JP6742013B1 - Asphalt mixture temperature measuring system, asphalt plant and asphalt mixture temperature measuring method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a temperature measuring system for an asphalt mixture, an asphalt plant, and a temperature measuring method for an asphalt mixture, capable of accurately measuring the temperature of an asphalt mixture without being affected by dust, water vapor and the like. SOLUTION: In an asphalt plant 1 that mixes asphalt and aggregate by a mixer 7 to produce an asphalt mixture, a temperature measurement system 8 that measures the temperature of the asphalt mixture in the mixer 7 The image pickup device 81 for picking up images at predetermined time intervals, and the calculation device 82 for calculating the temperature t of the asphalt mixture based on the plurality of picked-up images picked up by the image pickup device 81 are provided. The temperature t of the asphalt mixture is calculated based on the change that has occurred between the plurality of captured images. [Selection diagram]

Description

The present invention relates to a temperature measuring system and a temperature measuring method for an asphalt mixture, and an asphalt plant using the temperature measuring system.

BACKGROUND ART An asphalt plant that manufactures an asphalt mixture known as a road paving material mixes an aggregate that is heated and dried and an asphalt that is heated and melted in a liquid state with a mixer to produce an asphalt mixture. Asphalt changes in viscosity depending on the temperature, and the lower the temperature, the higher the viscosity.Therefore, the asphalt mixture shipped from the asphalt plant has a predetermined viscosity so that it can be spread to roads. It is laid on the road at a leveling temperature (for example, 110°C or higher). Therefore, the shipping temperature of the asphalt mixture is set in consideration of the temperature decrease until the asphalt mixture is transported to the construction site and spreading on the road is started.

In the asphalt plant, the temperature of the asphalt mixture in the mixer is measured as the shipping temperature of the asphalt mixture. A radiation thermometer such as an infrared thermometer that measures the surface temperature of the object to be measured without contact is used for the measurement of the shipping temperature (see, for example, Patent Document 1). The infrared thermometer includes a spot thermometer that measures the surface temperature of one point of the measured object and a thermography that measures the surface temperature of a wide range of the measured object as a temperature distribution.

JP, 10-018215, A

However, since the spot thermometer can measure only a partial temperature of the asphalt mixture, it is not reliable to use as a shipping temperature. Further, since the infrared light receiving portion of the spot thermometer has a small diameter of about several millimeters, it is easily affected by dust generated when the aggregate is put into the mixer, and water vapor generated when the aggregate and the asphalt are mixed, False measurement is likely to occur. In addition, the spot thermometer will not be able to measure the accurate temperature if dust or dirt adheres to the infrared receiver, so clean the infrared receiver regularly to ensure that the measured temperature matches the standard thermometer. Need to calibrate. On the other hand, in thermography, since the optical system for receiving infrared rays is larger than the infrared receiving section of the spot thermometer, it is unlikely to be affected by dust, water vapor, dirt, and the like. However, since the thermal image captured by the thermography has a low resolution, there is a problem that a measurement error increases when the thermal image is color-determined and the temperature is measured.

The problem to be solved by the present invention is, without being affected by dust, water vapor, etc., a temperature measuring system for an asphalt mixture capable of accurately measuring the temperature of an asphalt mixture, an asphalt plant, and a temperature measuring method for an asphalt mixture. Is to provide.

[1] A temperature measuring system for an asphalt mixture according to the present invention is a temperature measuring system for measuring a temperature of the asphalt mixture in the mixer in an asphalt plant for producing an asphalt mixture by mixing asphalt and aggregate with a mixer. And an imaging device that images the asphalt mixture in the mixer at predetermined time intervals, and a computing device that computes the temperature t of the asphalt mixture based on a plurality of captured images captured by the imaging device. , The arithmetic device, for each of the plurality of captured images, obtains a difference indicating the magnitude of movement of the asphalt mixture by comparing with the captured image captured immediately before, and digitizes the difference. An image analysis unit that calculates the fluctuation value and generates waveform data indicating the temporal change of the fluctuation value, and a characteristic value d that is a characteristic value indicating the characteristics of the waveform data by analyzing the waveform data. And a temperature calculation unit that calculates a temperature t of the asphalt mixture based on the characteristic value d, and the temperature calculation unit, as the characteristic value d, is the maximum value of the fluctuation values within a predetermined time. , An average value of a plurality of peaks within a predetermined time period, an average value of amplitudes within a predetermined time period, or the number of peaks within a predetermined time period is calculated, and the temperature t of the asphalt mixture increases as the characteristic value d increases. Thus, it is a temperature measurement system for an asphalt mixture that calculates the temperature t .

[2] Another asphalt mixture temperature measuring system according to the present invention is a temperature for measuring the temperature of the asphalt mixture in the mixer in an asphalt plant for producing an asphalt mixture by mixing asphalt and aggregate with a mixer. A measurement system, which is an imaging device that images the asphalt mixture in the mixer at predetermined time intervals, and an operation that calculates a temperature t of the asphalt mixture based on a plurality of captured images captured by the imaging device. And a device for calculating the difference indicating the magnitude of movement of the asphalt mixture by comparing the captured image captured immediately before with each of the plurality of captured images. An image analysis unit that calculates a numerically converted variation value to generate waveform data indicating a temporal change of the variation value, and a Fourier transform of the waveform data as a characteristic value d that is a characteristic value indicating the characteristic of the waveform data. A temperature calculation unit that calculates an array of a plurality of sine wave amplitudes having different angular frequencies, characteristic value data that is the characteristic value that indicates the characteristic of the waveform data that has been actually measured, and the characteristic value data And temperature data which is a pre-measured temperature of the asphalt mixture associated with, as the characteristic value data, a plurality of sine of different angular frequency obtained by Fourier transforming the pre-measured waveform data. Master data using an array of wave amplitudes, and the temperature calculation unit has a degree of similarity indicating a degree of similarity between the characteristic value d and the plurality of characteristic value data included in the master data. Is calculated, and the temperature data corresponding to the characteristic value data having the highest degree of similarity is set as the temperature t of the asphalt mixture, which is a temperature measuring system for the asphalt mixture.

[3] Another temperature measuring system for an asphalt mixture according to the present invention measures the temperature of the asphalt mixture in the mixer in an asphalt plant that mixes asphalt and aggregate with a mixer to produce an asphalt mixture. A temperature measuring system, wherein a temperature t of the asphalt mixture is calculated based on an imaging device that images the asphalt mixture in the mixer at predetermined time intervals and a plurality of captured images captured by the imaging device. An arithmetic unit, wherein the arithmetic unit obtains a difference indicating the magnitude of movement of the asphalt mixture by comparing each of the plurality of captured images with a captured image captured immediately before, and the difference Is calculated within a predetermined time as the characteristic value d which is a characteristic value indicating the characteristic of the waveform data. Maximum value of fluctuation value, average value of a plurality of peaks within a predetermined time, average value of amplitude within a predetermined time, or number of peaks within a predetermined time, or angular frequency obtained by Fourier transforming the waveform data Of a plurality of different sine wave amplitudes for calculating an array, a characteristic value data which is the characteristic value indicating the characteristic of the waveform data measured in advance, and a temperature measured in advance of the asphalt mixture. Temperature data, and a learned model that is generated by machine learning using teacher data as the teacher data and that uses the same characteristic value as the characteristic value d as the characteristic value data. The temperature measuring system for an asphalt mixture calculates the temperature t of the asphalt mixture by inputting the calculated characteristic value d to the learned model.

[4] In the above invention, the temperature calculation unit includes: a characteristic value data which is the characteristic value indicating characteristics of the waveform data previously measured, and the temperature data is a temperature of advance actually measured the asphalt mixture, The master data includes the master data, and the master data uses, as the characteristic value data, the characteristic value of the same kind as the characteristic value d, so that the larger the characteristic value data is, the higher the temperature data is. Characteristic value data and the temperature data are associated with each other, and the temperature calculation unit specifies the characteristic value data corresponding to the calculated characteristic value d from the master data, and specifies the specified characteristic value data. The temperature t of the asphalt mixture may be calculated based on the temperature data associated with .

[5] In the above invention, the arithmetic unit includes a database including a plurality of the master data, and the plurality of master data includes the characteristic value for each type of the asphalt mixture or the rotation speed of the mixer. Data and the temperature data may be associated with each other.

[6] In the above invention, the temperature calculation unit may calculate the temperature t of the asphalt mixture corresponding to the characteristic value d based on the following equation (1).

However, in the above formula (1),
n and n1 are the characteristic value data included in the master data,
n is the nth characteristic value data smaller than the characteristic value d analyzed from the waveform data,
n1 is the n1th characteristic value data larger than the characteristic value d,
m and m1 are the temperature data included in the master data,
m is the m-th temperature data corresponding to the n-th characteristic value data,
m1 is mth temperature data corresponding to the n1th characteristic value data.

[ 7 ] In the above invention, the image analysis unit may set an observation area in a region that does not include a mixing blade of the mixer in the captured image, and may obtain a difference in the observation area.

[ 8 ] In the above invention, the image analysis unit specifies, as a first pixel, a pixel in which the captured image is changed with respect to a captured image captured immediately before, based on a difference between the captured images, and A pixel in which the captured image has not changed with respect to the captured image captured immediately before is specified as a second pixel, and a difference image binarized by the first pixel and the second pixel is generated, and the difference image is generated. The variation value may be calculated based on the number of the first pixels of.

[ 9 ] In the above invention, the variation value may be calculated so as to increase as the number of the first pixels of the difference image increases.

[ 10 ] An asphalt plant according to the present invention is an asphalt plant for producing an asphalt mixture by mixing asphalt and aggregate, and a mixer for mixing the asphalt and the aggregate, and a temperature of the asphalt mixture. An asphalt plant equipped with a measuring system.

[ 11 ] The temperature measuring method for an asphalt mixture according to the present invention is a temperature measuring method for measuring the temperature of the asphalt mixture in the mixer in an asphalt plant for producing an asphalt mixture by mixing asphalt and aggregate with a mixer. The first step of imaging the asphalt mixture in the mixer at a predetermined time interval by an imaging device, and, for each of the plurality of the captured images, comparing the captured image captured immediately before with the asphalt. A second step of obtaining a difference indicating the magnitude of the movement of the mixture, calculating a variation value by digitizing the difference, and generating waveform data indicating a temporal variation of the variation value; and analyzing the waveform data. And a third step of calculating a characteristic value d, which is a characteristic value indicating the characteristics of the waveform data, and calculating the temperature t of the asphalt mixture based on the characteristic value d. The step of, as the characteristic value d, the maximum value of the fluctuation value within a predetermined time, the average value of a plurality of peaks within a predetermined time, the average value of the amplitude within a predetermined time, or the number of peaks within a predetermined time. computes the like temperature t of the characteristic value d larger the the asphalt mixture increases, the temperature measurement method of the asphalt mixture comprising calculating the temperature t.

[ 12 ] In the above invention, the third step includes calculating a temperature t of the asphalt mixture based on the calculated characteristic value d and master data provided in advance. may include a characteristic value data which is the characteristic value indicating characteristics of the waveform data previously measured, and the temperature data is a temperature of advance actually measured the asphalt mixture, wherein the master data, the characteristic value data As the characteristic value d, the characteristic value data and the temperature data are associated with each other so that the larger the characteristic value data is, the higher the temperature data is. The step of 3 specifies the characteristic value data corresponding to the calculated characteristic value d from the master data, and based on the temperature data associated with the specified characteristic value data, the asphalt mixture. The temperature t may be calculated .

[ 13 ] Another temperature measuring method for an asphalt mixture according to the present invention is a temperature for measuring a temperature of the asphalt mixture in the mixer in an asphalt plant for producing an asphalt mixture by mixing asphalt and aggregate with a mixer. In the measurement method, the first step of imaging the asphalt mixture in the mixer at a predetermined time interval by an imaging device, and for each of the plurality of the captured images, comparing with the captured image captured immediately before. A second step of obtaining a difference indicating the magnitude of movement of the asphalt mixture, calculating a variation value by digitizing the difference, and generating waveform data indicating a temporal variation of the variation value; As a characteristic value d, which is a characteristic value indicating a characteristic, a third step of calculating an array of amplitudes of a plurality of sine waves having different angular frequencies obtained by Fourier transforming the waveform data, and the calculated characteristic value. d, and a fourth step of calculating the temperature t of the asphalt mixture based on the pre-stored master data, wherein the master data has the characteristics of the previously measured waveform data. Characteristic value data that is the characteristic value shown, and temperature data that is a pre-measured temperature of the asphalt mixture associated with the characteristic value data, and as the characteristic value data, the previously measured waveform An array composed of the amplitudes of a plurality of sine waves having different angular frequencies obtained by Fourier transforming the data is used, and the fourth step includes the characteristic value d and the plurality of characteristic value data included in the master data. Is a method of measuring the temperature of an asphalt mixture, which comprises calculating a degree of similarity indicating the degree of similarity with the temperature data of the asphalt mixture, the temperature data corresponding to the characteristic value data having the highest degree of similarity. ..

[ 14 ] Another method for measuring the temperature of an asphalt mixture according to the present invention is, in an asphalt plant that mixes asphalt and aggregate by a mixer to produce an asphalt mixture, a temperature at which the temperature of the asphalt mixture in the mixer is measured. In the measurement method, the first step of imaging the asphalt mixture in the mixer at a predetermined time interval by an imaging device, and for each of the plurality of the captured images, comparing with the captured image captured immediately before. A second step of obtaining a difference indicating the magnitude of movement of the asphalt mixture, calculating a variation value by digitizing the difference, and generating waveform data indicating a temporal variation of the variation value; As the characteristic value d which is a characteristic value indicating the characteristic, the maximum value of the fluctuation value within a predetermined time, the average value of a plurality of peaks within a predetermined time, the average value of the amplitude within a predetermined time, or the peak within a predetermined time Or a third step of calculating an array of amplitudes of a plurality of sine waves having different angular frequencies obtained by Fourier transforming the waveform data, and the calculated characteristic value d. And a fourth step of calculating the temperature t of the asphalt mixture based on the learned model, the learned model having the characteristic value indicating the characteristic of the waveform data measured in advance. It is generated by machine learning using certain characteristic value data and temperature data that is the temperature measured in advance of the asphalt mixture as teacher data, and the characteristic value of the same kind as the characteristic value d is used as the characteristic value data. It is used, and the fourth step is a method for measuring the temperature of an asphalt mixture, which includes calculating the temperature t of the asphalt mixture by inputting the calculated characteristic value d to the learned model. ..

[ 15 ] In the above invention, the second step may include setting an observation area in a region of the captured image that does not include the mixing blade of the mixer, and obtaining a difference in the observation area.

According to the present invention, since the asphalt mixture is imaged by the imaging device, it is less affected by the adhesion of dust, water vapor, and dirt as compared with the spot thermometer. Moreover, since the temperature is measured based on the imaged image of the asphalt mixture imaged by the imaging device, the temperature can be measured more accurately than when the surface temperature is measured by thermography.

FIG. 1 is a schematic diagram showing a configuration of an asphalt plant and a temperature measurement system according to a first embodiment of the present invention. FIG. 2 is a frame image of the asphalt mixture in the mixer taken by the image pickup apparatus shown in FIG. FIG. 3 is an explanatory diagram showing a difference image generated by the image analysis unit of FIG. FIG. 4 is a graph showing an example of the waveform data generated by the image analysis unit shown in FIG. FIG. 5 is a graph for explaining a state in which an average value of peaks of a predetermined width including the maximum value of fluctuation values is obtained from the waveform data shown in FIG. FIG. 6 is a table showing the structure of the master data shown in FIG. FIG. 7 is a flowchart showing a processing procedure of the temperature measurement system shown in FIG. FIG. 8 is a graph showing an example of waveform data generated by the image analysis unit of the asphalt plant and the temperature measurement system according to the second embodiment of the present invention. FIG. 9 is a table showing a configuration of master data used in the temperature calculation unit of the asphalt plant and the temperature measurement system according to the second embodiment of the present invention. FIG. 10 is a graph showing an example of waveform data generated by the image analysis unit of the asphalt plant and the temperature measurement system according to the third embodiment of the present invention. FIG. 11 is a table showing a configuration of master data used in the temperature calculation unit of the asphalt plant and the temperature measurement system according to the third embodiment of the present invention. FIG. 12 is waveform data used for explaining a method of calculating the characteristic value d in the asphalt plant and the temperature measurement system according to the fourth embodiment of the present invention. FIG. 13 is a conceptual diagram showing a state in which the waveform data shown in FIG. 12 is decomposed by Fourier transform to obtain a plurality of sine waves. FIG. 14 is an explanatory diagram showing the amplitudes and angular frequencies of the plurality of sine waves shown in FIG. FIG. 15 is a conceptual diagram showing a state of obtaining an angular frequency array from the amplitudes and angular frequencies of the plurality of sine waves shown in FIG. FIG. 16 is a table showing a configuration of master data used in the temperature calculation unit of the asphalt plant and the temperature measurement system according to the fourth embodiment of the present invention. FIG. 17 is a graph showing an example of waveform data generated by the image analysis unit of the asphalt plant and the temperature measurement system according to the fourth embodiment of the present invention. FIG. 18 is a graph showing a characteristic value d composed of an angular frequency array obtained from the waveform data shown in FIG. FIG. 19 is a graph showing the angular frequency array of the master data shown in FIG. 20 is a conceptual diagram showing a calculation state of the similarity between the characteristic value d having the angular frequency array shown in FIG. 18 and the master data shown in FIG. FIG. 21 is a conceptual diagram showing a neural network included in the temperature calculation unit of the asphalt plant and the temperature measurement system according to the fifth embodiment of the present invention. FIG. 22 is a conceptual diagram showing a state in which machine learning is being performed by the neural network shown in FIG. FIG. 23 is a conceptual diagram showing a state in which the temperature t corresponding to the characteristic value d input by the neural network shown in FIG. 21 is output. FIG. 24 is a cross-sectional view showing a configuration for capturing an image of the asphalt mixture from the opening provided on the side surface of the mixing tank shown in FIG. FIG. 25 is a cross-sectional view showing a configuration for capturing an image of the asphalt mixture through a window portion having a transparent member provided on the side surface of the mixing tank shown in FIG.

<<First Embodiment>>
Hereinafter, a first embodiment to which a temperature measuring system for an asphalt mixture, an asphalt plant, and a temperature measuring method for an asphalt mixture according to the present invention are applied will be described with reference to the drawings. As shown in FIG. 1, the asphalt plant 1 includes a new aggregate supplying unit 2, a recycled aggregate supplying unit 3, a stone powder supplying unit 4, an asphalt supplying unit 5, an additive supplying unit 6, a mixer 7, a temperature measuring system 8 and The measurement result management device 9 and the like are provided. The new aggregate supply unit 2 is new aggregate, the recycled aggregate supply unit 3 is recycled aggregate, the stone powder supply unit 4 is stone powder, the asphalt supply unit 5 is new asphalt, and the additive supply unit 6 is addition for regeneration. The agents are supplied to the mixer 7, respectively. The mixer 7 mixes the supplied new aggregate, recycled aggregate, stone powder, new asphalt, and additive for regeneration to produce an asphalt mixture. The temperature measurement system 8 measures the shipping temperature of the asphalt mixture manufactured in the asphalt plant 1. The measurement result management device 9 manages the measurement result of the temperature of the asphalt mixture. The asphalt mixture produced by the asphalt plant 1 is loaded on a truck 10 and transported to a road construction site. In FIG. 1, the flow of aggregate, stone powder, new asphalt, additives, etc. is shown by a solid line, and the flow of image data, control signals, etc. is shown by a broken line.

The new aggregate supply unit 2 has a hot bottle 21 and a new aggregate measuring tank 22. The new aggregate is crushed stone, sand or the like which has not been used as an aggregate of an asphalt mixture. The new aggregate is supplied to the hot bottle 21 via a dryer, a screen, and the like (not shown). The inside of the hot bottle 21 is divided into, for example, five compartments, and each compartment constitutes a storage bin 21a. The new aggregate is heated and dried by a dryer, then sieved by a screen according to the classification according to the particle size, and stored in any of the five storage bins 21a according to the classification according to the particle size.

The new aggregate stored in each storage bin 21a is appropriately discharged to the new aggregate measuring tank 22 by opening the discharge port 21b of the storage bin 21a. The new aggregate introduced into the new aggregate measuring tank 22 is cumulatively measured by a measuring device (not shown) provided in the new aggregate measuring tank 22. When the weight of the new aggregate loaded in the new aggregate weighing tank 22 reaches the target amount, the hot bottle 21 receives an instruction from the new aggregate weighing tank 22 and closes the discharge port 21b of each storage bin 21a. State. Then, the new aggregate measured by the new aggregate measuring tank 22 is supplied to the mixer 7.

The recycled aggregate supply unit 3 includes a surge bin 31 and a recycled aggregate measuring tank 32. Recycled aggregate is used to recycle the aggregate contained in waste pavement material, and is composed of aggregate and old asphalt attached to the aggregate. Recycled aggregate obtained by crushing the waste pavement material with a crusher and heating and drying it with a recycling dryer is supplied to the surge bin 31. The surge bin 31 has a heat retention function, and stores the recycled aggregate heated by the regeneration dryer in a warm state. The recycled aggregate stored in the surge bin 31 is appropriately discharged to the recycled aggregate measuring tank 32 when the discharge port 31a of the surge bin 31 is opened. The recycled aggregate put into the recycled aggregate weighing tank 32 is weighed by a weighing device (not shown) provided in the recycled aggregate weighing tank 32. When the weight of the recycled aggregate charged into the recycled aggregate measuring tank 32 reaches the target amount, the surge bin 31 receives the instruction from the recycled aggregate measuring tank 32 and closes the discharge port 31a. Then, the recycled aggregate measured by the recycled aggregate measuring tank 32 is supplied to the mixer 7.

The stone powder supply unit 4 has a stone powder silo 41 and a stone powder measuring tank 42. Stone powder increases the viscosity of the asphalt mixture and fills the voids of the asphalt mixture as aggregate. The stone powder stored in the stone powder silo 41 is discharged to the stone powder measuring tank 42 by opening the discharge port 41a of the stone powder silo 41. The stone powder charged into the stone powder measuring tank 42 is measured by a measuring device (not shown) provided in the stone powder measuring tank 42. When the weight of the stone powder charged into the stone powder measuring tank 42 reaches the target amount, the stone powder silo 41 receives the instruction from the stone powder measuring tank 42 and closes the discharge port 41a of the stone powder silo 41. Then, the stone powder measured by the stone powder measuring tank 42 is supplied to the mixer 7.

The asphalt supply unit 5 has an asphalt tank 51 and an asphalt measuring tank 52. The new asphalt stored in the asphalt tank 51 having a heat retaining function is discharged to the asphalt measuring tank 52 when the discharge port 51a of the asphalt tank 51 is opened. The new asphalt put into the asphalt weighing tank 52 is weighed by a weighing device (not shown) provided in the asphalt weighing tank 52. When the weight of the new asphalt put into the asphalt measuring tank 52 reaches the target amount, the asphalt tank 51 receives the instruction from the asphalt measuring tank 52 and closes the discharge port 51a of the asphalt tank 51. Then, the new asphalt measured by the asphalt measuring tank 52 is supplied to the mixer 7.

The additive supply unit 6 has an additive tank 61 and an additive measuring tank 62. The additive restores the penetration etc. of the deteriorated old asphalt contained in the recycled aggregate. The regeneration additive stored in the additive tank 61 is discharged to the additive measuring tank 62 when the discharge port 61a of the additive tank 61 is opened. The regenerating additive charged in the additive measuring tank 62 is measured by a measuring device (not shown) provided in the additive measuring tank 62. When the weight of the additive for regeneration put into the additive measuring tank 62 reaches the target amount, the additive tank 61 receives the instruction from the additive measuring tank 62 and closes the discharge port 61a of the additive tank 61. State. Then, the additive for regeneration measured by the additive measuring tank 62 is supplied to the mixer 7. Note that the additive supply unit 6 may include a flow meter for measuring the flow rate of the additive for regeneration, instead of the additive measuring tank 62, and the additive for regeneration supplied to the mixer 7 using this flow meter. The agent may be weighed.

The mixer 7 includes a mixing tank 71 and a pair of mixing blades 72 that are rotationally driven in the mixing tank 71. The mixing tank 71 includes a new aggregate, a recycled aggregate, a stone aggregate, a new aggregate, a recycled aggregate, a stone aggregate, and an asphalt aggregate. And store the additive for regeneration. The pair of mixing blades 72 are arranged parallel to each other in the horizontal direction, and the blades provided on the outer periphery of the rotating shaft allow the new aggregate, recycled aggregate, stone powder, new asphalt, and additive for regeneration in the mixing tank 71. To produce an asphalt mixture. An openable/closable discharge port 71a is provided in the lower portion of the mixing tank 71, and the discharge port 71a is opened when the asphalt mixture is shipped.

The viscosity of asphalt changes depending on the temperature. The viscosity of asphalt mixture mainly composed of asphalt also changes with temperature, and the lower the temperature, the higher the viscosity. The asphalt mixture shipped from the asphalt plant 1 is spread on the road at a predetermined spreading temperature (for example, 110° C. or higher) so that a viscosity suitable for spreading on the road can be obtained. Therefore, when shipping the asphalt mixture, predict the temperature decrease of the asphalt mixture based on the season, weather, transportation time to the construction site, etc., and transport the asphalt mixture to the construction site and spread it on the road. The shipping temperature is set so as to reach a predetermined laying temperature when the start of the process. The shipping temperature of the asphalt mixture is adjusted by the temperature of the recycled aggregate, new aggregate and asphalt to be mixed.

The asphalt mixture temperature measuring system 8 measures the temperature of the asphalt mixture in the mixer 7 as the shipping temperature of the asphalt mixture. Thus, the temperature of the asphalt mixture can be measured before shipping, so that the shipping of the asphalt mixture having an inappropriate shipping temperature can be prevented. Further, by measuring the temperature of the asphalt mixture before shipment, the abnormality of the asphalt plant 1 can be detected early. For example, when the measured temperature is lower than or higher than the set shipping temperature, the dryer of the new aggregate supply unit 2 or the recycled aggregate supply unit 3 is out of order, and the heat retention function of the surge bin 31 and the asphalt tank 51 is maintained. It is possible to detect a failure in the early stage.

Based on the above-mentioned property that the viscosity of asphalt changes with temperature, the inventors of the present application have found that the intensity of movement of the asphalt mixture changes depending on the temperature of the asphalt mixture in the mixer 7. It was In particular, even with the same kind of asphalt mixture, the violent movement of the asphalt mixture caused by stirring with the mixing blade 72 of the mixer 7 is different between the low temperature and the high temperature. Specifically, at low temperatures, the viscosity of the asphalt mixture is small and slow, whereas at high temperatures, the viscosity of the asphalt mixture is low and the motion of the asphalt mixture is large and fast. Considering the area of the fixed point in the image of the asphalt mixture, the movement of the asphalt mixture as described above can be grasped as a wave of the asphalt mixture generated in the fixed point area. In this embodiment, the movement of the asphalt mixture is acquired as a wave based on such knowledge, and the temperature of the asphalt mixture is measured based on the pattern of the wave (that is, waveform data described below).

The temperature measurement system 8 images the asphalt mixture in the mixer 7 at predetermined time intervals, and calculates the temperature of the asphalt mixture based on the plurality of captured images. More specifically, the temperature of the asphalt mixture is calculated based on the change occurring between the plurality of captured images. In order to measure the temperature of the asphalt mixture by such a method, the temperature measurement system 8 includes an image pickup device 81 and a calculation device 82.

The imaging device 81 is a so-called digital camera that supports moving image shooting of a monochrome image or a color image. The imaging device 81 is fixed to the periphery of the mixer 7 so that the asphalt mixture in the mixing tank 71 can be imaged from diagonally above the mixer 7, and the asphalt mixture in the mixing tank 71 can be imaged at a fixed point. Has become. The imaging device 81 of the present embodiment performs high-speed moving image shooting at a frame rate of 60 frames (60 fps) per second, for example, in order to image the waves of the asphalt mixture in the mixer 7. The imaging device 81 and the arithmetic device 82 are connected by wire or wirelessly, and the captured moving images are sequentially output from the imaging device 81 to the arithmetic device 82. The frame rate of the imaging device 81 is not limited to the above 60 fps, and can be set to any value. For example, the imaging device 81 may be able to capture a moving image of 120 frames (120 fps) per second.

The calculation device 82 includes an image analysis unit 821, a temperature calculation unit 822, and a database 823. The image analysis unit 821 performs an image analysis process on each frame image of the moving image input from the imaging device 81. The temperature calculation unit 822 calculates the temperature of the asphalt mixture based on the result of the image analysis processing of the image analysis unit 821. The database 823 stores a plurality of master data 823 _{1 to} 823 _n used for calculating the temperature of the asphalt mixture by the temperature calculation unit 822.

The image analysis unit 821 sets an observation area in a region where the mixing blades 72 of the mixer 7 are not imaged with respect to the plurality of frame images input from the imaging device 81, and cuts out the image in the observation area from the frame image. FIG. 2 shows a frame image F1 as an example of the frame image. In this frame image F1, two lines L1 and L2 indicated by a two-dot chain line indicate the centers of the rotation axes of the pair of mixing blades 72 and 72. The pair of mixing blades 72, 72 rotate around the lines L1, L2 to mix the new aggregate, regenerated aggregate, stone powder, new asphalt, and regenerating additive. The image analysis unit 821 sets an observation area Oa in a region where the mixing blade 72 is not imaged, for example, a region between the pair of mixing blades 72, with respect to the frame image F1, and cuts out an image in the set observation area Oa. ..

The image analysis unit 821 compares the observation area Oa with the frame image captured immediately before for each of the plurality of frame images to obtain a difference. Here, the more intense the movement of the asphalt mixture (the larger and faster the movement), the more pixels there are changes in the observation area Oa, and the slower the movement of the asphalt mixture (the smaller and the slower the movement) the observation area Oa. There are few pixels that change in. Therefore, in this embodiment, the difference is obtained for each of the plurality of frame images. Hereinafter, this is referred to as calculation of a difference between frame images. That is, the image analysis unit 821 calculates the difference between the observation areas Oa of the first frame image and the second frame image, and then the observation areas Oa of the second frame image and the third frame image. For the observation area Oa between a plurality of frame images.

The image analysis unit 821 generates a difference image based on the result of the difference calculation performed on the observation area Oa of the plurality of frame images. Specifically, for example, based on the difference between the observation area Oa of the first frame image and the observation area Oa of the second frame image, the observation area Oa of the second frame image is A pixel that has changed with respect to the observation area Oa of the first frame image captured is identified as a first pixel, and a pixel that has not changed is identified as a second pixel. Then, a difference image is generated by binarizing the first pixel and the second pixel with, for example, white and black. FIG. 3 shows the difference image Di as an example of the difference image. In this difference image Di, the black pixel indicates the first pixel, and the white pixel indicates the second pixel.

Next, the image analysis unit 821 calculates the magnitude of the difference, that is, the variation value indicating the magnitude of the movement of the asphalt mixture, based on the generated difference image. The image analysis unit 821 calculates, for example, the number of first pixels of the difference image, and calculates the variation value based on the calculated number of pixels. This variation value is calculated so as to increase as the number of first pixels (black pixels in FIG. 3) of the difference image increases.

The image analysis unit 821 obtains a numerical value group including a plurality of fluctuation values by calculating a difference between a plurality of frame images, generating a difference image, and calculating a fluctuation value, and obtains waveform data indicating a temporal change of the fluctuation value. To generate. FIG. 4 shows the waveform data Wd1 as an example of the waveform data. In the waveform data Wd1, the vertical axis represents the fluctuation value and the horizontal axis represents the time. It can be seen from the waveform data Wd1 that the asphalt mixture periodically undulates according to the mixing operation of the mixer 7.

The temperature calculation unit 822 analyzes the waveform data and obtains the characteristic value d indicating the characteristics of the waveform data. The temperature calculation unit 822 of the present embodiment obtains, for example, the average value of the peaks included in the range of the predetermined width including the maximum value of the variation value from the waveform data within the predetermined time, and sets the average value as the characteristic value d. To do. As described above, as the temperature of the asphalt mixture becomes high, the viscosity of the asphalt mixture becomes low, so that the movement of the asphalt mixture becomes large. Therefore, the temperature t of the asphalt mixture can be accurately calculated by using the average value of the peaks of the fluctuation values representing the magnitude of movement of the asphalt mixture as the characteristic value d. Here, an example of the predetermined time is not particularly limited, but is 800 unit time in the example shown in FIG. One unit time depends on the frame rate of the image pickup device 81. For example, when the frame rate of the image pickup device 81 is 60 fps, one unit time is 1/60 sec, and 800 unit time is about 13.3 sec.

For example, in the example shown in FIG. 5, the temperature calculation unit 822 specifies the maximum value P1 of the variation value from the waveform data Wd1 within the predetermined time, and sets the maximum value width W1 including the maximum value P1. This maximum value width W1 may use a preset value, or may be set according to the type of asphalt mixture, the rotation speed of the mixer 7, and the like. The temperature calculation unit 822 calculates the average value of the peaks (for example, P1 to P5) of the fluctuation values included in the maximum value width W1 and sets the average value as the characteristic value d. For example, when the average value of the fluctuation value peaks P1 to P5 is “10100”, this average value becomes the characteristic value d.

When obtaining the average value of the peaks of the fluctuation value from the waveform data Wd1, for example, the average value of several peaks (for example, 5 peaks) may be obtained from the maximum value without setting the maximum value width W1. Further, instead of the average value, the maximum value itself of the variation value in the waveform data Wd1 may be used as the characteristic value d.

The temperature calculator 822 calculates the temperature t of the asphalt mixture based on the characteristic value d of the waveform data. Specifically, the temperature calculation unit 822 first selects, from the database 823, master data matching the type of asphalt mixture being manufactured and the rotation speed of the mixer 7 from the plurality of master data 823 _{1 to} 823 _n. .. As shown in FIG. 6, the master data 823 _{1 to} 823 _n are registered with characteristic value data indicating the characteristics of the waveform data by the magnitude of the variation value and the temperature data of the asphalt mixture. The data and the temperature data are associated with each other. The master data 823 _{1 to} 823 _n are data measured in advance for each combination of the type of asphalt mixture and the rotation speed of the mixer 7, and the larger the characteristic value data, that is, the larger the movement of the asphalt mixture. High temperature data for asphalt mix. The master data 823 _{1 to} 823 _n are actually measured using an image pickup apparatus having the same frame rate as the image pickup apparatus 81. Then, the temperature calculation unit 822 refers to the selected master data (for example, the master data of FIG. 6), and when the characteristic value d obtained from the waveform data Wd1 is “11600”, the temperature t of the asphalt mixture is “170°. C". Further, when the characteristic value d is "8700", the temperature t of the asphalt mixture is specified to be "140°C".

When the characteristic value d is not registered in the selected master data, the temperature calculation unit 822 first corresponds to the characteristic value data n smaller than the characteristic value d and the characteristic value data n from the master data. Temperature data m, characteristic value data n1 larger than the characteristic value d, and temperature data m1 corresponding to the characteristic value data n1 are extracted. Next, the temperature t of the asphalt mixture corresponding to the characteristic value d is calculated using the following mathematical expression (2). The mathematical expression (2) corresponds to the mathematical expression (1) of the present invention.

However, in the above equation (2), n and n1 are characteristic value data registered in the master data, and n is the nth characteristic value data smaller than the characteristic value d analyzed from the waveform data. Yes, n1 is the n1th characteristic value data larger than the characteristic value d. Further, m and m1 are temperature data of the asphalt mixture registered in the master data, m is mth temperature data corresponding to the nth characteristic value data, and m1 is the n1th characteristic. It is the m-th temperature data corresponding to the value data.

For example, when the characteristic value d is d=10100 obtained from the waveform data Wd1 in FIG. 5, the characteristic value data n smaller than the characteristic value d corresponds to 9800 and the characteristic value data n from the master data shown in FIG. The temperature m is 150° C., the characteristic value data n1 larger than the characteristic value d is 10800, and the temperature m1 corresponding to the characteristic value data n1 is 160° C. By substituting these values obtained from the master data into the above equation (2) and calculating, the temperature t of the asphalt mixture corresponding to the characteristic value d becomes 153°C.

The measurement result management device 9 includes a so-called personal computer, a display connected to the personal computer, an input device such as a keyboard and a mouse, a printer, and the like. The measurement result management device 9 manages the temperature t of the asphalt mixture output from the temperature measurement system 8 as a shipping temperature associated with a manufacturing lot number and the like by a data management program. Further, the measurement result management device 9 displays the managed data such as the shipping temperature on the display according to the operation of the user. Furthermore, when shipping the asphalt mixture, the measurement result management device 9 prints a shipping slip in which the type of asphalt mixture, the shipping amount, the shipping temperature, and the like are described. The shipping slip is delivered to the construction site along with the asphalt mixture.

Next, the operation of this embodiment will be described with reference to the flowchart shown in FIG. 7. The mixer 7 includes a new aggregate, a recycled aggregate supply unit 3, a stone powder supply unit 4, an asphalt supply unit 5, and an additive supply unit 6 for supplying new aggregate, recycled aggregate, stone powder, new asphalt, and regeneration. When the additive is supplied, the pair of mixing blades 72, 72 are rotated and mixed to produce an asphalt mixture. The temperature measurement system 8 starts the imaging of the asphalt mixture in the mixing tank 71 by the imaging device 81 after a lapse of a predetermined time from the start of mixing by the mixer 7 (step S1). The moving images of the asphalt mixture captured by the image capturing device 81 are sequentially output to the computing device 82.

The image analysis unit 821 of the arithmetic device 82 sets the observation area Oa in the plurality of frame images of the moving image input from the imaging device 81, and cuts out the image in the observation area Oa (step S2). The image analysis unit 821 calculates the difference in the observation area Oa between each of the plurality of frame images and the frame image captured immediately before. The image analysis unit 821 changes the observation area Oa of the first frame image captured immediately before based on the result of the difference calculation (step S3) performed on the observation area Oa of the plurality of frame images. The difference image binarized by the first pixel that is present and the second pixel that is not changed is generated (step S4). Next, the image analysis unit 821 calculates a variation value based on the number of first pixels of the generated difference image (step S5).

The image analysis unit 821 obtains a numerical value group consisting of a plurality of fluctuation values by calculating a difference between a plurality of frame images, generating a difference image, and calculating a fluctuation value, and calculates the temporal change of these fluctuation values. The waveform data Wd1 shown is generated (step S6).

The temperature calculation unit 822 analyzes the waveform data Wd1 and obtains the characteristic value d indicating the characteristics of the waveform data Wd1 (step S7). The temperature calculation unit 822 obtains, for example, the average value of the peaks included in the range of the predetermined width including the maximum value of the variation value from the waveform data Wd1 within the predetermined time, and sets the average value as the characteristic value d.

The temperature calculation unit 822 refers to the master data selected from the database 823, and specifies the temperature t of the asphalt mixture corresponding to the characteristic value d (step S8). When the characteristic value d is not registered in the selected master data, the temperature calculation unit 822 calculates the temperature t of the asphalt mixture corresponding to the characteristic value d using the above formula (2). ..

The computing device 82 outputs the measurement result of the temperature t of the asphalt mixture to the measurement result management device 9 (step S9). The measurement result management device 9 manages the temperature t of the asphalt mixture output from the temperature measurement system 8 as a shipping temperature associated with a manufacturing lot number and the like by a data management program. Further, the measurement result management device 9 prints a shipping slip in which the type of the asphalt mixture, the shipping amount of the asphalt mixture, the shipping temperature, etc. are described. The shipping slip is delivered to the construction site together with the asphalt mixture loaded on the truck 10.

As described above, according to the temperature measurement system 8, the temperature measurement method, and the asphalt plant 1 for the asphalt mixture of the present embodiment, the image pickup device 81 for picking up the image of the asphalt mixture in the mixer 7 and the image pickup device 81 for picking up the image. And a computing device 82 that computes the temperature t of the asphalt mixture based on the moving image. The computing device 82 computes the temperature t of the asphalt mixture based on the changes that have occurred between the plurality of frame images. .. As described above, since the temperature of the asphalt mixture is measured by the image pickup device 81, the influence of dust, water vapor, and dirt is less likely to occur than in the conventional spot thermometer. Further, since the temperature t of the asphalt mixture is calculated based on the change generated between the plurality of frame images, the temperature t can be measured more accurately than the case where the surface temperature is measured by thermography or the like.

Further, according to the present embodiment, the arithmetic device 82 obtains the difference for each of the plurality of frame images by comparing it with the frame image captured immediately before, and calculates the variation value by digitizing the difference to obtain the variation value. The image analysis unit 821 that generates the waveform data Wd1 indicating the temporal change of the temperature and the temperature calculation unit 822 that calculates the temperature t of the asphalt mixture based on the waveform data Wd1. As described above, since the waveform data of the present embodiment shows the wave pattern of the asphalt mixture, the temperature t of the asphalt mixture is calculated based on the waveform data Wd1 to determine the wave pattern of the asphalt mixture. Thus, the temperature t of the asphalt mixture can be calculated. Therefore, the temperature t can be measured more accurately than when the surface temperature is measured by thermography or the like.

Further, according to the present embodiment, the temperature calculation unit 822 analyzes the waveform data Wd1 to calculate the characteristic value d indicating the characteristics of the waveform data, and calculates the temperature t of the asphalt mixture based on the characteristic value d. To do. As a result, the processing load can be reduced as compared with the case where the temperature t of the asphalt mixture is calculated from the waveform data Wd1 itself. In addition, since the temperature t of the asphalt mixture is calculated only from the characteristic value d indicating the characteristics of the waveform data, it is possible to calculate by excluding the non-characteristic portion of the waveform data, and the temperature t of the asphalt mixture can be accurately measured. can do.

Further, according to the present embodiment, the temperature calculation unit 822 calculates the characteristic value data that is the characteristic value indicating the characteristics of the waveform data Wd1 and the temperature data that is the temperature of the asphalt mixture associated with the characteristic value data. It has the master data 823 _{1 to} 823 _n that it has, and the temperature calculation unit 821 calculates the temperature t of the asphalt mixture based on the calculated characteristic value d and the master data. As described above, since the master data 823 _{1 to} 823 _n are provided in advance, the temperature t of the asphalt mixture can be quickly calculated with a low processing load. In addition, since the master data 823 _{1 to} 823 _n are data measured in advance for each combination of the type of asphalt mixture and the rotation speed of the mixer 7, the temperature t of the asphalt mixture is accurately measured based on the measured value. Can be measured. Furthermore, the asphalt mixture has different viscosity characteristics with respect to temperature depending on the type of asphalt used, the mixing ratio with the aggregate, the addition amount of the regeneration additive, and the like. Therefore, by providing a plurality of master data 823 _{1 to} 823 _n for each type of asphalt mixture or the rotation speed of the mixer 7, the temperature of the asphalt mixture can be measured more accurately. Further, in the master data 823 _{1 to} 823 _n , the higher the characteristic value data, the higher the temperature data according to the temperature characteristics of the asphalt mixture. Therefore, the temperature of the asphalt mixture can be accurately measured by using the master data 823 _{1 to} 823 _n .

Further, according to the present embodiment, the temperature calculation unit 822 calculates the characteristic value d based on the peak of the waveform data Wd1. As described above, the waveform data of the present embodiment shows the wave pattern of the asphalt mixture, but the peak of the waveform data shows the wave pattern most characteristically. Therefore, by calculating the characteristic value d based on the peak of the waveform data Wd1, the temperature t of the asphalt mixture can be accurately calculated based on the wave pattern of the asphalt mixture.

Further, according to the present embodiment, the average value of a plurality of peaks of the fluctuation value within the unit time is used as the characteristic value d. As described above, as the temperature of the asphalt mixture becomes high, the viscosity of the asphalt mixture becomes low, so that the movement of the asphalt mixture becomes large. Therefore, the temperature t of the asphalt mixture can be accurately calculated by using the average value of the peaks of the fluctuation values representing the magnitude of movement of the asphalt mixture as the characteristic value d.

Further, according to the present embodiment, the image analysis unit 821 sets the observation area Oa in a region that does not include the mixing blade 72 of the mixer 7 in the frame image, and obtains the difference within the observation area Oa. As a result, the processing load can be reduced as compared with the case where the difference between the entire frame images is obtained, so that the temperature of the asphalt mixture can be measured quickly.

<<Second Embodiment>>
Next, a second embodiment to which the temperature measuring system for asphalt mixture, the asphalt plant, and the temperature measuring method for asphalt mixture of the present invention are applied will be described based on the drawings. In this embodiment, the method of calculating the characteristic value d by the temperature calculation unit 822 and the contents of the database are different from those in the above-described first embodiment, but the other configurations are the same as in the first embodiment. The same components as those in the first embodiment are designated by the same reference numerals and detailed description thereof will be omitted.

In this embodiment, the temperature t of the asphalt mixture is calculated based on the difference in the amplitude of the waveform of the waveform data. In the asphalt mixture, the size of the movement and the time required for the movement to stop change depending on the viscosity. Specifically, when the viscosity of the asphalt mixture is low, the movement of the asphalt mixture is increased and the movement is quickly converged, so that the amplitude is increased. On the contrary, when the viscosity of the asphalt mixture is high, the movement of the asphalt mixture becomes small and the convergence of the movement becomes gentle (slow), so that the amplitude becomes small. In this embodiment, paying attention to this point, the temperature t of the asphalt mixture is calculated based on the amplitude of the waveform of the waveform data within the predetermined time.

FIG. 8 shows the waveform data Wd2 as an example of the waveform data used in this embodiment. The temperature calculation unit 822 extracts the amplitude including the peak and the bottom continuous to the peak from the waveform data Wd2, and calculates the difference between the peak fluctuation value and the bottom fluctuation value as the amplitude value. Specifically, the temperature calculation unit 822 extracts four amplitudes A1 to A4 from the waveform data Wd2 and calculates the difference between the fluctuation values between the peak and bottom of each amplitude A1 to A4 as the amplitude values Ad1 to Ad4. To do. Then, the temperature calculation unit 822 calculates the average value of the amplitude values Ad1 to Ad4 as the characteristic value d. For example, when the average value of the amplitude values Ad1 to Ad4 is “8480”, this average value becomes the characteristic value d.

The temperature calculator 822 calculates the temperature t of the asphalt mixture based on the characteristic value d of the waveform data Wd2 that is the average value of the amplitude values. Specifically, the temperature calculation unit 822 first selects, from the database 823, master data matching the type of asphalt mixture being manufactured and the rotation speed of the mixer 7 from the plurality of master data 823 _{1 to} 823 _n. .. As shown in FIG. 9, characteristic data indicating the characteristics of the waveform data Wd2 by the magnitude of amplitude and temperature data of the asphalt mixture are registered in the master data 823 _{1 to} 823 _n . The data and the temperature data are associated with each other. The master data 823 _{1 to} 823 _n are data measured in advance for each combination of the type of asphalt mixture and the rotation speed of the mixer 7, and the larger the characteristic value data, that is, the larger the movement of the asphalt mixture. High temperature data for asphalt mix. The master data 823 _{1 to} 823 _n are actually measured using an image pickup apparatus having the same frame rate as the image pickup apparatus 81, as in the first embodiment. Then, the temperature calculation unit 822 refers to the selected master data (for example, the master data of FIG. 9), and when the characteristic value d obtained from the waveform data Wd2 is “9640”, the temperature t of the asphalt mixture is “170°. C". Further, when the characteristic value d is “6820”, the temperature t of the asphalt mixture is specified to be “140° C.”.

When the characteristic value d is not registered in the selected master data, the temperature calculation unit 822 first corresponds to the characteristic value data n smaller than the characteristic value d and the analysis value data n from the master data. Temperature data m, analysis value data n1 larger than the characteristic value d, and temperature data m1 corresponding to the characteristic value data n1 are extracted. Next, the temperature t of the asphalt mixture corresponding to the characteristic value d is calculated using the above equation (2). For example, when the characteristic value d is “8480”, “153.4° C.” is calculated as the temperature t of the asphalt mixture by using the master data shown in FIG. 9 and the above equation (2). be able to.

As described above, according to the temperature measuring system 8, the temperature measuring method, and the asphalt plant 1 of the asphalt mixture of the present embodiment, the amplitude obtained from the waveform data Wd2 is used as the characteristic value d. The temperature t of the asphalt mixture can be calculated with high accuracy based on the magnitude of the above and the speed of convergence of the movement. Further, since the average value of the amplitude is used, the temperature t of the asphalt mixture can be accurately calculated even when the wave changes with the passage of time. Although the average amplitude is used as the characteristic value d, the maximum value of the amplitude may be used as the characteristic value d.

<<Third Embodiment>>
Next, a third embodiment to which the asphalt mixture temperature measuring system, the asphalt plant and the asphalt mixture temperature measuring method of the present invention are applied will be described with reference to the drawings. In this embodiment, the method of calculating the characteristic value d by the temperature calculation unit 822 and the contents of the database are different from those in the above-described first embodiment, but the other configurations are the same as in the first embodiment. The same components as those in the first embodiment are designated by the same reference numerals and detailed description thereof will be omitted.

In the present embodiment, when the temperature is low and the viscosity is high, the asphalt mixture exhibits a wave pattern of the same cycle as the mixing blade 72 of the mixer 7, but when the viscosity becomes low, the asphalt mixture is extruded by the mixing blade 72 and the Also shows a wave pattern of a short cycle, and is a calculation method focusing on the fact that the cycle (that is, the speed of movement of the asphalt mixture) is different. Specifically, the number of peaks of the fluctuation value included in the waveform data within a predetermined time (frequency per predetermined time) is obtained, and this frequency is set as the characteristic value d.

FIG. 10 shows the waveform data Wd3 as an example of the waveform data used in this embodiment. The temperature calculation unit 822 obtains the number of peaks included in the predetermined width range from the waveform data Wd3. Specifically, the peak included in the range of the predetermined width is, for example, the maximum value of the variation value (for example, “10100” in FIG. 10) and the minimum value of the variation value (for example, “0” in FIG. 10). Is calculated, and the number of peaks equal to or more than the intermediate value between the maximum value and the minimum value (for example, “5050” in FIG. 10) is counted. Further, the predetermined time for obtaining the number of peaks is set based on the frame rate of moving image shooting and the operation time of the mixer 7. For example, in the present embodiment, when the frame rate for capturing a moving image is 60 fps and the operation time of the mixer 7 is 30 seconds, "1800", which is the number of frame images captured in this 30 seconds, is set as the predetermined time. In the waveform data Wd3 shown in FIG. 10, only up to 800 unit time is shown. Therefore, for the sake of convenience, the number of peaks (20) within 600 unit time is counted, and a value obtained by multiplying the count value by 3 (60) is calculated. It was calculated as the number of peaks included within a predetermined time. In the present embodiment, this peak number “60” becomes the characteristic value d.

The temperature calculator 822 calculates the temperature t of the asphalt mixture based on the characteristic value d of the waveform data Wd3 including the number of peaks per predetermined time. Specifically, the temperature calculation unit 822 first selects, from the database 823, master data matching the type of asphalt mixture being manufactured and the rotation speed of the mixer 7 from the plurality of master data 823 _{1 to} 823 _n. .. As shown in FIG. 11, characteristic data indicating the characteristics of the waveform data Wd3 by the number of peaks included within a predetermined time and temperature data of the asphalt mixture are registered in the master data 823 _{1 to} 823 _n . The characteristic value data and the temperature data are associated with each other. The master data 823 _{1 to} 823 _n are data measured in advance for each combination of the type of asphalt mixture and the rotation speed of the mixer 7. The larger the characteristic value data, that is, the wave cycle of the asphalt mixture is. The shorter (the faster the asphalt mixture moves), the higher the temperature data of the asphalt mixture. The master data 823 _{1 to} 823 _n are actually measured using an image pickup apparatus having the same frame rate as the image pickup apparatus 81, as in the first embodiment. Then, the temperature calculation unit 822 refers to the selected master data (for example, the master data of FIG. 11), and when the characteristic value d obtained from the waveform data Wd3 is “62.6”, the temperature t of the asphalt mixture is “ 180°C". Further, when the characteristic value d is "57.2", the temperature t of the asphalt mixture is specified to be "130°C".

When the characteristic value d is not registered in the selected master data, the temperature calculation unit 822 first corresponds to the characteristic value data n smaller than the characteristic value d and the characteristic value data n from the master data. Temperature data m, characteristic value data n1 larger than the characteristic value d, and temperature data m1 corresponding to the characteristic value data n1 are extracted. Next, the temperature t of the asphalt mixture corresponding to the characteristic value d is calculated using the above equation (2). For example, when the characteristic value d is “60”, “155.4° C.” is calculated as the temperature t of the asphalt mixture by using the master data shown in FIG. 11 and the above equation (2). be able to.

As described above, according to the temperature measurement system 8, the temperature measurement method, and the asphalt plant 1 of the asphalt mixture of the present embodiment, the number of peaks included in the predetermined time, which is obtained from the waveform data Wd3, as the characteristic value d. Therefore, the temperature t of the asphalt mixture can be accurately calculated based on the wave cycle of the asphalt mixture (that is, the moving speed of the asphalt mixture).

<<4th Embodiment>>
Next, a fourth embodiment to which the asphalt mixture temperature measuring system, the asphalt plant, and the asphalt mixture temperature measuring method of the present invention are applied will be described with reference to the drawings. In this embodiment, the method of calculating the characteristic value d by the temperature calculation unit 822 and the contents of the database are different from those in the above-described first embodiment, but the other configurations are the same as in the first embodiment. The same components as those in the first embodiment are designated by the same reference numerals and detailed description thereof will be omitted.

In the present embodiment, the temperature of the asphalt mixture is calculated using the characteristic of the waveform obtained by Fourier transforming the waveform data, that is, the characteristic of the wave of the asphalt mixture as the characteristic value d. As is well known, the Fourier transform can capture waveform data as a combination of a plurality of sine waves and decompose the waveform data into a plurality of sine waves. In the present embodiment, the waveform data is decomposed into a plurality of sine waves by Fourier transform, the amplitude and the angular frequency are acquired from each sine wave, and the angular frequency array is used as the characteristic value d of the waveform data. Below, the method of acquiring an angular frequency array from waveform data using Fourier transform is demonstrated.

The complex waveform data Wd4 shown in FIG. 12 is created by combining sinusoidal waves (sin waves) Wd4a, Wd4b and Wd4c as shown in FIG. Therefore, when the waveform data Wd4 is Fourier-transformed and decomposed, sine waves Wd4a, Wd4b, and Wd4c can be obtained. The sine wave can be represented by “r sin ω x ”, as is well known. In addition, "r" shows an amplitude and "ω" shows an angular frequency. As shown in FIG. 14, the temperature calculation unit 822 of the present embodiment obtains the amplitude r and the angular frequency ω from the sine waves Wd4a, Wd4b, and Wd4c obtained by Fourier transforming the waveform data Wd4, and obtains the angular frequency array. .. For example, the amplitude r and the angular frequency ω of the sine wave Wd4a are “amplitude 5” and “angular frequency 2”. The amplitude r and the angular frequency ω of the sine wave Wd4b are “amplitude 4” and “angular frequency 5”. Similarly, the amplitude r and the angular frequency ω of the sine wave Wd4c are “amplitude 3” and “angular frequency 7”. This angular frequency array can be represented by a graph G1 in which the vertical axis represents amplitude and the horizontal axis represents angular frequency, as shown in FIG. That is, the angular frequency array of the sine waves Wd4a, Wd4b, and Wd4c is {0,5,0,0,4,0,3,0}.

The temperature calculation unit 822 calculates the temperature t of the asphalt mixture based on the characteristic value d of the waveform data Wd4 including the angular frequency array. Specifically, the temperature calculation unit 822 first selects, from the database 823, master data matching the type of asphalt mixture being manufactured and the rotation speed of the mixer 7 from the plurality of master data 823 _{1 to} 823 _n. .. In the master data 823 _{1 to} 823 _n , as shown in FIG. 16, characteristic value data composed of an angular frequency array showing the characteristics of the waveform data and temperature data of the asphalt mixture are registered. And temperature data are associated with each other. The master data 823 _{1 to} 823 _n are data measured in advance for each combination of the type of asphalt mixture and the rotation speed of the mixer 7. The master data 823 _{1 to} 823 _n are actually measured using an image pickup apparatus having the same frame rate as the image pickup apparatus 81, as in the first embodiment.

The temperature calculator 822 calculates the degree of similarity between the characteristic value d obtained from the waveform data Wd4 and the characteristic value data of the selected master data, and obtains the temperature data corresponding to the characteristic value data most similar to the characteristic value d. Let the temperature t of the asphalt mixture. The method of calculating the similarity between the characteristic value d and the characteristic value data can be exemplified by, for example, a method of obtaining the Euclidean distance between the characteristic value d and the characteristic value data described above. The method of calculating the degree of similarity is not particularly limited to the above, and a function for obtaining the degree of similarity of Jaccard or the similarity of sequences included in a programming language, for example, similar() may be used.

Hereinafter, an example will be described in which the temperature t of the asphalt mixture is obtained based on the characteristic value d formed by the angular frequency array of the actual waveform data of the asphalt mixture. FIG. 17 shows the waveform data Wd5 as an example of the waveform data used in this embodiment. The temperature calculation unit 822 performs a Fourier transform on the waveform data Wd5 to decompose it into seven sine waves, obtains the amplitude r and the angular frequency ω of each sine wave, and acquires the characteristic value d composed of the angular frequency array. Specifically, the angular frequency array of the waveform data Wd5 is, for example, {0, 5, 1, 3, 4, 3, 2, 1} as shown in the graph G2 of FIG. 18, and this angular frequency array is The characteristic value d of the waveform data Wd5 is obtained.

The temperature calculation unit 822 calculates the temperature t of the asphalt mixture by identifying the characteristic value data most similar to the characteristic value d of the waveform data Wd5. Here, as an example, a calculation example using characteristic value data of temperature data of “170° C.”, “160° C.”, and “150° C.” is shown. Graphs G3a to G3c of the characteristic value data when the temperature data is “170° C.”, “160° C.”, and “150° C.” are shown in FIGS. 19(A), (B) and (C). As shown in FIG. 20, the temperature calculation unit 822 uses, for example, similar(), which is a function for obtaining the similarity of sequences, and the characteristic value d of the waveform data WD5 shown in FIG. The angular frequency arrays of the characteristic value data shown in (B) and (C) are respectively compared. As a result, the characteristic value data of “150° C.” can be calculated as the characteristic value data having the angular frequency array having the highest similarity to the characteristic value d. In addition, the accuracy of calculating the temperature t of the asphalt mixture is further enhanced by providing a large number of characteristic value data in advance. Further, by refining the similarity calculation algorithm, it is possible to calculate between temperatures.

As described above, according to the temperature measuring system 8, the temperature measuring method, and the asphalt plant 1 of the asphalt mixture of the present embodiment, a plurality of different angular frequencies obtained by Fourier transforming the waveform data Wd5 as the characteristic value d. An array (angular frequency array) composed of sine wave amplitudes of is used. Further, the temperature calculation unit 822 calculates the similarity between the analysis value d and the analysis value data of the plurality of master data, and sets the temperature data of the characteristic value data having the highest similarity as the temperature of the asphalt mixture. Thus, the angular frequency array representing the characteristics of the wave is obtained from the waveform data of the asphalt mixture, and the temperature t of the asphalt mixture is calculated using this angular frequency array. Therefore, based on the wave characteristics of the asphalt mixture, the asphalt mixture is calculated. The temperature t of the mixture can be accurately calculated.

<<Fifth Embodiment>>
Next, a fifth embodiment to which the asphalt mixture temperature measuring system, the asphalt plant and the asphalt mixture temperature measuring method of the present invention are applied will be described with reference to the drawings. The present embodiment is different from the above-described first embodiment in that the method for calculating the characteristic value d by the temperature calculation unit 822 and the database are not provided, but the other configurations are the same as those in the first embodiment. The same components as those in the first embodiment are designated by the same reference numerals and detailed description thereof will be omitted.

In the present embodiment, the temperature t of the asphalt mixture is obtained from the characteristic value d of the waveform data using a learned model that uses a known machine learning algorithm such as a neural network. Specifically, as shown in FIG. 21, the temperature calculation unit 822 of this embodiment includes a neural network NNW as a learned model. This neural network NNW shows characteristic value data composed of an angular frequency array obtained by Fourier transforming waveform data obtained from an asphalt mixture at an arbitrary temperature, and the temperature of the asphalt mixture associated with the characteristic value data. It is a model that is generated by machine learning using temperature data and as teacher data, and has a characteristic value d indicating a change in wave of the asphalt mixture as an input and a temperature t of the asphalt mixture as an output. FIG. 22 shows a state of the neural network NNW during machine learning. The characteristic value data input as the teacher data is actually measured using an image pickup apparatus having the same frame rate as the image pickup apparatus 81.

The temperature calculation unit 822 of the present embodiment performs a Fourier transform on the waveform data indicating the waves of the asphalt mixture, and calculates the characteristic value d consisting of an angular frequency array. Next, as shown in FIG. 23, when the temperature calculation unit 822 inputs the characteristic value d to the learned neural network NNW, the neural network NNW estimates the temperature t of the asphalt mixture corresponding to the characteristic value d. Output.

As described above, according to this embodiment, the learned neural network NNW generated by machine learning is provided by using the characteristic value data indicating the wave motion of the asphalt mixture and the temperature data of the asphalt mixture as the teacher data. .. When the characteristic value d calculated by analyzing the waveform data is input to the neural network NNW, the temperature t of the asphalt mixture is output. Therefore, unlike the first to fourth embodiments described above, the temperature t can be obtained from the characteristic value d without having a large number of master data. In addition, the temperature t of the asphalt mixture can be accurately calculated by performing the machine learning of the neural network NNW using a large number of teacher data.

Although the analysis value data including the angular frequency array is used as the teacher data of the neural network NNW, the invention is not limited to this. For example, the learned neural network NNW is generated by using any of the characteristic value data used in the first to fourth embodiments as teacher data, and the characteristic value d of the first to fourth embodiments is set in the neural network NNW. The temperature t may be output when input. According to this, the temperature t of the asphalt mixture can be accurately obtained by the learned model that is machine-learned using various characteristic value data indicating the characteristics of the waveform data as teacher data.

In the above embodiment, a digital camera capable of high-speed moving image shooting is used as the imaging device 81, but instead of this, a plurality of (for example, 60 or more) still images can be continuously shot per second. A digital camera may be used. In this case, the image analysis unit 821 calculates the difference between the captured images to generate the waveform data.

Further, the asphalt mixture was imaged by the image pickup device 81 from diagonally above the mixer 7, but as shown in FIG. 24, an opening 71b was formed in the upper side surface of the mixing tank 71, and the asphalt mixture was imaged through this opening 71b. You may. Further, as shown in FIG. 25, a window portion 71d in which a transparent member 71c is fitted may be formed in the opening portion 71b, and the asphalt mixture may be imaged through the window portion 71d. The transparent member 71c is a member that can transmit visible light, and for example, glass, plastic, acrylic, resin, vinyl chloride resin, polystyrene, polycarbonate, polystyrene, hybrid glass, or the like can be used. In this embodiment, the transparent member 71c is made of tempered glass. Further, the transparent member 71c is not limited to this, and may be a material capable of transmitting infrared rays. According to these, it is possible to reduce adverse effects on the image pickup device 81 due to dust generated when the aggregate is put into the mixer 7, water vapor generated when mixing the aggregate and the asphalt, and the like.

Further, the characteristic value d is not limited to the above-mentioned one as long as it shows the characteristic of the waveform data. For example, as the characteristic value d representing the speed of movement of the asphalt mixture, the average value of the peak intervals of the waveform data, the steepness of the waveform data (for example, the rising slope from the bottom to the peak), or the like may be used.

In the temperature measurement system 8 of each of the above-described embodiments, for each of a plurality of picked-up images of the asphalt mixture in the mixer 7, a difference is obtained by comparing the picked-up image taken immediately before, and a variation value obtained by digitizing the difference. Is calculated to generate the waveform data indicating the temporal change of the fluctuation value, and the temperature t of the asphalt mixture is calculated based on the generated waveform data. However, the present invention is not limited to the temperature measurement system 8 as in the above-described embodiments. That is, if the asphalt mixture in the mixer 7 is imaged at a predetermined time interval and the temperature t of the asphalt mixture is calculated based on the changes generated in the plurality of imaged images, the temperature measuring system of the present invention can be used. included.

For example, as a temperature measurement system using another example of the present invention, the asphalt mixture in the mixer 7 is imaged at a predetermined time interval, image analysis processing is performed on a plurality of imaged images, and image analysis is performed. The flow rate of the asphalt mixture may be calculated based on the result of the treatment, and the temperature t of the asphalt mixture may be calculated from this flow rate. More specifically, feature points in the asphalt mixture are extracted from one of the plurality of captured images, and the feature points are tracked in the plurality of captured images captured after this one captured image. Then, as the flow velocity of the asphalt mixture, the moving distance d of the characteristic point per predetermined time is calculated based on the tracking result of the characteristic point, and the temperature t of the asphalt mixture is calculated based on the calculated moving distance d. To do. According to the temperature measurement system using another example of the present invention, since the temperature t is measured based on the flow velocity of the asphalt mixture, that is, the viscosity of the asphalt mixture, the surface temperature of a conventional spot thermometer, thermography, etc. The temperature t of the asphalt mixture can be measured more accurately than a thermometer that measures

Furthermore, the example in which the temperature measuring system 8 is provided in the asphalt plant 1 provided with the mixing line for mixing the new aggregate and the recycled aggregate has been described, but the present invention is not particularly limited to this. For example, the temperature measuring system 8 may be applied to a line containing only new aggregate.

1... Asphalt plant 2... New aggregate supply part 3... Recycled aggregate supply part 4... Stone powder supply part 5... Asphalt supply part 6... Additive supply part 7... Mixer 71... Mixing tank 72... Mixing blade 8... Temperature measurement system 81... Imaging device 82... Calculation device 821... Image analysis part 822... Temperature calculation part 823... Database 823 _{1 to} 823 _n ... Master data 9... Measurement result management device Oa... Observation area di... Difference image Wd1 to Wd5... Waveform data

Claims (9)

In an asphalt plant that mixes asphalt and aggregate with a mixer to produce an asphalt mixture, a temperature measurement system for measuring the temperature of the asphalt mixture in the mixer,
An imaging device for imaging the asphalt mixture in the mixer at predetermined time intervals,
A computing device that computes a temperature t of the asphalt mixture based on a plurality of captured images captured by the imaging device,
The arithmetic unit is
For each of the plurality of captured images, a difference indicating the magnitude of movement of the asphalt mixture is calculated by comparing with the captured image captured immediately before, and a variation value that is a numerical value of the difference is calculated, and the variation value of the variation value is calculated. An image analysis unit that generates waveform data indicating a temporal change,
A temperature calculation unit that analyzes the waveform data, calculates a characteristic value d that is a characteristic value indicating the characteristics of the waveform data, and calculates the temperature t of the asphalt mixture based on the characteristic value d. Cage,
The temperature calculation unit, as the characteristic value d, the maximum value of the fluctuation value within a predetermined time, the average value of a plurality of peaks within a predetermined time, the average value of the amplitude within a predetermined time, or the peak within a predetermined time. The temperature measurement system for an asphalt mixture , wherein the temperature t is calculated such that the temperature t of the asphalt mixture increases as the characteristic value d increases . In an asphalt plant that mixes asphalt and aggregate with a mixer to produce an asphalt mixture, a temperature measurement system for measuring the temperature of the asphalt mixture in the mixer,
An imaging device for imaging the asphalt mixture in the mixer at predetermined time intervals,
A computing device that computes a temperature t of the asphalt mixture based on a plurality of captured images captured by the imaging device,
The arithmetic unit is
For each of the plurality of captured images, a difference indicating the magnitude of movement of the asphalt mixture is calculated by comparing with the captured image captured immediately before, and a variation value that is a numerical value of the difference is calculated, and the variation value of the variation value is calculated. An image analysis unit that generates waveform data indicating a temporal change,
As a characteristic value d, which is a characteristic value indicating the characteristics of the waveform data, a temperature calculation unit that calculates an array of amplitudes of a plurality of sine waves having different angular frequencies obtained by Fourier transforming the waveform data,
Characteristic value data that is the characteristic value indicating the characteristic of the waveform data that has been measured in advance, and temperature data that is the temperature that is measured in advance of the asphalt mixture that is associated with the characteristic value data, and the characteristic As value data, a master data using an array consisting of amplitudes of a plurality of sinusoidal waves having different angular frequencies obtained by Fourier transforming the previously measured waveform data is provided,
The temperature calculation unit calculates a degree of similarity between the characteristic value d and a plurality of characteristic value data included in the master data, and corresponds to the characteristic value data having the highest degree of similarity. A temperature measuring system for an asphalt mixture, wherein the temperature data is the temperature t of the asphalt mixture. In an asphalt plant that mixes asphalt and aggregate with a mixer to produce an asphalt mixture, a temperature measurement system for measuring the temperature of the asphalt mixture in the mixer,
An imaging device for imaging the asphalt mixture in the mixer at predetermined time intervals,
A computing device that computes a temperature t of the asphalt mixture based on a plurality of captured images captured by the imaging device,
The arithmetic unit is
For each of the plurality of captured images, a difference indicating the magnitude of movement of the asphalt mixture is calculated by comparing with the captured image captured immediately before, and a variation value that is a numerical value of the difference is calculated, and the variation value of the variation value is calculated. An image analysis unit that generates waveform data indicating a temporal change,
As the characteristic value d which is a characteristic value indicating the characteristic of the waveform data, the maximum value of the fluctuation value within a predetermined time, the average value of a plurality of peaks within a predetermined time, the average value of the amplitude within a predetermined time, or a predetermined value A number of peaks in time, or a temperature calculation unit for calculating an array of amplitudes of a plurality of sine waves having different angular frequencies obtained by Fourier transforming the waveform data,
Characteristic value data that is the characteristic value that indicates the characteristics of the waveform data that has been measured in advance, and temperature data that is the temperature that has been measured in advance of the asphalt mixture, are generated by machine learning using teacher data, and the characteristic value As data, a trained model that uses the same characteristic value as the characteristic value d is provided,
The temperature calculation unit for calculating a temperature t of the asphalt mixture by inputting the calculated characteristic value d to the learned model, the temperature measurement system of the asphalt mixture. A temperature measuring system for an asphalt mixture according to claim 1 , wherein
The temperature calculation section is provided with a characteristic value data which is the characteristic value indicating characteristics of the waveform data previously measured, and the temperature data is a temperature of advance actually measured the asphalt mixture, the master data containing the ,
The master data uses the same characteristic value as the characteristic value d as the characteristic value data, and the characteristic value data and the temperature data are set such that the larger the characteristic value data, the higher the temperature data. Is associated with,
The temperature calculation unit specifies the characteristic value data corresponding to the calculated characteristic value d from the master data, and based on the temperature data associated with the specified characteristic value data, the asphalt An asphalt mixture temperature measuring system for calculating the temperature t of the mixture. A temperature measuring system for the asphalt mixture according to any one of claims 1 to 4 ,
The image analysis unit sets an observation area in a region that does not include a mixing blade of the mixer in the captured image, and obtains a difference in the observation area. An asphalt plant for producing an asphalt mixture by mixing asphalt and aggregate.
A mixer for mixing the asphalt and the aggregate,
Asphalt plant with a temperature measuring system, the asphalt mixture according to any one of claims 1-5. In an asphalt plant for producing an asphalt mixture by mixing asphalt and aggregate with a mixer, a temperature measuring method for measuring the temperature of the asphalt mixture in the mixer,
A first step of imaging the asphalt mixture in the mixer with an imaging device at predetermined time intervals;
For each of the plurality of picked-up images picked up by the image pickup device, a difference indicating the magnitude of movement of the asphalt mixture is calculated by comparing with the picked-up image picked up immediately before, and a variation value obtained by digitizing the difference is calculated. And a second step of generating waveform data showing the temporal change of the fluctuation value,
A third step of analyzing the waveform data to calculate a characteristic value d, which is a characteristic value indicating the characteristics of the waveform data, and calculating a temperature t of the asphalt mixture based on the characteristic value d. And
In the third step, as the characteristic value d, a maximum value of the fluctuation value within a predetermined time, an average value of a plurality of peaks within a predetermined time, an average value of amplitudes within a predetermined time, or calculates the number of peaks, the so that the temperature t of the characteristic value d larger the the asphalt mixture increases, the temperature measurement method of the asphalt mixture comprising calculating the temperature t. In an asphalt plant for producing an asphalt mixture by mixing asphalt and aggregate with a mixer, a temperature measuring method for measuring the temperature of the asphalt mixture in the mixer,
A first step of imaging the asphalt mixture in the mixer with an imaging device at predetermined time intervals;
For each of the plurality of picked-up images picked up by the image pickup device, a difference indicating the magnitude of movement of the asphalt mixture is calculated by comparing with the picked-up image picked up immediately before, and a variation value obtained by digitizing the difference is calculated. And a second step of generating waveform data showing the temporal change of the fluctuation value,
A third step of calculating an array of amplitudes of a plurality of sine waves having different angular frequencies obtained by Fourier transforming the waveform data as a characteristic value d which is a characteristic value indicating the characteristic of the waveform data,
A fourth step of calculating the temperature t of the asphalt mixture on the basis of the calculated characteristic value d and the master data provided in advance,
The master data is characteristic value data that is the characteristic value indicating the characteristic of the waveform data that has been actually measured in advance, and temperature data that is the previously measured temperature of the asphalt mixture that is associated with the characteristic value data, Including, as the characteristic value data, using an array composed of a plurality of sine wave amplitude of different angular frequency obtained by Fourier transforming the waveform data measured in advance,
In the fourth step, the degree of similarity between the characteristic value d and the plurality of characteristic value data included in the master data is calculated, and the characteristic value data having the highest similarity is calculated. A method for measuring the temperature of an asphalt mixture, which comprises setting the temperature data to the temperature t of the asphalt mixture. In an asphalt plant for producing an asphalt mixture by mixing asphalt and aggregate with a mixer, a temperature measuring method for measuring the temperature of the asphalt mixture in the mixer,
A first step of imaging the asphalt mixture in the mixer with an imaging device at predetermined time intervals;
For each of the plurality of picked-up images picked up by the image pickup device, a difference indicating the magnitude of movement of the asphalt mixture is calculated by comparing with the picked-up image picked up immediately before, and a variation value obtained by digitizing the difference is calculated. And a second step of generating waveform data showing the temporal change of the fluctuation value,
As the characteristic value d which is a characteristic value indicating the characteristic of the waveform data, the maximum value of the fluctuation value within a predetermined time, the average value of a plurality of peaks within a predetermined time, the average value of the amplitude within a predetermined time, or a predetermined value A third step of calculating the number of peaks in time, or an array of amplitudes of a plurality of sine waves having different angular frequencies obtained by Fourier transforming the waveform data;
A fourth step of calculating the temperature t of the asphalt mixture on the basis of the calculated characteristic value d and a learned model provided in advance,
The learned model is a machine learning using characteristic value data, which is the characteristic value indicating the characteristic of the waveform data measured in advance, and temperature data, which is the temperature measured in advance of the asphalt mixture, as teacher data. The characteristic value that is generated and uses the same characteristic value as the characteristic value d as the characteristic value data,
The fourth step is a method for measuring a temperature of an asphalt mixture, including calculating the temperature t of the asphalt mixture by inputting the calculated characteristic value d to the learned model.