KR102656576B1 - System and method for smart farm light source control - Google Patents

Abstract

The present invention relates to a system and method for controlling smart farm light sources. The control method of the smart farm light source control system according to the present invention is to select a combination of light sources corresponding to at least one combination of the output size and whether to output each light source having an individual wavelength range according to the setting information stored by the smart farm manager. setting each star and controlling the light source output according to the corresponding light source combination; determining crop information and growth status for each zone within the smart farm; determining a ranking of excellent light source combinations within the smart farm for the corresponding crops using the determined light source combination for each zone and crop information and growth status for each zone; It is characterized by comprising the step of comparing the excellent light source combinations in each smart farm for the same crop and determining the final excellent light source combination for each crop according to statistical processing.

Description

System and method for smart farm light source control {SYSTEM AND METHOD FOR SMART FARM LIGHT SOURCE CONTROL}

The present invention relates to a system and method for controlling smart farm light sources, and more specifically, to a device and method for controlling a combination of various light sources to improve the growth conditions of crops.

Recently, farms in the form of 'smart farms' have been widely used. Smart farms refer to places where certain crops are grown using smart farm technology.

Here, smart farm technology combines cutting-edge ICT technologies such as the Internet of Things, cloud, big data, mobile and artificial intelligence technologies with the existing agricultural industry to create high added value such as improved productivity, efficiency, and quality throughout the agricultural industry, including production, distribution, and consumption. It means the technology you are pursuing.

In other words, in a smart farm, many devices can be automated using ICT technology, thereby increasing crop production efficiency.

In these smart farms, light sources with specific wavelengths for each crop are deployed to replace the sun.

However, conventionally, light sources (LEDs) with a specific wavelength band for each crop must be purchased and used, so smart farm farmers have no choice to improve crop growth.

Managers or farm owners who operate a large number of smart farms hope to find the optimal light source conditions for the crops they are growing by directly controlling the wavelength and intensity of the light source, but the needs of these smart farm managers have not been met in the past. No solution has been proposed to satisfy it.

Furthermore, when many smart farms are operated for the same crop, there is a request for a method to derive optimal light source conditions for the crop.

Publication Patent No. 10-2022-0122180

Therefore, the present invention was created to respond to the above-described conventional requests, and its purpose is to create a system in which the light source conditions are changed variously according to the selection of the smart farm manager, and the optimal light source conditions for crops are derived according to these changes in the light source conditions. and methods are provided.

In order to achieve the above object, the control method of the smart farm light source control system according to the present invention is a combination of at least one of the output size and whether or not each light source has an individual wavelength range according to the setting information stored by the smart farm manager. setting the corresponding light source combination for each zone and controlling the light source output according to the corresponding light source combination; determining crop information and growth status for each zone within the smart farm; determining a ranking of excellent light source combinations within the smart farm for the corresponding crops using the determined light source combination for each zone and crop information and growth status for each zone; It may include a step of comparing excellent light source combinations within each smart farm for the same crop and determining the final excellent light source combination for each crop according to statistical processing.

Here, the ranking of light source combinations within each smart farm is ranked, the ranking of light source combinations within each smart farm for the same crop is compared to calculate the influence rate of other growth conditions, and the calculated influence rate of other growth conditions is less than or equal to a preset value. After selecting crops, the final best light source combination for the selected crops can be determined.

Here, the combination of light sources may vary depending on the date and time zone.

Here, a step of generating two-dimensional images corresponding to each of the determined light source combinations for each zone and performing machine learning using the generated two-dimensional images as input and the growth status of each crop as an output label may be further included. .

Here, after specifying whether to output and a color corresponding to the output size by wavelength for the determined combination of light sources for each region, a two-dimensional image can be generated by arranging grids colored in the specified color by wavelength order.

In addition, in order to achieve the above object, the smart farm light source control system according to the present invention controls at least one combination of the output size and whether to output each light source having an individual wavelength range according to the setting information stored by the smart farm manager. a light source combination processing unit that sets the corresponding light source combination for each zone and outputs the light source according to the corresponding light source combination; a crop growth determination unit that determines crop information and growth status for each zone within the smart farm; a ranking determination unit that determines the ranking of the best light source combination within the smart farm for the corresponding crop using the light source combination for each zone determined by the light source combination processing unit and the crop information and growth status for each zone determined by the crop growth determination unit; It may include a final excellent light source combination determination unit that compares the rankings of excellent light source combinations within each smart farm for the same crop and determines the final excellent light source combination for each crop according to statistical processing.

Here, the ranking determination unit ranks the rankings of light source combinations within each smart farm, and the final superior light source combination determination unit calculates the influence rate of other growth conditions by comparing the rankings of light source combinations within each smart farm for the same crop. After selecting crops whose growth condition influence rate is less than or equal to a preset value, the final excellent light source combination for the selected crops can be determined.

Here, the light source combination of the light source combination processing unit may vary depending on the date and time zone.

Here, a machine learning execution unit that generates two-dimensional images corresponding to the light source combinations for each zone determined by the light source combination processing unit, and performs machine learning using the generated two-dimensional images as input and the growth status of each crop as an output label. More may be included.

Here, the machine learning performing unit specifies whether to output by wavelength and a color corresponding to the output size for the regional light source combination determined by the light source combination processing unit, and then creates a two-dimensional image by arranging the grid colored in the specified color by wavelength order. can be created.

As described above, according to the present invention, a smart farm manager can try various combinations of light sources to suit the crops grown within his or her smart farm.

In particular, by comparing the growth status of crops grown in each zone for the light source combinations performed within each smart farm, the optimal light source combination for a specific cultivated crop can be derived.

1 is a schematic configuration diagram of the entire system including a smart farm light source control system according to an embodiment of the present invention;
Figure 2 is a functional block diagram of the smart farm light source control system of Figure 1,
Figure 3 is an example of a two-dimensional image corresponding to a combination of light sources,
Figures 4 and 5 are control flow diagrams of a smart farm light source control system according to an embodiment of the present invention.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

Each embodiment according to the present invention below is only an example to aid understanding of the present invention, and the present invention is not limited to these embodiments. In particular, the present invention may be comprised of a combination of at least one of the individual components, individual functions, or individual steps included in each embodiment.

In particular, for convenience, alphabet letters such as ‘(a)’ are included in some claims, but these alphabet letters do not specify the order of each step.

The schematic configuration of the entire system including the smart farm light source control system 100 according to an embodiment of the present invention is as shown in FIG. 1.

In the same drawing, each smart farm 200 automatically performs operations related to crop growth/cultivation according to preset values, as mentioned above. In particular, in this embodiment, a plurality of light sources each having different wavelength bands are used. Assume it is installed.

In this way, whether or not the light source is output and the output size of the plurality of light sources installed in each smart farm 200 are determined according to the setting information (setting value) by the smart farm 200 manager, and according to this, the smart farm 200 manager Depending on your choice, crops can be grown using a combination of various light sources.

At this time, the smart farm 200 can be divided into predetermined zones and set to have a combination of light sources for each zone.

For example, if red light sources, blue light sources, and yellow light sources are installed in each zone in the smart farm 200, the manager sets 50% red light sources, 30% blue light sources, 20% yellow light sources, and 20% yellow light sources in area A of the smart farm 200. Zone B can be set to output 40% red light, 20% blue light, and 30% yellow light, and zone C can be set to output 20% red light, 40% blue light, and 40% yellow light.

According to the settings of the smart farm 200 manager, various light sources with individual wavelengths provided in the smart farm 200 are output in a specific combination (i.e., a combination with different ratios for each light source described above). ) may be achieved by a control device provided within the control device, but may also be achieved by a remote smart farm light source control system 100 as shown in FIG. 1 .

In particular, information about the types or growth conditions of crops grown within the smart farm 200 may also be collected by a device provided within the smart farm 200, but as described above, such information can also be collected from a remote smart farm light source. It may also be collected by control system 100.

In addition, the smart farm light source control system 100 according to an embodiment of the present invention may be configured as a separate server that is distinct from each smart farm 200, as shown in FIG. 1, or may be configured as an external server and each It may be comprised of a total combination of control devices/sensing devices provided in the smart farm 200.

However, in this embodiment, as an example, the smart farm light source control system 100 is operated in a form distinct from each smart farm 200 as shown in FIG. 1, and in this case, the smart farm light source control system 100 It is assumed that the light source output within each smart farm 200 is remotely controlled while communicating with a sensor device or setting device provided within the smart farm 200.

An example of a specific functional block of this smart farm light source control system 100 is shown in FIG. 2.

As shown in the figure, the smart farm light source control system 100 includes a light source combination processing unit 110, a crop growth determination unit 120, a ranking determination unit 130, a final excellent light source combination determination unit 140, and a machine It may be configured to include a learning performance unit 150.

The light source combination processing unit 110 sets a predetermined light source combination for each zone of each smart farm 200 and then performs a function of controlling light source output according to the light source combination.

The light source combination set here is made according to the setting information stored by the smart farm 200 manager, and may be a combination of at least one of the output size and whether or not each light source has an individual wavelength range.

For example, the light source combination may include a combination of the output size and whether to output a red light source, a blue light source, and a yellow light source, as in the above-mentioned example.

That is, the manager of each smart farm 200 can divide his or her smart farm 200 into predetermined zones and enter setting information for the combination of light sources for each divided zone. In this way, each smart farm 200 ) The information input by the manager is collected by the light source combination processing unit 110 and can perform the functions described above.

Here, communication between each smart farm 200 and the smart farm light source control system 100 can be accomplished by wired or wireless methods, and the technology itself for communicating between remote devices using IoT technology, etc. is known. A more detailed description is omitted.

The light source combination of the light source combination processing unit 110 may not be a fixed value but may vary depending on the date (season) and time zone.

The crop growth determination unit 120 performs the function of determining crop information and growth status for each zone within the smart farm 200.

Here, crop information for each zone within the smart farm 200 may have been previously input by the manager of the smart farm 200.

In addition, crop information for each zone may also be input by the manager of the smart farm 200, and further, the crop growth determination unit 120 may analyze images captured by imaging equipment such as a camera provided in the smart farm 200. ) may have made his own judgment.

For example, the crop growth determination unit 120 can determine the growth status of crops in each zone by using a preset image analysis algorithm to determine the size of the cultivated crop, the state of the leaves, etc. from the captured image.

The ranking determination unit 130 uses the light source combination for each zone determined by the light source combination processing unit 110 and the crop information and growth status for each zone determined by the crop growth determination unit 120 to determine the smart farm 200 for the corresponding crop. It performs the function of determining the ranking of the best light source combinations within the system.

For example, when the smart farm 200 manager cultivates the same crop and sets different light source combinations for each zone, the ranking determination unit 130 determines the light source combination for each zone and crop growth determined by the light source combination processing unit 110. Using the crop information and growth status for each zone determined by the determination unit 120, the light source combination ranking for the corresponding crop can be ranked.

For example, as described above, in area A of the smart farm 200, red light source 50%, blue light source 30%, yellow light source 20%, area B 40% red light source, blue light source 20%, yellow light source 30%, C After setting the zones to output at a ratio of 20% red light source, 40% blue light source, and 40% yellow light source, if the crop growth condition is best in zone B and worst in zone C, the ranking determination unit 130 For the relevant crops (red light source 40%, blue light source 20%, yellow light source 30%), (red light source 50%, blue light source 30%, yellow light source 20%), (red light source 20%, blue light source 40%, yellow light source Ranking can be given in order of light source (40%).

The final excellent light source combination determination unit 140 compares the rankings of excellent light source combinations within each smart farm 200 for the same crop and determines the final excellent light source combination for each crop according to statistical processing.

For example, the light source combination ranking may be different for each smart farm 200, and the final excellent light source combination determination unit 140 statistically processes the light source combination ranking determined within the plurality of smart farms 200 to determine the best growth condition. It is possible to determine the optimal range of light source combinations.

That is, since the range of light source combinations may be different within different smart farms 200, statistical processing is performed on them, for example, after calculating the average value of light source combinations with high rankings of excellent light sources, the calculated average value can be judged as the final excellent light source combination.

In particular, the final excellent light source combination determination unit 140 calculates the influence rate of other growth conditions by comparing the rankings of the light source combinations in each smart farm 200 for the same crop, and the calculated influence rate of other growth conditions is less than or equal to a preset value. After selecting suitable crops, the final excellent light source combination for the selected crops can be determined.

For example, as a result of applying the first light source combination, the second light source combination, the third light source combination, and the fourth light source combination in the first smart farm 200 and the second smart farm 200 to a specific crop, the first light source combination In the smart farm 200, the order of high growth status of the cultivated crops is the first light source, the fourth light source combination, the third light source, and the second light source, while in the second smart farm 200, the order of high growth status of the cultivated crops is When the second light source, third light source, fourth light source combination, and first light source are in reverse order, the final excellent light source combination determination unit 140 determines other growth conditions (e.g., types and amounts of nutrients supplied, cultivation Temperature, humidity, carbon dioxide concentration, etc.) can be ruled out by determining that the influence of the light source on the crop is greater than the light source.

As another example, as a result of applying the first light source combination, the second light source combination, the third light source combination, and the fourth light source combination in the first smart farm 200 and the second smart farm 200 to specific cultivated crops, 1 In the smart farm 200, the order of high growth status of the cultivated crops is the first light source, the fourth light source combination, the third light source, and the second light source, whereas in the second smart farm 200, the order of high growth status of the cultivated crops is the first light source, the fourth light source combination, the third light source, and the second light source. In the case of a combination of the first light source, third light source, fourth light source, and second light source, although the order of the third light source and fourth light source is reversed, the first light source is the one in which the growth state of the cultivated crop is the highest. Since the lowest state of the second light source is the same for both the first smart farm 200 and the second smart farm 200, in this case, the final excellent light source combination determination unit 140 sets the influence rate of other growth conditions to a preset value. By determining that the following is achieved, the final excellent light source combination (i.e., the first light source combination) can be determined for the crop.

Meanwhile, the machine learning unit 150 generates two-dimensional images corresponding to each zone-specific light source combination determined by the light source combination processing unit 110, uses the generated two-dimensional images as input, and uses the growth status of each crop as an output label. Machine learning can be performed.

That is, in the case of supervised learning among machine learning, input and labeling values (output labels) are required. In this embodiment, two-dimensional images corresponding to each light source combination for each zone of each smart farm 200 are generated, and then generated. Machine learning is performed using a two-dimensional image as input and the growth status of each crop as an output label.

For example, the machine learning unit 150 specifies whether to output by wavelength and a color corresponding to the output size for the combination of light sources for each region determined by the light source combination processing unit 110, and then determines the color corresponding to the output size by wavelength order of the grid colored with the specified color. Two-dimensional images can be created by arrangement.

A specific example of this is shown in Figure 3.

For example, if there are a first light source, a second light source, a third light source, a fourth light source, and a fifth light source in the order of the wavelength band, a specific light source considering color, saturation, and brightness is determined according to the output status and output size of each light source. A color can be specified, and grids colored with the designated color can be arranged horizontally as shown in FIGS. 3(a) and 3(b).

For reference, the second light source in FIG. 3(a) and the fourth light source in FIG. 3(b) are expressed in black, which means that the corresponding light source is not output.

Recently, machine learning using images has been widely used in the field of artificial intelligence. For example, after building a machine learning model using CNN (Convolutional Neural Network), the two-dimensional image generated as described above, that is, the grid array image, is used as input. Machine learning can be performed.

Depending on the machine learning performance, the parameter values of each layer included in the CNN may be determined. Since the machine learning process itself corresponds to a known technology, a more detailed description will be omitted.

Hereinafter, the control flow of the smart farm light source control system 100 according to an embodiment of the present invention will be described with reference to FIGS. 4 and 5.

First, the smart farm light source control system 100 can set a combination of light sources based on information set by the manager of each smart farm 200 (step S1), for example, whether the light source for each wavelength is turned on and the output size. etc. can be set for each zone of each smart farm 200.

Subsequently, the smart farm light source control system 100 can ensure that each smart farm 200 outputs light sources according to the set light source combination for each zone (step S3).

At this time, the light source output can be output for a preset time every day, and in particular, the output time and output maintenance time can be changed depending on the season.

At this time, the smart farm light source control system 100 determines the crop growth status for each zone of each smart farm 200 (step S5) and determines the ranking of the best light source combination within each smart farm 200 (step S7).

This process may be processed at the time crop cultivation is completed, or may be processed at a preset period, for example, weekly or monthly.

Next, the smart farm light source control system 100 compares the rankings of the best light source combinations in each smart farm 200 and determines the final best light source combination for each crop (step S9).

The judgment of the final light source combination for each crop may be classified by schedule or season.

Meanwhile, in determining the final excellent light source combination for each crop, the process of FIG. 4 described above may be followed, but an artificial intelligence module constructed through machine learning may also be used.

To achieve this, machine learning must be continuously performed, and Figure 5 shows the process in which machine learning is performed in this way.

First, as shown in FIG. 4, it is assumed that the first mode is a mode in which the ranking of excellent light source combinations for each smart farm 200 is determined, and then the final excellent light source combination is determined by comparing them.

With the first mode set in this way (step S11), the smart farm light source control system 100 generates a two-dimensional image for each light source combination (step S13), uses the generated two-dimensional image as an input, and sets the corresponding light source combination. Machine learning is performed by labeling the growth status of each crop (step S15).

Here, the growth status may be a percentage of 100 or a specific score determined after standardization.

This machine learning process may be continuously performed with the first mode set until the number of machine learning executions reaches a preset number (step S17).

This is because a sufficient amount of machine learning is required to build a proper artificial intelligence module.

If it is determined that sufficient machine learning has been performed (step S17), the smart farm light source control system 100 is set to the second mode (step S17).

In the second mode, an artificial intelligence module built by machine learning can be used to determine the final best light source combination for a specific crop.

Specifically, when a light source combination for a specific crop is received from an external source, for example, each smart farm 200, the artificial intelligence module in the second mode determines the growth state in response to the input light source combination for the specific crop. A corresponding specific score may be output.

That is, in the second mode, the growth status score of a specific crop can be calculated in response to various types of light source combinations, so the final excellent light source combination for the specific crop can be determined by comparing the calculated scores.

Of course, the machine learning process in the first mode of operation can be continuously performed and provide a final combination of excellent officials for a specific crop upon request from an external party, for example, a smart farm manager.

Meanwhile, of course, the process of performing each of the above-described embodiments can be performed by a program or application stored in a predetermined recording medium (eg, computer-readable). Here, recording media include electronic recording media such as RAM (Random Access Memory), magnetic recording media such as hard disks, and optical recording media such as CDs (Compact Disk). At this time, the program stored in the recording medium can be executed on hardware such as a computer or smartphone to perform each of the above-described embodiments. In particular, at least one of the functional blocks of the smart farm light source control system according to the present invention described above may be implemented by such a program or application.

In addition, the present invention is not limited to the specific embodiments described above, but can be implemented with various changes and modifications without departing from the gist of the present invention. It will be apparent that such changes and modifications are included in the present invention if they fall within the scope of the appended claims.

100: Smart farm light source control system 200: Smart farm
110: Light source combination processing unit 120: Crop growth determination unit
130: Ranking judgment unit 140: Final excellent light source combination judgment unit
150: Machine learning execution unit

Claims (10)

(a) According to the setting information saved by the smart farm manager, the light source combination corresponding to at least one combination of the output size and the output status of each light source having an individual wavelength range is set for each zone to determine the light source output according to the light source combination. controlling;
(b) determining crop information and growth status for each zone within the smart farm;
(c) Using the zone-specific light source combination determined in step (a) and the zone-specific crop information and growth status determined in step (b), the ranking of the best light source combination within the smart farm for the corresponding crop is determined. Steps and;
(d) comparing the excellent light source combinations within each smart farm for the same crop and determining the final excellent light source combination for each crop according to statistical processing;
(e) Generating two-dimensional images corresponding to the light source combinations for each zone determined in step (a) above, and performing machine learning using the generated two-dimensional images as input and the growth status of each crop as an output label.
Including,
In step (e), the color corresponding to the output size and whether to output by wavelength is specified for the combination of light sources for each region determined in step (a), and then a two-dimensional image is created by arranging the grid colored in the specified color by wavelength order. A control method of a smart farm light source control system characterized by generating. According to paragraph 1,
In step (c), the light source combinations within each smart farm are ranked,
In step (d), the ranking of light source combinations within each smart farm for the same crop is compared to calculate the influence rate of other growth conditions, and the crops for which the calculated influence rate of other growth conditions is less than or equal to a preset value are selected. A control method of a smart farm light source control system characterized by determining the final excellent light source combination for crops. According to paragraph 1,
A control method of a smart farm light source control system, characterized in that the light source combination in step (a) varies depending on the date and time zone. delete delete According to the setting information stored by the smart farm manager, a light source combination corresponding to at least one combination of the output and output size of each light source having an individual wavelength range is set for each zone, and the light source output according to the light source combination is output to the light source combination processing unit. and;
a crop growth determination unit that determines crop information and growth status for each zone within the smart farm;
a ranking determination unit that determines the ranking of the best light source combination within the smart farm for the corresponding crop using the light source combination for each zone determined by the light source combination processing unit and the crop information and growth status for each zone determined by the crop growth determination unit;
a final excellent light source combination determination unit that compares the rankings of excellent light source combinations within each smart farm for the same crop and determines the final excellent light source combination for each crop according to statistical processing;
A machine learning execution unit that generates two-dimensional images corresponding to the light source combinations for each zone determined by the light source combination processing unit, and performs machine learning using the generated two-dimensional images as input and the growth status of each crop as an output label.
Including,
The machine learning unit specifies whether to output by wavelength and a color corresponding to the output size for the regional light source combination determined by the light source combination processing unit, and then generates a two-dimensional image by arranging the grid colored in the specified color by wavelength order. A smart farm light source control system characterized by: According to clause 6,
The ranking determination unit ranks the ranking of light source combinations within each smart farm,
The final excellent light source combination determination unit calculates the influence rate of other growth conditions by comparing the rankings of light source combinations in each smart farm for the same crop, selects crops for which the calculated influence rate of other growth conditions is less than or equal to a preset value, and selects A smart farm light source control system characterized by determining the final excellent light source combination for the selected crops. According to clause 6,
A smart farm light source control system, characterized in that the light source combination of the light source combination processing unit varies depending on the date and time zone. delete delete