partially missing is stored as a missing value, and the temperature data is not missing Checking whether the temperature data is within a predetermined tolerance range of Lindeberg conditions;
Excluding temperature data not within the tolerance range as missing values;
In the case where there is the missing value for the temperature data after the checking in the checking step and the excluding step, the average value of the temperature data on the same month of the year excluding the year in which the missing value exists is calculated as the missing data. A step to complement as an alternative value,
A step of calculating a past average temperature for the temperature data after complementing the missing value;
Based on the past average temperature and the minimum growth temperature and maximum growth temperature of the crop, the effective temperature of the crop is accumulated daily from the cultivation start date,
Outputting the day when the integrated effective integrated temperature exceeds a predetermined target effective integrated temperature as a predicted harvest date of the crop;
Harvest time prediction program characterized by causing

Images