KR20220125642A - Temperature control system for plant based on learning using data - Google Patents

Abstract

The present invention relates to a learning-based temperature control system for plants using data. The learning-based temperature control system for plants using data according to one embodiment of the present invention comprises: a growth information providing unit that provides plant growth information; an environment information providing unit that provides indoor environment information; a data receiving unit that receives data from the growth information providing unit and the environment information providing unit; a monitoring unit that stores and monitors the data received from the data receiving unit; a scenario derivation unit that derives an optimal scenario based on the result of the monitoring unit; an appropriate temperature derivation unit that derives an appropriate temperature currently required for the derived optimal scenario; and a temperature control unit that controls the temperature based on the derived appropriate temperature. Accordingly, the present invention enables optimal temperature management depending on the plant variety or growth requirement environment.

Description

Temperature control system for plants based on learning using data {TEMPERATURE CONTROL SYSTEM FOR PLANT BASED ON LEARNING USING DATA}

The present invention relates to a learning-based temperature control system for plants using data.

The temperature control of a house or greenhouse for plants according to the prior art has been managed at a level that determines whether or not the heating and cooling device operates so that the temperature is maintained within a predetermined range by measuring the temperature.

In this case, when there are different varieties of plants in one house, the temperature is controlled in a limited state in which the harvest results of each variety cannot be predicted. In other words, concentrating on one breed may not be the optimal environment for other breeds, but they have been collectively managed.

In addition, even for the same variety, it is managed while ignoring the requirements for different environments depending on the state of growth or health.

SUMMARY OF THE INVENTION An object of an embodiment of the present invention is to provide a system that enables optimal temperature management according to the environment required for plant varieties or growth, which has been proposed to solve the above-described problems.

To achieve the above object, a learning-based temperature control system for plants using data according to an embodiment of the present invention for achieving the above object is a growth information providing unit providing plant growth information, an environmental information providing unit providing indoor environment information, growth information A data receiving unit that receives data from the providing unit and the environmental information providing unit, a monitoring unit that stores and monitors data received from the data receiving unit, a scenario derivation unit that derives an optimal scenario based on the results of the monitoring unit, and the present for the derived optimal scenario It includes an appropriate temperature derivation unit for deriving a required appropriate temperature, and a temperature controller for controlling the temperature based on the derived appropriate temperature.

Preferably, the monitoring unit may compare the correlation between the received growth information and the environment information, and compare the actual growth flow data and the pre-stored recommended growth flow data based on a DTW (Dynamic Time Warping) operation.

Preferably, the scenario derivation unit derives a plurality of scenarios based on the environment required for the current growth level of each of the plurality of varieties and the future growth level of each of the plurality of varieties after a predetermined time, and selects an optimal scenario among the plurality of scenarios. can be derived

Preferably, the scenario derivation unit is based on the environmental information including temperature, humidity, soil pH, soil nitrogen concentration, and light quantity, the current health state of the plant, and changes in the health state and growth state of future plants reflecting the environmental information can be predicted to derive a scenario.

Preferably, it further includes a management server and a house server, the house server includes a growth information providing unit, an environmental information providing unit, a data receiving unit, a monitoring unit, and a temperature control unit, and the management server includes a scenario deriving unit and an appropriate temperature deriving unit. It may further include a second monitoring unit connected to a communication network and a monitoring result receiving unit for receiving the results of the monitoring unit, and a data storage unit for storing the received monitoring result and providing data for the scenario deriving of the scenario deriving unit. .

Specific details of other embodiments are included in the detailed description and drawings.

A learning-based temperature control system for plants using data according to an embodiment of the present invention enables optimal temperature management according to the environment required for plant varieties or growth.

The effects of the present invention are not limited to the above-mentioned effects, and other effects not mentioned will be clearly understood by those skilled in the art from the description of the claims.

1 is a diagram illustrating a learning-based temperature control system for plants using data according to an embodiment of the present invention. 2 is a diagram illustrating a learning-based plant temperature control system using data according to another embodiment of the present invention.

Advantages and features of the present invention, and a method of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in a variety of different forms, and only these embodiments allow the disclosure of the present invention to be complete, and are common in the technical field to which the present invention pertains. It is provided to fully inform those with knowledge of the scope of the invention, and the present invention is only defined by the scope of the claims. The term "part" in the specification includes a unit realized by hardware, a unit realized by software, and a unit or module realized using both.

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

1 is a diagram illustrating a learning-based plant temperature control system 10 using data according to an embodiment of the present invention.

Referring to FIG. 1 , the temperature control system 10 for plants according to an embodiment of the present invention includes a growth information providing unit 112 , an environmental information providing unit 114 , a data receiving unit 132 , a monitoring unit 134 , It includes a scenario deriving unit 138 , an appropriate temperature deriving unit 152 , and a temperature control unit 154 .

The growth information providing unit 112 provides plant growth information to the data receiving unit 132 . The growth information providing unit 112 does not necessarily have a spatial concept, but may include a sensor connected to the vicinity of or in contact with a plant to measure the growth state of the plant. The growth information providing unit 112 may include a wired line or a wireless communication module that receives information from a sensor and provides information to the data receiving unit 132 .

Growth information may include plant length or size information. The length or size information can be measured through color contrast for each pixel using a 3D scanning technology or a camera technology. The color contrast for each pixel means to distinguish the background from the plant by dividing the color for each pixel of the image measured by the camera, and it is possible to distinguish even the leaves, fruits, and stems of plants.

The growth information may include information on the growth rate of a plant. The growth rate can be calculated by measuring the length or size over time.

Growth information may include health status information of plants. Growth information may include plant shape and color data.

The growth information may include information on a distribution degree of a plant, a batch density, and information on a change amount of the batch density.

The environment information providing unit 114 provides indoor environment information to the data receiving unit 132 . The environmental information providing unit 114 does not necessarily have a spatial concept, but may include a sensor connected to an indoor space such as a house, the ground surface, the soil, or a nearby or contacted plant to measure indoor environmental information. have.

The environmental information may include at least one of temperature, humidity, soil pH, nitrogen concentration, light quantity, flow rate of water flowing to the plant, flow rate, indoor wind speed, far infrared rays, anion emission amount, and blowing information.

The data receiving unit 132 receives data from the growth information providing unit 112 and the environmental information providing unit 114 . The data receiving unit 132 does not necessarily have a spatial concept, but may be located in the same indoor space as the growth information providing unit 112 and the environmental information providing unit 114, or may be located in a space where wireless communication is possible. .

The monitoring unit 134 stores and monitors the data received by the data receiving unit 132 .

Preferably, the monitoring unit 134 compares the correlation between the received growth information and the environment information, extracts feature data based on a DTW (Dynamic Time Warping) operation, and recommends growth stored in advance with the actual growth flow data. Flow data can be compared.

The monitoring unit 134 may check the rate of plant growth and health status. It stores past information of temperature, humidity, soil pH, nitrogen concentration, light quantity, flow rate, flow velocity, wind speed, far-infrared rays, and negative ions, stores information on each current environment, and displays past and present growth and health status by time flow. By monitoring, the correlation between each environmental information and growth information is compared.

The monitoring unit 134 may determine whether growth is abnormal by comparing the normal and preferred growth flow data and the actual growth flow data received from the growth information providing unit 112 .

Convert and store the codec of the recommended growth flow data, convert the resolution of the actual growth flow to the resolution of the recommended growth flow, and compare the RGB average value or edge histogram of the converted codec and the converted resolution of the actual growth flow. Based on it, it can be compared with the recommended growth flow.

The recommendation growth flow may classify image and audio data, and extract feature data of the image. Data codec, resolution, bit rate, and frames per second (FPS) can be considered. It is more preferable to convert and store the recommended growth flow data in advance so that the recommended growth flow data and the actual growth flow data are converted to the same reference codec, and match the actual growth flow information therewith. By inquiring/finding the resolution, which is the number of horizontal/vertical pixels of the image, mutually recommended growth flow and actual growth flow information is converted into the same/similar standard resolution. In addition, image data of different file formats and compression methods are unified with the same file format and compression method. For example, the file format of the recommended growth flow data is avi (audio video interleaving), the compression method is DivX MPEG-4 Video v4 (OpenDivX), the resolution, audio sample rate, channels, frames per second, normalizer An operation is performed to match the corresponding actual growth flow data in the same way. Audio related data may not be required.

Feature data can be extracted based on comparison of RGB average values or edge histograms, and changes in positions of shapes in a scene can be extracted as vector values.

The feature data may be color data, structure data of shapes in an image, and shape data based on RGB average value comparison, color luminance information, or central object extraction.

RGB is a method of displaying colors using red, green, and blue as the three primary colors of light, and the average value thereof can be calculated. Each pixel is made of three combinations of red, green, and blue, and each pixel contains 1 byte of red, 1 byte of green, and 1 byte of blue using a byte unit. One byte represents 256 levels of color from 0 to 255. Colors using RGB can be displayed in the same way as 'RGB (red level, green level, blue level)', and the difference in features can be calculated by the average value of each location cell.

Compared to the general histogram, the edge histogram is less sensitive to image changes and shows the spatial distribution of the edge as characteristic information of the growth flow.

In addition, a similarity measurement algorithm may be further used. For example, based on a dynamic time warping (DTW) operation, it is possible to compare with the feature data of each actual growth stream and measure the degree of similarity between the feature data. Using DTW, it is possible to measure the similarity between two sequences even if the rates of the recommended growth flow and the actual growth flow are different. Data that can be converted into a linear sequence can be analyzed with DTW. DTW calculates the best match between the recommended growth flow data and the actual growth flow data, two given sequences. These sequences are non-linearly warped in the time dimension to determine similarity. DTW is an algorithm that measures the similarity of the wavelengths on the time axis, and the growth similarity can be confirmed even if there is acceleration and deceleration in the observation process. The time difference between the recommended growth flow and the actual growth flow is informational and the data can be reflected in the monitoring and scenario process. The characteristic data of the recommended growth flow is a flow expected to appear in actual growth, and the characteristic data of the actual growth flow that is different from the characteristic data of the recommended growth flow is a flow that is not expected to appear in actual growth. The characteristic data of the actual growth flow may be used to predict a bad scenario by comparing it with existing stored data or a plurality of recommended growth flow data. In addition, if the characteristic data of the actual growth flow shows a different pattern from the previously stored data, the scenario prediction may be delayed and the corresponding characteristic section may be prioritized to focus. After showing a different pattern from the previously stored data, if it is a bad result, it can be reflected in the data again. Conversely, if, after showing a pattern different from the previously stored data, an unexpected high-quality fruit is obtained, in the cause analysis, the corresponding characteristic section can be extracted as a cause candidate, reflected in the recommended growth flow data, and recommended It can be applied as a scenario. Based on the results of repeated experiments, improved results that were not previously expected can be obtained, and effective research is possible. By using the characteristic data of the actual growth flow, it can help to derive an optimal scenario, and effective research is possible.

DTW (Dynamic Time Warping) operation is calculated in a many-to-one relationship with feature data, and when the DTW operation value is 0, it means a perfect match. Depending on the degree of similarity can be judged.

A high degree of distinguishing power, for example, feature data in which a feature appears strongly, is assigned a high rank and arranged. It is more preferable to extract the difference information by giving priority to the characteristic data in which the characteristics between the growth flow data can be clearly distinguished.

The actual growth flow data is compared with the pre-stored recommended growth flow data to extract difference information. The degree of difference can be quantified for each section, and it is possible to extract sections with high differences. It is possible not only to compare the corresponding growth information of the extracted difference section, but also to check the environmental information and changes at the time.

It is more preferable that a plurality of recommended growth flow data is stored, and if the difference information from the actual growth flow data to be compared is large or if there is a lot of characteristic data, it can be determined that the degree of abnormality of the corresponding plant increases.

The scenario deriving unit 138 derives an optimal scenario based on the result of the monitoring unit 134 .

The scenario derivation unit 138 may derive a future scenario based on the stored past environment information and growth information at a corresponding time. When the value of any one of temperature, humidity, soil pH, nitrogen concentration, light quantity, flow rate, flow rate, wind speed, far infrared ray, and anion is different from the general flow, the environmental variable through growth information after a predetermined time has elapsed can determine the extent of the spread of In addition, by extracting growth information in the case of abnormal growth, it is possible to extract a past environmental variable that caused the corresponding result and an abnormal range of the corresponding environmental variable that adversely affects the growth. In addition, by storing variable information of an external environment (eg, outside the house), a correlation with the corresponding environmental variable and growth information may be estimated. A number of environmental variables and normal ranges of the corresponding environmental variables and growth scenarios may be derived, and an optimal scenario may be selected from among them. A lot of growth information and environmental information are stored, and the more the scenario derivation process is repeated, the more accurate the results are derived.

The appropriate temperature deriving unit 152 derives an appropriate temperature currently required for the derived optimum scenario.

It is possible to know the normal range of each environment variable for the optimal scenario, and to derive an appropriate temperature currently required. For example, when a lower humidity is more desirable in the growth process of a given variety, the humidity is lowered within a normal range, and in this case, an appropriate temperature range that can bring the best results can be derived. In this process, the degree of ripple of each of many environment variables can be reflected together.

The temperature controller 154 controls the temperature based on the derived appropriate temperature.

The temperature control unit 154 may include a controller (not shown), a carbon fiber heating element (not shown), a blower (not shown), and a wireless communication module (not shown). The wireless communication module may receive the appropriate temperature information received from the appropriate temperature deriving unit 152 , and the controller may control the carbon fiber heating element and the blower.

Preferably, the scenario derivation unit 138 derives a plurality of scenarios based on the environment required for the current growth level of each of the plurality of varieties and the future growth level of each of the plurality of varieties after a predetermined time, and selects an optimal method among the plurality of scenarios. Scenarios can be selected and derived.

The current level of growth includes the level of growth and health at the present time.

The future growth level includes the level of growth and the level of health when a predetermined time has elapsed.

When there are multiple varieties in the same indoor space, the current growth level of each variety is different, the growth rate of each variety is different, the future growth level of each variety is different, and the optimal environment required by each variety is different at each time point. different. In addition, for the best fruit or harvest, there is a good period for efficiency as the growth rate is faster during the entire growth cycle, and sometimes there is a growth section in which a high-quality fruit is obtained only when the growth rate is slow. This is the reason why it is necessary not only to promote growth, but also to control growth.

The scenario derivation unit 138 may derive a plurality of future scenarios based on the stored past environmental information and the growth information of each variety at a corresponding time, and may select an optimal scenario among them.

Preferably, the scenario derivation unit 138 is based on the environmental information including temperature, humidity, soil pH, soil nitrogen concentration, and light quantity, the current health state of the plant, and the change in the health state of the plant in the future reflecting the environmental information And it is possible to derive a scenario by predicting a change in the growth state. The environmental information may further include far-infrared rays and anion emission amounts.

2 is a diagram illustrating a learning-based plant temperature control system 10 using data according to another embodiment of the present invention.

Referring to FIG. 2 , the temperature control system 10 for plants according to another embodiment of the present invention further includes a management server 210 and a house server 250 . The house server 250 includes a growth information providing unit 112 , an environmental information providing unit 114 , a data receiving unit 132 , a monitoring unit 134 , and a temperature control unit 154 . The management server 210 includes a scenario deriving unit 138 , an appropriate temperature deriving unit 152 , a monitoring result receiving unit 142 , and a data storage unit 144 .

Growth information providing unit 112 , environmental information providing unit 114 , data receiving unit 132 , monitoring unit 134 , temperature control unit 154 , scenario deriving unit 138 , and appropriate temperature deriving unit 152 , 1, in this embodiment, only the management server 210, the house server 250, the monitoring result receiving unit 142, and the data storage unit 144 will be described using FIG. 2 .

The management server 210 and the house server 250 do not necessarily have a spatial concept, but in general, the management server 110 will be located in a place convenient for an administrator or a person in charge to manage, and the house server 250 is a plastic house. , it will be located in a place where plants are located, such as greenhouses, or in a convenient place to manage plants in the vicinity. The management server 210 and the house server 250 may be connected by wireless communication.

The monitoring result receiving unit 142 receives the results of the second monitoring unit 136 and the monitoring unit 134 connected to the communication network.

The second monitoring unit 136 may be located not only in a space separated from the management server 210 , but also in a space separated from the monitoring unit 134 . The second monitoring unit 136 may be connected to or included with other second and third house servers (not shown), and provides a result value according to the monitoring process to the monitoring result receiving unit 142 .

The monitoring result receiving unit 142 includes a third monitoring unit (not shown), ... , n-th monitoring unit (not shown), etc. can receive many monitoring results.

The data storage unit 144 stores the received monitoring result, and provides data for the scenario derivation of the scenario derivation unit 138 .

The data storage unit 144 stores more information over time, and allows the scenario derivation unit 138 to make more optimal choices.

The system according to the present invention can not only obtain optimal environmental information according to the environment required for plant varieties or growth, but also can provide various scenarios through the information, and select the optimal scenario from among various variables. There is a remarkable effect that it can be managed with the best growth, and this becomes even more so as time goes by and information from various places is accumulated. In addition, effective research is possible by analyzing and predicting, together with environmental information, whether growth is positively or negatively affected when a different pattern is shown compared to the previously known recommended growth flow.

The system according to the present invention provides an appropriate environment (temperature, humidity, soil pH, nitrogen concentration, light quantity, flow rate, flow velocity, wind speed, far infrared ray, negative ion) derivation unit and environment (temperature, humidity, soil pH, nitrogen concentration, light quantity, flow rate) , flow velocity, wind speed, far infrared rays, negative ions) control system, each further comprising a control system for learning-based plants using data (temperature, humidity, soil pH, nitrogen concentration, light quantity, flow rate, flow velocity, wind speed, far infrared rays, negative ions) control system, (For example, a learning-based humidity control system for plants using data, a learning-based far-infrared emission system for plants using data, etc.) may be included. Plants include flowers, trees, crops, cultivation, seeds, flowers, and medicinal herbs, and the present system can be used as a system for other purposes (eg, a learning-based nitrogen concentration control system for crops using data, etc.). Indoors, including the inside of a plastic house, a (glass) greenhouse, a plastic film greenhouse, an orchard, a garden, and a flower, may include a system for a purpose (eg, a flow control system inside a plastic house, etc.). This system can also be used as a plant growth control system using indoor environmental data. (For example, a crop growth control system using indoor environment data) This system can be used as a storage system for various data that can represent the best plant growth, and can also be used as a learning system. Machine learning of the system according to the present invention can be utilized as a control system for plants using artificial intelligence (AI). (e.g. machine learning systems for plant environments for artificial intelligence)

In the above, preferred embodiments of the present invention have been illustrated and described, but the present invention is not limited to the specific embodiments described above, and in the technical field to which the present invention pertains without departing from the gist of the present invention as claimed in the claims. Various modifications may be made by those of ordinary skill in the art, and these modifications should not be individually understood from the technical spirit or perspective of the present invention.

10 Temperature control system for plants 112 Growth Information Department 114 Environmental Information Service Department 132 data receiver 134 monitoring unit 136 Second monitoring unit 138 Scenario Derivation Unit 152 Proper temperature derivation part 154 temperature control 210 Management Server 250 house servers 142 Monitoring result receiver 144 data storage

Claims (5)

Growth information providing unit for providing information on the growth of plants; Environmental information providing unit for providing indoor environment information; Data receiving unit for receiving data from the growth information providing unit and the environmental information providing unit; Storing the data received from the data receiving unit and monitoring unit for monitoring; Scenario deriving unit deriving an optimal scenario based on the result of the monitoring unit; Appropriate temperature deriving unit deriving an appropriate temperature currently required for the derived optimal scenario; Based on the derived appropriate temperature a temperature controller to control the temperature; Learning-based plant temperature control system using data including The learning of claim 1, wherein the monitoring unit compares the correlation between the received growth information and the environment information, and compares the actual growth flow data with the pre-stored recommended growth flow data based on a DTW (Dynamic Time Warping) operation. Temperature control system for base plants The method of claim 1, wherein the scenario derivation unit derives a plurality of scenarios based on the current growth level of each of the plurality of varieties and the environment required for the future growth level of each of the plurality of varieties after a predetermined time, and selects an optimal one among the plurality of scenarios. A learning-based temperature control system for plants using data derived by selecting a scenario of According to claim 1, wherein the scenario derivation unit, based on the environmental information including temperature, humidity, soil pH, soil nitrogen concentration, and light quantity, the current health state of the plant, and the future health of the plant reflecting the environmental information A learning-based temperature control system for plants using data that derives scenarios by predicting changes in state and growth state The management server of claim 1, further comprising a management server and a house server, wherein the house server comprises the growth information providing unit, the environment information providing unit, the data receiving unit, the monitoring unit, and the temperature control unit. a second monitoring unit including the scenario deriving unit and the appropriate temperature deriving unit, a second monitoring unit connected to a communication network, and a monitoring result receiving unit receiving the result of the monitoring unit, and storing the received monitoring result, and deriving a scenario from the scenario deriving unit Learning-based plant temperature control system using data further comprising a data storage unit that provides data for

Images