JP2002196085A - System for predicting road surface condition - Google Patents

Abstract

PROBLEM TO BE SOLVED: To rationally predict a road surface condition ranging over from a short term upto medium and long terms. SOLUTION: An actual locale data from an actual locale meteorological-data collecting device comprising a road surface-shaped sensor for a monitored road surface and a meteorological sensor group, and a Japan Meteorological Agency GPV weather forecast data are input periodically, the short term prediction for the monitored road surface condition is carried out based on the locale data, and the medium and long term predictions for the monitored road surface condition are conducted based on the Japan Meteorological Agency GPV weather forecast data.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention provides road surface condition data necessary for management (display of road conditions, spraying of snow-melting material) for prevention of slip accidents and the like in winter, from the current time to short-term (5 hours), medium-term (24 hours) and a long-term (72 hours) road surface state prediction system.

[0002]

2. Description of the Related Art Road managers such as the Ministry of Construction take measures to spray road freeze inhibitors for safety management of winter roads. At this time, long-term forecast (3 days ahead), medium-term forecast (1 day ahead), short-term forecast (5
It is said that it is necessary to predict the road surface state (time ahead). The road surface state prediction performs four state predictions of freezing, snow compaction, dryness, and wetness, and road surface temperature prediction.

[0003]

When road surface condition prediction is performed using only numerical calculation prediction data (GPV data or the like) distributed by the Japan Meteorological Agency or the like, the tendency of medium- to long-term prediction matches, but not that of short-term prediction. When performing road surface condition prediction using only local weather data, short-term prediction is possible, but medium- to long-term prediction is difficult.
An object of the present invention is to provide a system that rationally predicts a short- to medium-long term road surface condition by combining the two.

[0004]

In order to solve the above-mentioned problems, the invention according to claim 1 is characterized in that local data from a local weather data collection device comprising a road surface shape sensor of a monitored road surface and a group of weather sensors; Japan Meteorological Agency GPV (Grid Point Val
ue) periodically inputting weather forecast data, performing short-term prediction of the monitored road surface condition based on the local data, and performing medium-term and long-term prediction of the monitored road surface condition based on the Meteorological Agency GPV weather forecast data. It is characterized by.

[0005] Further, the invention according to claim 2 is characterized in that the short-term prediction, the medium-term prediction and the long-term prediction are four states of road surface freezing, road surface wetness, road surface snow compaction, and road surface drying.

Further, the invention according to claim 3 is characterized in that the short-term prediction, the medium-term prediction and the long-term prediction are four states of road surface freezing, road surface wetness, road surface snow compaction, and road surface drying.

Further, according to the invention of claim 4, in the local weather data collecting apparatus, there is no road surface convex state,
And it is determined that the road surface is frozen when the road surface has moisture and the road surface temperature is minus, the road surface is not in a convex state, and the road surface is wet when the road surface temperature is plus and the road surface temperature is positive, and the road surface is in the convex state and the road surface temperature is plus. It is distinguished as road surface wet / sherbet-like snow under the condition of 凸, road surface snow condition is determined under the condition of road surface convexity and negative road surface temperature, and road surface dry is determined under the condition of no road surface convex condition and no road surface moisture. And

[0008] In the invention according to claim 5, the short-term prediction predicts a road surface temperature by an extrapolation method based on a temperature change before a predetermined time, and calculates a road surface condition at the predicted road surface temperature and the predicted time. The road surface condition is determined based on

Further, in the invention according to claim 6, the medium-term and long-term forecasts are obtained by converting the temperature data of the Meteorological Agency GPV weather forecast data into a road surface temperature by a correction obtained by a statistical method, and estimating the road surface temperature. It is characterized by predicting road surface conditions based on a combination of temperature, temperature, precipitation, and precipitation type.

Further, in the invention according to claim 7, the medium-term and long-term predictions are characterized in that the short-term prediction is used as an initial value.

Further, the invention according to claim 8 is characterized in that a communication unit for transmitting the prediction result to a remote place via a wired or wireless communication line is provided.

[0012]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a conceptual diagram showing a system configuration according to an embodiment of the present invention. Reference numeral 1 denotes an information processing unit which is a main part of the system of the present invention, and includes an interface unit 1a and a prediction operation unit 1b. 2 is a local data collection unit, 3 is a GPV data distribution unit from the Meteorological Agency, etc.
Reference numeral 4 denotes the Internet, and reference numeral 5 denotes a receiving unit for determining / predicting result information.

In the information processing unit 1, an interface unit 1a includes local data M1 and GPV data M2 transmitted via the Internet 4 in the form of an electronic mail attachment.
And passes it to the prediction calculation unit 1b. The prediction calculation unit 1b performs a short-term, middle-term, and long-term road surface state determination / prediction calculation based on a predetermined algorithm, and determines a determination / prediction result M3.
Is transmitted from the interface unit 1a to the information receiving unit via the Internet 4 in the form of an electronic mail attachment.

An embodiment of the collection unit 2 for local data M1 will be described with reference to FIGS. Reference numeral 6 denotes a monitored road surface, and reference numeral 7 denotes a pole fixed to the side of the road surface. Reference numeral 8 denotes a road surface shape sensor mounted on the top of the pole 7, and in this embodiment, a level meter using a laser beam is used.
The difference ΔL between the dry level L0 of the road surface and the level L1 of the convex state due to snow or ice 9 is measured based on the reflection time of the irradiation laser beam B on the road surface 6.

Reference numeral 10 denotes an air temperature sensor attached to the pole 7, reference numeral 11 denotes a dew point sensor also attached to the pole 7, and reference numeral 12 denotes a temperature sensor embedded in the monitored road surface 6. Reference numeral 13 denotes a data transmission device, which inputs measurement signals from these sensor groups, normalizes the A / D conversion, and transmits the local data M1 to the distant data processing unit via the Internet in the form of an e-mail attachment at regular intervals. . 14 is W
It is an eb-compatible camera, and performs real-time monitoring of road conditions using a browser.

Reference numeral 13 in FIG. 4 indicates a component of the data transmission apparatus, and includes an A / D converter 13a, a data processing unit 13b, a data storage unit 13c, and a communication unit 13d. The measurement value ΔL of the laser type road surface shape sensor 3, the measurement value T1 of the air temperature sensor, the measurement value T2 of the dew point sensor, and the measurement value T3 of the road surface temperature sensor are input to the A / D converter 13a, and are normalized and digitized. Storage unit 13c
The data is temporarily stored in the data processing unit 13b, is collected into a predetermined file format by the data processing unit 13b, and is transmitted at regular intervals to the Internet via the communication unit 13d. Similarly, an image signal of the Web-compatible camera 14 is transmitted to the Internet.

Next, the local data M in the information processing section 1
The algorithm of the short-term prediction judgment by No. 1 will be described.
First, the real-time road surface determination algorithm determines from the measured value of the road surface shape sensor 8 that the road surface has a convex state when the change ΔL from the steady level of the monitored road surface 6 is equal to or more than a predetermined value β.

Next, the measured value T of the road surface temperature sensor 12
When the difference between 3 and the measured value T2 of the dew point sensor 11 is equal to or more than a predetermined value α, it is determined that the road surface has moisture.

Further, it is determined that the road surface is frozen under the condition that the road surface is not in a convex state, the road surface has moisture, and the measured T3 value of the road surface temperature sensor is negative. It is determined that the road surface is wet on the condition that there is no road surface convex state, there is road surface moisture, and the road surface temperature is plus.

Under the condition that the road surface is in a protruding state and the road surface temperature is positive, it is determined that the road surface is wet or sherbet-like snow. Under the condition that the road surface is in a convex state and the road surface temperature is minus, it is determined that the road surface is snow-covered. Road surface drying is determined under the condition that there is no road surface convex state and there is no road surface moisture.

The algorithm for short-term prediction uses the above-mentioned real-time local weather data M1 as an initial value. As for the road surface temperature, the temperature change one hour before the predicted time is extrapolated to the predicted time. However, the simple extrapolation uses an extrapolation method that takes into account the temperature rise due to solar radiation in the morning and the temperature fall in the evening.
The road surface state prediction is determined from the road surface temperature prediction and the presence or absence of rain (rain, sleet, snow) at the predicted time. The determination of the presence / absence of rain (rain, sleet, snow) is made by direct visual observation at the site or by monitoring information of a Web-compatible camera image by a browser.

(1) The prediction is made at every hour and the prediction time is 1
Predict the time, 2 hours, 3 hours, 4 hours, 5 hours ahead. (2) Prediction of road surface temperature Based on the local road surface temperature data, the road surface temperature up to 5 hours ahead is simply extrapolated. In the morning (at 7:00 and 8:00), extrapolation is performed in consideration of the rise in road surface temperature due to solar radiation. When the weather is sunny: road surface temperature rise = γ ° C / h (γ value can be changed as a parameter) When the weather is not sunny: road surface temperature rise = δ ° C / h (δ value can be changed as a parameter) Evening (15 At 16:00, 17:00, and 18:00), extrapolation is performed in consideration of the road surface temperature drop due to radiant cooling. When the weather is sunny: road surface temperature drop = γ ° C / h (γ value can be changed as a parameter) When the weather is not sunny: road surface temperature drop = δ ° C / h (δ value can be changed as a parameter)

FIG. 4 is a table showing a road surface condition determination result based on a determination condition applying a threshold value set to the predicted road surface temperature TF, corresponding to the road surface condition at the predicted time. Are predicted and determined.

Next, the Meteorological Agency GPV weather forecast data M2
Medium-term and long-term forecasts by using GIS will be described. At the Japan Meteorological Agency, as a result of numerical forecasts using a large computer,
It provides rainfall prediction data (no precipitation, rain, sleet, snow, etc.) at points corresponding to the intersections of meshes at intervals of 20 km throughout the country. This data is GPV (Grid Point Valu
It is called e), and the weather forecasting company purchases the necessary number of points for this data and adds its own analysis to make the weather forecast. This data can be purchased from a distributor in the form of an e-mail attachment.

When this GPV data is used for the prediction of the road surface condition of the road or the weather at the site of the road surface observation, the road surface condition observation point may be exactly at the intersection of the meshes. Surrounding 4 points or nearby 2
It is necessary to interpolate the data of the 6 points to which the points have been added, to further consider the altitude and the terrain, and to add the observation data at the site to correct the prediction value, thereby performing more accurate prediction.

In the present invention, first, the temperature data of the GPV point is converted into the local temperature by a statistical method. Next, the temperature is converted into the road surface temperature. Using the short-term forecast results as initial values, four-step prediction is performed using combinations of the predicted road surface temperature, air temperature, high water volume, and types of precipitation (no precipitation, rain, sleet, snow, etc.).

The mid-term prediction is performed twice a day (at 9 o'clock and 16 o'clock), and prediction is made up to 24 hours ahead at one-hour intervals. The long-term prediction is performed once a day (at 9:00), and prediction is made up to 72 hours ahead at one-hour intervals. However, the prediction from 1 hour to 24 hours ahead uses the medium-term prediction result. First, G close to the predicted location
The road surface temperature is calculated from the temperature of the PV data. Road surface temperature = GPV air temperature data + ζ Here, 変 え る changes depending on the time of day. The value of ζ is determined in advance for each time zone in consideration of local conditions.

Derivation of the precipitation prediction is performed by the following judgment formula based on the humidity data of GPV. Humidity data <80% → No precipitation Humidity data> 80% and temperature 2.50 ° C or more (settable) → Rain Humidity data> 80% and temperature 1.20 ° C to 2.50 ° C (settable) → Sleet Humidity data> 80% and temperature 1.20 Below (can be set) → Snow

FIG. 5 is a table showing a road surface state determination result based on a determination condition applying a threshold value set to a predicted road surface temperature TF, corresponding to information on a road surface state obtained by short-term prediction one hour before the predicted time. Yes, four states of wet, snow compaction, freezing, and drying are predicted and determined.

FIG. 6 shows a specific configuration example of the information processing section 1 shown in FIG. Local weather data and GPV weather data to which CSV format data M1 and M2 are attached are distributed via the Internet in the form of e-mail. The mail receiving interface 1a checks these mails at 5-minute intervals, stores the CSV data in the database 1c, and updates them periodically.

The arithmetic processing unit 1b executes short-term, medium-term, and long-term prediction calculations based on the data in the database 1c.
The prediction result is written to a mail attachment CSV file in the mail transmission interface 1a '. This data is distributed to the Internet as distribution data M3 by the mail transmission function in the form of an e-mail attached file.

Further, the arithmetic processing unit 1b outputs a daily report 1e on the result of the prediction operation based on the form format information from the master file 1d. The master file 1d has calculation parameter setting and mail CSV template setting functions. This file is modified when there is a change in the operation parameters or the like.

In implementing the present invention, a coefficient value specific to a target area is used in the prediction algorithm in both the short term and the medium term. Therefore, in order to determine this coefficient, test operation is required prior to actual operation of the system. The longer the trial operation period, the better, but it is possible to obtain a practical coefficient value even in one winter or one month.

[0034]

As described above, according to the present invention, the following effects can be expected. (1) By the short-term prediction (3 hours) of the road surface condition, the road manager can obtain support information for determining whether or not to execute measures such as spraying of a freeze inhibitor at night or the like on the day. (2) With the medium-term forecast (one day ahead), support information for judging whether or not to prepare for measures such as materials and staffing can be obtained. (3) With the long-term forecast (three days ahead), support information for preparing a preparation plan for materials and staffing can be obtained.

(4) During the test operation period, by collecting actual data while operating the system and feeding the result back to the redetermination of the coefficient value, it is expected that the accuracy of the road surface state prediction is further improved. (5) Although the coefficient values are specific to each region, the algorithm is considered to be common in a wide area. Therefore, by simply replacing the region-specific coefficient values in the system, a system suitable for the target region can be constructed. Becomes possible.

[Brief description of the drawings]

FIG. 1 is a conceptual diagram illustrating a system configuration according to an embodiment of the present invention.

FIG. 2 is a configuration diagram of a local weather data collection device according to another embodiment of the present invention.

FIG. 3 is a block diagram illustrating a configuration example of a data processing unit in FIG. 3;

FIG. 4 is a table showing determination results of short-term prediction according to the present invention.

FIG. 5 is a table showing a determination result of the medium-term and long-term predictions of the present invention.

FIG. 6 is a block diagram illustrating a specific configuration example of an information processing unit according to the present invention.

[Explanation of symbols]

 Reference Signs List 1 Information processing unit 2 Local data collection unit 3 GPV data distribution unit 4 Internet 5 Discrimination / prediction result receiving unit

Claims (8)

[Claims] 1. Periodically inputting local data from a local weather data collection device comprising a road surface shape sensor of a monitored road surface and a weather sensor group, and the Meteorological Agency GPV weather forecast data, and receiving data based on the local data. A road surface condition prediction system that performs short-term prediction of a monitored road surface condition and performs medium-term and long-term prediction of the monitored road surface condition based on the Meteorological Agency GPV weather forecast data. 2. The road surface condition discrimination system according to claim 1, wherein the short-term prediction, the medium-term prediction, and the long-term prediction include four states: road surface freezing, road surface wetness, road surface snow compaction, and road surface drying. 3. The on-site weather data collection device includes a road surface shape sensor for measuring a level change by a laser means, an air temperature sensor, a dew point sensor, and a road surface temperature sensor. It is determined that the road surface has a protruding state when the change of the road surface is equal to or greater than a predetermined value, and it is determined that the road surface has moisture when the difference between the measured value of the road surface temperature sensor and the measured value of the dew point temperature sensor is equal to or greater than a predetermined value. The road surface condition determination system according to claim 1 or 2, wherein: 4. The local weather data collection device,
It is determined that the road surface is frozen when there is no road surface convex state and there is road surface moisture and the road surface temperature is minus, and it is determined that the road surface is wet when there is no road surface convex state and there is road surface moisture and the road surface temperature is positive and there is a road surface convex state. And, when the road surface temperature is positive, the road surface is determined to be wet and sherbet-like snow, and when the road surface is convex, and when the road surface temperature is negative, the road snow is determined. The road surface condition prediction system according to any one of claims 1 to 3, wherein the determination is performed. 5. The short-term prediction predicts a road surface temperature by an extrapolation method based on a temperature change before a predetermined time with respect to current local data, and based on the predicted road surface temperature and a road surface state at a predicted time. The road surface state prediction system according to any one of claims 1 to 4, wherein the state is determined. 6. The medium-term and long-term forecast is based on the JMA G
The method is characterized in that the temperature data of the PV weather forecast data is converted and estimated into the local temperature and the road surface temperature by correction obtained by a statistical method, and the road surface condition is predicted by a combination of the road surface temperature, the temperature, the amount of precipitation, and the type of precipitation. Claims 1 to 5
The road surface condition prediction system according to any one of the above. 7. The road surface state prediction system according to claim 6, wherein the medium-term and long-term predictions use a short-term short-term prediction as an initial value. 8. The road surface state prediction system according to claim 1, further comprising a communication unit that transmits the prediction result to a remote place via a wired or wireless communication line.