JPH1096790A - Weather condition forecasting method and device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To precisely forecast a local future weather condition from present weather condition information of an object point by searching and finding a weather condition changing degree of the object point corresponding to past weather condition information of a peripheral point from area data in a database. SOLUTION: A weather condition observing device 2 observes an atmospheric temperature, atmospheric pressure, humidity, wind force, the wind direction and the sunshine to constitute weather condition information, and supplies these to an information processor to constitute an atmospheric temperature anticipating device 3. The atmospheric temperature forecasting device 3 obtains weather condition analyzing information through a modem 1 from an external information means besides them. Next, past area data matching to this is searched from a statistical database by using a forecast object point and present weather condition information of its peripheral point as a key, and a future temperature in an object point is forecasted from an atmospheric temperature changing degree corresponding to the searched area data. Afterwards, present weather condition information is successively observed, and observation data is compared with forecast data, and when they are different from each other, statistical information is corrected. In this way, a precise atmospheric temperature can be forecasted while polishing the statistical information.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a weather forecasting method and apparatus for accurately performing a local weather forecast based on general-purpose weather information provided by the Japan Meteorological Agency or the like. The present invention relates to a weather forecasting method and apparatus suitable for application to cultivation, temperature control of a large-scale warehouse, and the like.

[0002]

2. Description of the Related Art Generally, a feedback control system such as PID control is adopted in a large-scale temperature control system, for example, a temperature control system in a paddy field, a vinyl cultivation of fruit trees, a large-scale warehouse and the like. This kind of feedback control method is effective for a process with a relatively small disturbance due to its basic principle of maintaining a controlled process at a target value while detecting and correcting the fluctuation. However, in the case of the above-described large-scale temperature control system, fluctuations in weather conditions act on the control system as large disturbances, so that it is difficult to stably control this with only the feedback control method.

In the above-mentioned large-scale temperature control system, if a change in the weather environment surrounding the large-scale temperature control system can be quickly predicted, a more accurate and stable system can be obtained by adding a feedforward control method using the predicted value. It seems that temperature control can be realized.

[0004]

However, conventional local weather forecasting techniques are not always satisfactory enough in terms of their forecasting accuracy.
Executing feedforward control using such a prediction result often leads to the temperature of the controlled object in the direction opposite to the actual temperature change (disturbance) due to unexpectedness, and rather makes the controlled process unstable. There was a problem.

The present invention has been made in view of such conventional problems, and an object thereof is to accurately perform local weather forecasting based on general-purpose weather information provided by the Japan Meteorological Agency or the like. It is an object of the present invention to provide a weather forecasting method and apparatus which can be carried out at a time.

[0006]

The invention described in claim 1 of the present application is a forecast target point weather information input means for inputting current weather information of a weather forecast target point, and a surrounding area of the weather forecast target point. A peripheral point weather information input means for inputting current weather information of the point, and past weather information of the surrounding point of the weather prediction target point and a corresponding weather change degree of the weather prediction target point corresponding to the weather prediction target point are associated with each other. A weather information database in which a plurality of items are stored as regional data, and a search for regional data in the database using the current weather information of the inputted peripheral point as a key corresponds to past weather information of the corresponding peripheral point. Weather change degree search means for obtaining the weather change degree of the prediction target point to be predicted;
Weather calculation means for calculating a future weather at the weather prediction target point based on the inputted current weather information of the weather prediction target point and the degree of weather change of the weather prediction target point obtained by the search. And a weather forecasting device characterized by the following.

[0007] The invention according to claim 2 of the present application is a prediction target point weather information input means for inputting the current weather information of the weather prediction target point, and the present weather information of the surrounding points of the weather prediction target point. A surrounding point weather information input means for inputting, a wind direction information constituting the current weather information of the weather forecast target point, and a current corresponding to the windward of the weather forecast target point based on the current weather information of the surrounding point. Upwind weather information calculation means for obtaining weather information of a specific point, a deviation between the input weather information of the prediction target point and the weather information of a surrounding specific point corresponding to the windward, and a prediction target point corresponding to the deviation. A weather information database in which a plurality of pieces of information relating the degree of weather change are stored as weather gradient data, a current weather information of the input peripheral point, and a corresponding windward. Weather change degree search means for searching the weather gradient data in the database using the deviation from the weather information of the surrounding specific point as a key to obtain the degree of weather change of the prediction target point corresponding to the corresponding past weather deviation; Weather calculation means for calculating a future weather at the weather prediction target point based on the inputted current weather information of the weather prediction target point and the degree of weather change of the weather prediction target point obtained by the search. And a weather forecasting device characterized by the following.

The invention according to claim 3 of the present application is characterized in that the weather is temperature and the degree of weather change is a temperature change over time. It is in the weather forecasting device described.

The invention described in claim 4 of this application is characterized in that the forecast target point weather information input means comprises weather observation means for observing and inputting current weather information of the weather forecast target point. A weather forecasting device according to claim 1 or 2.

In the invention according to claim 5 of the present application, the surrounding point weather information input means inputs the surrounding weather information transmitted from a weather information transmitting station such as a weather service support center via a communication line. The weather forecasting device according to claim 1 or 2, wherein:

According to a sixth aspect of the present invention, there is provided a prediction target weather information input step for inputting current weather information of a weather prediction target point, and a current weather information of a surrounding point of the weather prediction target point. A surrounding point weather information input step for inputting, and a plurality of pieces of information relating the past weather information of the surrounding points of the weather forecast target point and the corresponding weather change degree of the weather forecast target point corresponding to each other as the regional data. By searching the stored weather information database using the current weather information of the input peripheral point as a key, the weather change degree of the prediction target point corresponding to the past weather information of the corresponding peripheral point is obtained. A search step, based on the current weather information of the input weather prediction target point and the degree of weather change of the weather prediction target point obtained by the search. In weather prediction method characterized by comprising the a weather calculating a future weather weather forecast object point to have.

According to a seventh aspect of the present invention, there is provided a prediction target point weather information input step for inputting current weather information of a weather prediction target point, and a current weather information of a surrounding point of the weather prediction target point. Surrounding point weather information input step for inputting, based on wind direction information constituting the current weather information of the weather forecast target point and current weather information of the surrounding point, weather information of a surrounding specific point corresponding to the windward Windward weather information calculation step of calculating, and the deviation between the input weather information of the prediction target point and the weather information of the surrounding specific point corresponding to the windward and the degree of weather change of the prediction target point corresponding to the deviation. A meteorological information database storing a plurality of items associated with each other as meteorological gradient data is stored in the form of a meteorological information of the inputted peripheral point and a peripheral area corresponding to the determined windward. By searching using the deviation from the weather information of the fixed point as a key, a weather change degree search step for obtaining the weather change degree of the prediction target point corresponding to the corresponding past weather deviation, a weather change degree search step of the input weather prediction target point A weather calculation step of calculating a future weather of the weather prediction target point based on the current weather information and the degree of weather change of the weather prediction target point obtained by the search. .

The invention according to claim 8 of the present application is characterized in that the weather is temperature and the degree of weather change is a change in temperature over time. It is in the described weather forecast method.

According to a ninth aspect of the present invention, in the prediction target point weather information input step, information from weather observation means for monitoring current weather information of the weather prediction target point is input. A weather forecasting method according to claim 6 or claim 7.

According to a tenth aspect of the present invention, in the surrounding point weather information input step, surrounding weather information transmitted from a weather information transmitting station such as a weather service support center is input via a communication line. A meteorological forecasting method according to claim 6 or claim 7, characterized in that:

According to an eleventh aspect of the present invention, there is provided a temperature control apparatus for executing feedforward control using predicted weather by the weather prediction apparatus according to the first or second aspect. is there.

In the above claims, "weather information" is a general term for wind direction, wind power, temperature, humidity, sunshine and the like.

The term "degree of weather change" means the degree of change of temperature with respect to time, which is obtained under predetermined weather conditions (wind direction, wind, temperature, humidity, sunshine, etc.).

According to the first or sixth aspect of the present invention, if the current weather information of the weather prediction target point and the current weather information of the surrounding points are known, the future weather of the weather prediction target point can be obtained by using the information alone. Can be accurately predicted.

According to the second or seventh aspect of the present invention, if the current weather information of the weather prediction target point and the current weather information of the surrounding specific point corresponding to the windward are known, the weather information can be obtained by using the information alone. The future weather at the prediction target point can be accurately predicted.

According to the third or eighth aspect of the invention, in the weather forecasting apparatus according to the first or second aspect and the sixth or seventh aspect, the temperature prediction target point The temperature in the future can be accurately predicted.

According to the invention set forth in claim 4 or claim 9, in the weather forecasting apparatus according to claim 1 or claim 2 or claim 6 or claim 7, the weather forecasting target point Currently, the task of inputting weather information can be automated.

According to the fifth or tenth aspect of the present invention, in the weather forecasting apparatus according to the first or second aspect and the sixth or seventh aspect, the current weather at a peripheral point is provided. Information entry can be automated.

Further, according to the present invention, the control accuracy in a large-scale temperature control system such as a paddy field, a vinyl cultivation of fruit trees, a large-scale warehouse, etc. can be remarkably improved.

[0025]

Preferred embodiments of the present invention will be described below in detail with reference to the accompanying drawings.

FIG. 1 schematically shows the configuration of the entire temperature control system for executing feedforward control using the temperature predicted according to the present invention. In the drawings, a portion surrounded by a broken line is a portion related to the present invention.

As shown in FIG. 1, this temperature control system is provided with current weather information (for example, weather radar) of a peripheral point transmitted from a weather information transmitting station such as a weather service support center via a wired or wireless line. 1 for receiving and demodulating weather analysis diagrams such as AMeDAS and AMeDAS, and a weather observation device 2 for automatically obtaining current weather information of a prediction target point.
And predicts the future temperature (weather) of the prediction target point based on the weather information, and executes the feedforward control using the predicted temperature to execute the heat source control device (the vinyl storage in a large-scale warehouse or orchard). And a weather forecasting device 3 for controlling a house temperature control boiler 4).

FIG. 4 shows an example of a weather analysis chart which is weather information from an external information organization which is received and demodulated by the modem 1.
Note that the weather analysis diagram shown in the same figure is obtained by photographing the northern hemisphere from the sky above the North Pole for convenience of description, but for local weather forecasting described later, it is called a 20 km mesh to 30 km mesh. It is needless to say that such a local meteorological analysis map having a higher resolution is suitable. This type of weather analysis diagram includes, for example, temperature information, humidity information, and wind information of the upper layer of 850 hp. Specifically, the temperature is represented by an isotherm, and the humidity is represented by a color density.
The winds are each represented by an L-shaped arrow symbol. The information processing device constituting the temperature prediction device 3 uses, for example, a well-known pattern matching method to convert the information of each point from the above-mentioned weather analysis diagram into, for example, temperature (4 degrees), humidity ( 50%), wind direction (S
E), it can be read like wind (10 knots).

FIG. 2 shows details of the main part of the temperature control system shown in FIG. As shown in the figure, a weather observation device 2 for obtaining weather information of a prediction target point includes a thermometer 21 for observing temperature, pressure, humidity, wind power, wind direction, and sunshine constituting the weather information. , Barometer 22,
A hygrometer 23, an anemometer 24, an anemometer 25, and a sunshine meter 26 are provided. Then, the information observed by these meters 21 to 26 is automatically read into the information processing device constituting the temperature prediction device 3 at an appropriate timing.

The information processing device constituting the temperature prediction device 3 includes a statistical information recording device 31 and an observation information recording device 32. As described in detail later, the statistical information recording device 31 records (stores) statistical data (weather information database), which is past weather information required for weather prediction.
Have been. As will be described later in detail, the observation information recording device 32 has the weather information of a peripheral point from an external information organization received and demodulated by the modem 1 and the prediction target point observed by the weather observation device 2. Weather information is recorded (stored).

FIG. 7 shows an example of statistical data constituting the weather information database. As shown in the figure, the statistical data is roughly classified into "regional data" and "temperature gradient data". Here, “regional data” refers to a plurality of peripheral points (points) included in a specific area centered on a weather prediction target point (hereinafter, also referred to as a “temperature control point because it is also a temperature control target point”). A, point B, point C, point D) and the temperature control points, the "weather data" obtained at a specific date and time in the past, and in the state where such weather data is obtained, the temperature of the temperature control point is changed to the time thereafter. "Temperature change data" that shows how it has changed over time
And a pair. Here, an example of the positional relationship between the “temperature control point” and its surrounding points (point A, point B, point C, point D) is shown in FIG.

That is, as shown in FIG. 7, "data 1", which is an example of the regional data, indicates that, at 20:00 on March 20 which is the past specific date and time, the weather information at the temperature control point indicates the wind direction ( E), wind (10 m / s), temperature (8 degrees), humidity (8%), sunshine (50%), upper layer temperature (2 degrees), upper layer humidity (50%), and its surroundings The weather information at the point (point A, point B, point C, point D) includes wind direction (NNE, NE, NW, NW), wind power (1
0 m / s, 15 m / s, 13 m / s, 14 m / s), temperature (10 degrees, 13 degrees, 14 degrees, 14 degrees), humidity (10 degrees)
%, 10%, 20%, 23%), sunshine (90%, 90%
%, 90%, 85%), and the temperature change at the temperature control point after such a weather condition is the temperature change (−0.4 degrees / hour). It shows that. Similarly, “data 2”, which is an example of regional data, is a past specific date and time of August 2
At 12:00 on the day, the weather information at the temperature control point includes wind direction (E), wind force (10 m / s), and temperature (28
Degrees), humidity (50%), sunshine (30%), upper layer temperature (8 degrees), upper layer humidity (80%), and its surrounding points (point A, point B, point C, point D) ), The wind direction (SSE, NE, NW, N), wind power (20
m / s, 15 m / s, 13 m / s, 14 m / s), temperature (30 degrees, 34 degrees, 33 degrees, 30 degrees), humidity (60%,
70%, 65%, 65%), sunshine (30%, 30%, 3
0%, 35%), and the temperature change at the temperature control point after such a weather condition was the temperature change (+0.2 degrees / hour). It is shown.

On the other hand, the "temperature gradient data" refers to the weather deviation between the weather information of the surrounding specific point corresponding to the windward of the temperature control point and the weather information of the temperature control point at a specific point in time in the past. In the state where the weather deviation is obtained, the data is paired with "temperature change data" indicating how the temperature at the temperature control point has changed over time thereafter.

FIG. 5 shows an example of the peripheral specific point E corresponding to the windward side of the temperature control point. In this case, the weather information of the peripheral specific point E can be obtained as a substantially average of the weather information of the peripheral points C and D. The meaning of using the “temperature gradient data” is based on the idea that the temperature change at the temperature control point is determined to some extent uniquely by the three-dimensional movement of the air existing on the wind.

That is, as shown in FIG. 7, "data 1" which is an example of the temperature gradient data corresponds to the windward of the temperature control point at 21:00 on March 20 which is the past specific date and time. The weather information at the surrounding specific points includes wind direction (NNE), wind force (10 m / s), humidity (30%), altitude difference from the temperature control point (100 m), sunshine (30%),
This indicates that the temperature was different from the temperature control point (4 degrees). Under such a weather deviation condition, the temperature change at the temperature control point was changed by the temperature change (−0.4 degrees / Time). Similarly,
"Data 2" which is an example of the temperature gradient data indicates that, at 12:00 on August 2 which is the past specific date and time, the weather information at the peripheral specific point corresponding to the windward of the temperature control point is wind direction (SSW), Wind (13m / s), humidity (70%),
Altitude difference from the temperature control point (10m), sunshine (30%),
This indicates that the temperature was different from the temperature control point (2 degrees), and the temperature change at the temperature control point was the temperature change (+0.8 degrees / hour) under such weather deviation conditions. ).

FIG. 9 shows an example of observation data recorded in the observation information recording device 32. That is, this observation data, “Data 1”, shows that the weather information of the temperature control point at 18:00 on March 20 includes wind direction (E) and wind force (10 m /
s), temperature (6 degrees), humidity (10%), sunshine or cloud cover (50%).

Next, the operation of the temperature prediction device 3 shown in FIG. 2 will be systematically described with reference to the flowchart of FIG.

First, the data processing device constituting the temperature prediction device 3 observes current weather information of a temperature control point (see FIG. 5) which is a weather prediction point (step 301). The observation of the weather information is performed by the thermometer 21, the barometer 22, the hygrometer 23, and the anemometer 2 constituting the weather observation device 2.
4. This is performed based on observation data (see FIG. 9) taken in from the anemometer 25 and the sunshine indicator 26.

Next, the data processing device constituting the temperature prediction device 3 obtains weather analysis information from an external information organization (step 302). As described above, this weather analysis information is obtained from the weather analysis chart received and demodulated via the modem 1 (see FIG. 4) by using a known pattern matching method at each of the surrounding points A to D (see FIG. 5). This is performed by reading weather data (see FIG. 6) such as temperature, humidity, wind direction, and wind force.

Next, the data processing device constituting the temperature predicting device 3 uses the temperature control point obtained in steps 301 and 302 and the current weather information of the surrounding points A to D as keys to match the past weather information. Search regional data (see FIG. 7) from the statistical data (step 30).
3). At this time, the allowable error range of the matching judgment condition between the observation data and the statistical data is, for example, within one month before and after the predicted date, within two hours before and after the time, within two directions before and after the wind direction, and within two directions before and after the temperature. Within 3 degrees before and after, within 30% before and after sunshine, and within 10 points for the others, correction is made by applying an appropriate correction coefficient depending on the region.

If the allowable error range is set in this manner, for example, in the case of the “data 1” of the regional data shown in FIG. 7, it will match with the weather information shown in FIG. It will be understood that there is no matching with the weather information shown in FIG.

Next, if local data matching the current weather information of the temperature control point and the surrounding weather points A to D is found (step 303 YES), the temperature change degree corresponding to the searched local data is used. To predict the future temperature at the temperature control point (step 3
04). This temperature prediction processing is performed according to the following equation: predicted temperature = current temperature + temperature change degree * time. Therefore, as long as the degree of temperature change can be accurately known, a local future temperature can be accurately predicted. For example, at 18:00 on March 20, the temperature at the temperature control point is 9.0 degrees, and the data obtained from the external information institution is "data 1" of the regional data shown in FIG.
, The corresponding degree of temperature change is -0.4 degrees / hour, and the temperatures at the 21:00 temperature control points on the same day, which are three hours later, are substituted into the above equation. Thus, it is calculated as 7.8 degrees.

On the other hand, the temperature control point and its surrounding point A
If no area data matching the current weather information of .about.D is found (NO in step 303), a prediction process of a future temperature at a temperature control point is performed using the above-mentioned temperature gradient data (step 305). As described above, this temperature prediction processing is performed in the vicinity specific point E corresponding to the windward side.
The temperature difference, humidity, and height difference between the two are obtained from the weather information of FIG. 7 and the weather information of the temperature control point.
The temperature is matched with the temperature gradient data shown in (1), and if the matched data exists, the prediction process of the future temperature at the temperature control point is performed using the temperature change degree corresponding to the data. The allowable error range in matching with the temperature gradient data does not need to be as strict as the matching process with the regional data described above. This temperature prediction process is also performed according to the following equation: predicted temperature = current temperature + degree of temperature change * time. Therefore, as long as the degree of temperature change can be accurately known, a local future temperature can be accurately predicted. For example, assuming that the temperature at the temperature control point is 29.0 degrees at 12:00 on August 2 and that the data obtained from the external information institution matches the "data 2" of the temperature gradient data shown in FIG. Since the degree of temperature change corresponding thereto is +0.8 degrees / hour, the temperature at the temperature control point at 15:00 on the same day corresponding to 3 hours later is 31.4 degrees by substituting them into the above equation. Is calculated.

Next, the data processing device constituting the temperature prediction device 3 predicts the control amount on the premise of the predicted future temperature of the control point (step 306), and based on the prediction, performs feedforward control of the boiler. (Step 307).

Thereafter, while observing the current weather information sequentially (step 308), the observed data is compared with the previously predicted data (step 309), and if the observed data is different from the predicted data Is to correct the statistical information (step 310) to perform accurate temperature prediction while refining the statistical information to increase the prediction accuracy.

The initial value of the temperature gradient data is calculated based on the principle of temperature change shown in FIG. 8 and the physical law of temperature change by compressing and expanding air containing water vapor. Since the calculated value is different from the actual value, it is preferable to correct the temperature gradient data as needed based on the observation data.

Thus, according to this embodiment, for example, in a large-scale temperature control system such as a paddy field, vinyl cultivation of fruit trees, or a large-scale warehouse, fluctuations in the weather environment surrounding the system are quickly predicted, and the predicted value is calculated. By adding a feed-forward control method using this method, more precise and stable temperature control can be realized.

The above-described embodiment can of course be implemented by a stored program type control device such as a personal computer. At that time, a computer program for carrying out the present invention includes a hard disk, Of course, it can be stored in various recording media such as a floppy disk and a CD-ROM.

[0049]

As is apparent from the above description, according to the present invention, a local weather forecast can be accurately performed based on general-purpose weather information provided by the Meteorological Agency or the like.

[Brief description of the drawings]

FIG. 1 is a block diagram schematically illustrating a configuration of an entire temperature control system that performs feedforward control using a temperature predicted according to the present invention.

FIG. 2 is a block diagram showing details of a main part of the temperature control system shown in FIG. 1;

FIG. 3 is a flowchart showing an operation procedure of the temperature control system shown in FIG. 2;

FIG. 4 is an explanatory diagram showing an example of weather information received via a modem.

FIG. 5 is an explanatory diagram showing an example of a positional relationship between a wind direction of a weather prediction target point and surrounding points.

FIG. 6 is an explanatory diagram showing an example of matching with regional data in a weather database (statistical data) and an example of non-matching of current weather information of a weather prediction target point and surrounding points.

FIG. 7 is an explanatory diagram showing an example of each of regional data and temperature gradient data constituting a weather database (statistical data).

FIG. 8 is an explanatory diagram for explaining a method for obtaining the temperature at the prediction target point based on the temperature and the wind speed at the windward point.

FIG. 9 is an explanatory diagram showing an example of observation data at a weather forecast point.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Modem 2 Meteorological observation device 3 Temperature prediction device 4 Heat source control device 21 Thermometer 22 Barometer 23 Hygrometer 24 Anemometer 25 Anemometer 26 Sunlight meter

Claims (11)

[Claims] 1. A prediction target weather information input means for inputting current weather information of a weather prediction target point, and a peripheral weather information input means for inputting current weather information of a peripheral point of the weather prediction target point A weather information database in which a plurality of pieces of past weather information of surrounding points of the weather prediction target point and a corresponding degree of weather change of the weather prediction target point corresponding thereto are stored as a plurality of regional data, By searching regional data in the database using the current weather information of the surrounding points as a key,
Weather change degree search means for obtaining the degree of weather change of the prediction target point corresponding to the past weather information of the corresponding peripheral point; and the current weather information of the input weather prediction target point and the weather prediction target obtained by the search A meteorological calculation unit that calculates a future weather at a weather prediction target point based on the degree of weather change at the point. 2. A forecast point weather information input means for inputting current weather information of a weather forecast target point, and a surrounding point weather information input means for inputting current weather information of surrounding points of the weather forecast target point Based on wind direction information constituting the current weather information of the weather forecast target point and current weather information of the surrounding point, windward weather for weather information of a specific surrounding point corresponding to the windward of the weather forecast target point An information calculation unit, and a correlation between the input weather information of the prediction target point and the weather information of the surrounding specific point corresponding to the windward and the degree of weather change of the prediction target point corresponding to the deviation. A weather information database stored as a plurality of pieces of weather gradient data, and a deviation between the input current weather information of the surrounding point and the weather information of the surrounding specific point corresponding to the obtained windward. By searching for weather gradient data in the database using the key as a key, a weather change degree search means for obtaining a weather change degree of a prediction target point corresponding to the corresponding past weather deviation, a weather change degree search means of the input weather prediction target point A weather calculation unit that calculates a future weather at the weather prediction target point based on current weather information and a degree of weather change at the weather prediction target point obtained by the search. 3. The weather forecasting apparatus according to claim 1, wherein the weather is temperature, and the degree of weather change is a temperature change over time. 4. The forecast target point weather information input means,
3. The weather forecasting device according to claim 1, comprising weather observation means for observing and inputting current weather information of the weather forecast target point. 5. The surrounding weather information input means inputs surrounding weather information sent from a weather information transmitting station such as a meteorological service support center via a communication line. Item 3. The weather forecasting device according to item 2. 6. A prediction target weather information input step for inputting current weather information of a weather prediction target point, and a peripheral weather information input step for inputting current weather information of a peripheral point of the weather prediction target point. The weather information database in which past weather information of surrounding points of the weather prediction target point and a corresponding degree of weather change of the weather prediction target point corresponding thereto are stored as a plurality of regional data as the local data, A weather change degree search step of obtaining a weather change degree of a prediction target point corresponding to past weather information of the corresponding surrounding point by searching using the current weather information of the obtained surrounding point as a key, and the input weather forecast Based on the current weather information of the target point and the degree of weather change of the weather prediction target point obtained by the search, the future weather of the weather prediction target point And a weather calculation step of calculating the weather forecast. 7. A prediction target weather information input step for inputting current weather information of a weather prediction target point, and a peripheral weather information input step for inputting current weather information of a peripheral point of the weather prediction target point. Based on the wind direction information and the current weather information of the surrounding points constituting the current weather information of the weather forecast target point, a windward weather information calculating step of obtaining weather information of a specific surrounding point corresponding to the windward; A plurality of data obtained by correlating the deviation between the inputted weather information of the prediction target point and the weather information of the surrounding specific point corresponding to the windward and the degree of weather change of the prediction target point corresponding to the deviation as weather gradient data. The stored weather information database is used as a key to store the deviation between the inputted current weather information of the surrounding point and the weather information of the surrounding specific point corresponding to the obtained windward. By searching as, the weather change degree search step of obtaining the weather change degree of the prediction target point corresponding to the corresponding past weather deviation, the current weather information of the input weather prediction target point and the current weather information obtained by the search A weather calculation step of calculating a future weather at the weather prediction target point based on a degree of weather change at the weather prediction target point. 8. The weather forecasting method according to claim 6, wherein the weather is a temperature, and the degree of weather change is a change in temperature over time. 9. The forecast target point weather information input step inputs information from weather observation means for observing current weather information of the weather forecast target point. Weather forecast method. 10. The method according to claim 6, wherein the step of inputting weather information around the surroundings inputs surrounding weather information sent from a weather information transmitting station such as a weather service support center via a communication line. Item 7. The weather forecasting method according to Item 7. 11. A temperature control device that performs feedforward control using predicted weather calculated by the weather prediction device according to claim 1 or 2.