KR20230060815A - Edge computing device for smart farm and smart farm including same - Google Patents

Abstract

The present invention provides an edge computing device for a smart farm that preemptively controls the cultivation environment by predicting changes in the cultivation environment, and a smart farm including the same. The present invention relates to a data receiving unit for receiving environmental information from an environmental sensor that senses the cultivation environment of a cultivation unit equipped with facilities and equipment for controlling the cultivation environment including temperature and humidity, in which crops are grown, and generates environment information, and time-series data. and a controller controlling the facility equipment by analyzing the received environmental information with an analysis model and predicting a future cultivation environment.

Description

Edge computing device for smart farm and smart farm including same {Edge computing device for smart farm and smart farm including same}

The present invention relates to an edge computing device for a smart farm and a smart farm including the same, and relates to an edge computing device for a smart farm that preemptively controls a cultivation environment by predicting changes in the cultivation environment, and a smart farm including the same.

Here, background art related to the present invention is provided, and they do not necessarily mean prior art (This section provides background information related to the present disclosure which is not necessarily prior art).

Until one generation ago, agriculture that cultivates crops has been classified as one of the representative labor-intensive industries. This is because even if many tasks are mechanized, it is not possible to mechanize all tasks 100%, let alone operate the machine.

In addition, since agricultural technology is highly influenced by climate, it is difficult to standardize and mechanize or automate.

However, this agriculture is an industry that secures food, one of the most important factors for living as 'food' for all members of society, and is an industry in which self-sufficiency is desperately needed while being protected, especially in Korea.

On the other hand, with the recent development of industrial technology, various smart farm systems have been introduced and are being used in various forms.

The most basic purpose of a smart farm is to mass-produce high-quality crops with a small workforce.

By using various automated equipment, the physical time of manual work required for crop cultivation can be greatly reduced. Therefore, by increasing the area where one person can grow crops, the number of workers per area could be greatly reduced.

Automation equipment is highly effective for mass production, and the influence of climate and seasons can be minimized through the development of various types of house cultivation methods.

However, there is a problem that it is difficult to easily achieve high quality of crops only with the above-mentioned factors.

In order to cultivate high-quality crops, it is essential to take various appropriate measures according to the shape and size of crops during the growth process.

So far, smart farm systems have been managed with the outside, such as houses, and automatically detect and collect the cultivation environment of the cultivation area where crops grow, such as temperature and humidity.

However, it is not possible to preemptively control the cultivation environment by controlling the environment such as temperature and humidity at the time when the cultivation environment is changed.

The present invention provides an edge computing device for a smart farm that preemptively controls the cultivation environment by predicting changes in the cultivation environment, and a smart farm including the same. However, the scope of the present invention is not limited thereby.

One aspect of the present invention is a data receiver for receiving environmental information from an environmental sensor that senses the cultivation environment of a cultivation unit in which crops are grown and is equipped with facility equipment for controlling the cultivation environment including temperature and humidity to generate environmental information. and a controller for controlling the facility equipment by analyzing the received environmental information using a time-series data analysis model and predicting a future cultivation environment.

The environment information may include a measurement time and measurement data of a cultivation environment.

The control unit acquires environment information up to the first point in time, estimates a parameter of a time series data analysis model using the acquired environment information, and uses the time series data analysis model using the estimated parameter to determine the first point in time. The mean and standard deviation of subsequent environmental information is predicted, and the average and The standard deviation is predicted, a confidence interval (±β) of the measurement data at the second time point is calculated using the predicted average and standard deviation, and the average value of the measurement data at the second time point in the future is calculated under the preset reliability, , It is possible to determine whether the average value of the measured data of the predicted second time point exists within a predetermined range, and control the facility equipment according to the determination result.

The first point of view may be a current point of view, and the second point of view may be a predetermined future point of time from the first point of time.

On the other hand, another aspect of the present invention is a cultivation unit including a space where crops are grown and isolated from the external environment, facility equipment for adjusting the cultivation environment of the cultivation unit, and environmental information by sensing the cultivation environment. It provides a smart farm, including an environmental sensor that generates and the above-mentioned edge computing device for a smart farm.

Other aspects, features, and advantages other than those described above will become clear from the detailed description, claims, and drawings for carrying out the invention below.

An edge computing device for a smart farm and a smart farm including the device according to an embodiment of the present invention can be preemptively controlled to grow crops in an appropriate environment.

In addition, the edge computing device for smart farm and the smart farm including the device according to an embodiment of the present invention can use the predicted result to save energy required for cultivation.

Of course, the scope of the present invention is not limited by these effects.

1 is a block diagram schematically illustrating an edge computing device for a smart farm and a smart farm including the same according to an embodiment of the present invention.
2 is a flowchart schematically illustrating an edge computing device for a smart farm and a smart farm including the same according to an embodiment of the present invention.

Hereinafter, various embodiments of the present disclosure will be described in conjunction with the accompanying drawings. Various embodiments of the present disclosure may have various changes and various embodiments, and specific embodiments are illustrated in the drawings and related detailed descriptions are described. However, this is not intended to limit the various embodiments of the present disclosure to specific embodiments, and should be understood to include all changes and/or equivalents or substitutes included in the spirit and technical scope of the various embodiments of the present disclosure. In connection with the description of the drawings, like reference numerals have been used for like elements.

Expressions such as “include” or “may include” that may be used in various embodiments of the present disclosure indicate the presence of disclosed functions, operations, or components, and may include one or more additional functions, operations, or components. components, etc. are not limited. In addition, in various embodiments of the present disclosure, terms such as "include" or "have" are intended to designate that there are features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, It should be understood that it does not preclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

In various embodiments of this disclosure, expressions such as “or” include any and all combinations of the words listed together. For example, "A or B" may include A, may include B, or may include both A and B.

Expressions such as "first", "second", "first", or "second" used in various embodiments of the present disclosure may modify various components of various embodiments, but do not limit the components. don't For example, the above expressions do not limit the order and/or importance of corresponding components. The above expressions may be used to distinguish one component from another. For example, the first user device and the second user device are both user devices and represent different user devices. For example, a first element may be termed a second element, and similarly, a second element may also be termed a first element, without departing from the scope of rights of various embodiments of the present disclosure.

When an element is referred to as being "connected" or "connected" to another element, the element may be directly connected or connected to the other element, but with the other element. It should be understood that other new components may exist between the other components. On the other hand, when an element is referred to as being “directly connected” or “directly connected” to another element, it will be understood that no new element exists between the element and the other element. should be able to

Terms used in various embodiments of the present disclosure are only used to describe a specific embodiment, and are not intended to limit various embodiments of the present disclosure. Singular expressions include plural expressions unless the context clearly dictates otherwise.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which various embodiments of the present disclosure belong.

Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in various embodiments of the present disclosure, ideal or excessively formal. not interpreted as meaning

Hereinafter, a system may be at least one of configurations of devices, a computer program that executes a method of operating the devices, and a medium in which the computer program is recorded, some of them, or all of them.

1 is a block diagram schematically showing an edge computing device for a smart farm and a smart farm 100 including the device according to an embodiment of the present invention.

The smart farm 100 includes a cultivation unit 110, facility equipment 120, and an environmental sensor 130. The cultivation area is a space isolated from the external environment, and crops are grown. For example, a farming department may be a house.

The facility equipment 120 may control a cultivation environment including temperature, humidity, concentration of carbon dioxide, etc. in which crops are grown. For example, the facility equipment 120 may include a boiler, an air conditioner, a dehumidifier, a humidifier, a ventilator, an air conditioner, and the like.

The environmental sensor 130 may generate environmental information by sensing the cultivation environment. For example, the environment sensor 130 may include sensors that measure the temperature, humidity, and concentration of carbon dioxide in the cultivation area. In this case, the environment information may include measurement data of measurement time and cultivation environment. For example, the environmental information may include data such as 13:00:01 on August 1, 2021, 27°C.

Additionally, the smart farm control unit 140 operates the facility equipment 120 based on the control information. For example, the control unit of the smart farm 100 may control the facility equipment 120 according to the predicted information when the edge computing device 200 for smart farm transmits predicted information, as will be described later. Alternatively, the smart farm controller 140 may control the facility equipment 120 when the edge computing device 200 for smart farm transmits control information according to the prediction information.

The edge computing device 200 for smart farms can control the facility equipment 120 by predicting the environmental information of the farming unit. In more detail, the edge computing device 200 for a svart farm may analyze environmental information using a time-series data analysis model, predict a future cultivation environment of a cultivating unit, and control the facility equipment 120 according to the prediction result.

Due to this, the edge computing device 200 for smart farms can be controlled preemptively to grow crops in an appropriate environment.

In addition, the edge computing device 200 for smart farms may use the prediction result to save energy required for cultivation.

For example, when the edge computing device 200 for smart farm predicts that the temperature will increase beyond an appropriate growth temperature range, the ventilation device may be operated. Due to this, it is possible to save energy and maintain an appropriate growth temperature. At this time, the edge computing device 200 for a smart farm may directly or indirectly control the facility equipment 120 .

The edge computing device 200 for a smart farm includes a data collection unit 210 that receives the environmental information from the environmental sensor 130, analyzes the received environmental information with a time-series data analysis model, and predicts a future cultivation environment. A controller 220 controlling the facility equipment 120 may be included.

Hereinafter, prediction using a time series data analysis model will be described in detail.

Referring to FIG. 2 , in order to predict environment information, the control unit 220 obtains environment information up to a specific point in time (eg, the current point in time) using environment information up to the past or the present as a modeling target (S10).

The control unit 220 estimates the parameters of the time series data analysis model using measurement data information for each measurement time included in the collected environment information (S20). Regarding time series analysis, univariate time series analysis is a technique of explaining the current state and predicting the direction of future change using only past information of the time series without using a specific explanatory variable. That is, the current state of time-series variables is seen as a result of internal trends and external shocks of time-series variables.

According to an embodiment of the present invention, the time series data analysis model used for predicting environmental information includes an autoregressive (AR) model (autoregressive model), a moving average (MA) model (moving average model), and an autoregressive moving average (ARMA) model. model and an autoregressive integrated moving average (ARIMA) model. In particular, according to embodiments of the present invention, it may be desirable to use the ARIMA model.

The ARIMA model is a differencing model by combining an autoregressive (AR) model and a moving average (MA) model. Autoregressive models reflect changes that arise from the effects of past conditions on the present, while moving average models reflect changes that arise only from the effects of pure shocks. Since the combination of the two models can be explained only when the shock factor is stable, the difference is used as a method of stabilizing unstable time-series variables.

The ARIMA model used for environmental information prediction is expressed as ARIMA(p,d,q), and the time series of which the order of the autoregressive process is p and the order of the moving average is q for the d-order differential time series of the original time series Y(t). indicates that it is a process. According to an embodiment of the present invention, the control unit 220 may use the ARIMA (p, 1, q) model. In this specification, the term parameter may be a concept including the parameters p, d, and q of the ARIMA model and/or coefficients used in equations of the ARIMA model determined by the parameters. Regarding the detailed operation of the ARIMA model, the control unit 220 estimates the optimal parameters while changing the values of the parameters (p, d, q) of the ARIMA model.

Then, by using the ARIMA model using the estimated parameters and parameters, the average and standard deviation of the measurement data of the second time point after a certain period from the first time point in which the hourly measurement data data exists are predicted (S30).

Then, a confidence interval (±β) of the measured data at the second time point is calculated using the predicted mean and standard deviation (S40). The confidence interval is obtained by adding and subtracting an appropriate confidence range (β) value from the predicted value.

Next, the control unit 220 calculates an average value of measurement data at a second time point in the future under a preset reliability (S50). Reliability refers to the probability that future measurement data is outside the range of predicted measurement data. When the reliability is set, the control unit 220 assumes a normal distribution and calculates expected mean measurement data. If the reliability is high, the possibility of realizing measurement data that deviate from the expectation is reduced, but the prediction interval becomes too large, resulting in poor prediction precision. When the reliability is low, precision prediction with a small range is possible, but the possibility of out of the prediction range increases. According to an embodiment of the present invention, the reliability may preferably have a value between 70% and 90%. More preferably, 80% is appropriate.

After calculating the predicted value of the measured data, the controller 220 determines whether the predicted average value of the measured data at the second point in time is within a predetermined range (S60).

And the control unit 220 controls the facility equipment 120 according to the determination result (S70). For example, in the case of temperature, the temperature range required for growth is 15 to 18 ° C, but the measured data of the predicted second time point The average value may be 20°C, and the current temperature may be 18°C.

Although the current temperature is within an appropriate temperature range, the controller 220 predicts that the temperature will rise in the future, and thus controls the facility equipment 120 by generating and transmitting control information for operating the air conditioner.

At this time, since the controller 220 can maintain the temperature by operating the air conditioner weakly, energy can be saved and the temperature can be effectively maintained.

Here, the first viewpoint is a current viewpoint, and the second viewpoint may be separated by a predetermined time from the first viewpoint. For example, the second point of view may be a point of time 30 minutes or one hour in the future from the first point of time, which is the present point of time.

The edge computing device 200 may repeat the prediction to predict future temperature trends and control accordingly.

On the other hand, to describe the operating method of the ARIMA (Autoregressive Integrated Moving Average) model in detail, the controller 220 sets the ARIMA (p, 1, q) model. Basically, the control unit 220 sets the difference value d between predicted values to 1 in order to explain non-stationary time series. However, it is not necessarily set to only 1. It can have a value of 0 or 2 to 5. Then, the value of the first parameter (p) and the value of the third parameter (q) are set. As described above, the first parameter (p) is a parameter meaning the lag of the AR model, and the third parameter (q) means the lag of the MA model.

The control unit 220 estimates the parameter value and the parameter value while changing the p value and the q value from 0 to 5. According to an embodiment of the present invention, the ARIMA (p,1,q) model converts the measured data value at time (t+1) to the previous measured data value Y(t) of p and q previous prediction error values e(t) It is expressed as a linear combination of , and the predicted value is calculated based on the expressed equation.

For example, an ARIMA model where p is 1 and q is 0 is Y(t) = Y(t-1) + A * (Y(t-1)-Y(t-2)) + e(t) , Here, the control unit 220 estimates the coefficient A for the difference value (Y(t-1)-Y(t-2)) of the previous measurement data. Here, e(t) may represent white noise.

For example, an ARIMA model where p is 1 and q is 1 is Y(t) = Y(t-1) + A * (Y(t-1)-Y(t-2)) + e(t) + B * It is expressed by the equation of e(t-1), where the control unit 220 moves as well as the coefficient A for the difference value (Y(t-1)-Y(t-2)) of the previous measurement data. Estimate the average coefficient B. e(t) is the error term at time t.

The control unit 220 estimates the optimal parameter in each case while changing the p and q values from 0 to 5, respectively

And, finally, the p, q values and parameters with the smallest prediction error are determined, and the measurement data at the future point is predicted using them.

On the other hand, according to another embodiment of the present invention, as in the above method, after randomly changing the p and q values, estimating the optimal parameters for each case and comparing all, the prediction error is calculated to indicate the smallest prediction error. Instead of estimating the parameters and/or parameters as optimal parameters, the ARIMA model is pre-determined by first estimating the parameters, and the parameters related to the coefficients of the equation of the ARIMA model, for example, A and/or B, are subsequently estimated. method can be used.

At this time, the controller 220 estimates the values of the parameters p and q as optimal values using an auto correlation function (ACF) and a partial auto correlation function (PACF). That is, the suitability of the ARIMA model is determined using the above two functions. When the time series data has AR characteristics, the ACF decreases slowly and the PACF decreases rapidly except for the first lag. Conversely, in the case of MA, the ACF decreases rapidly and the PACF decreases slowly. It is desirable to use rapidly decreasing lags as the parameters (p, q) of each AR and MA model.

Alternatively, various ARIMA models are created by varying the p and q values from 0 to 5, and if the model is not suitable, the model is modified according to the shape of the ACF and PACF. It is also possible to adopt the most appropriate model by repeating this process until the ACF and PACF of the modified model show the appearance of white noise.

According to the above method, the ARIMA model (p, 1, q) is determined, and the parameter values of A and/or B are estimated using the determined ARIMA model. When the parameter estimation is completed, the control unit 220 determines the suitability of the parameters of the ARIMA model by using previously secured actual data, that is, measurement data for a certain period of time as test data. That is, a prediction error between the predicted value and the actual measurement data is calculated, and a model having a parameter representing the smallest error is adopted as a final model.

Due to this, the edge computing device 200 for smart farms can be controlled preemptively to grow crops in an appropriate environment.

In addition, the edge computing device 200 for smart farms may use the prediction result to save energy required for cultivation.

In this way, the present invention has been described with reference to the embodiments shown in the drawings, but this is only exemplary, and those skilled in the art will understand that various modifications and equivalent other embodiments are possible therefrom. . Therefore, the true technical scope of protection of the present invention should be determined by the technical spirit of the appended claims.

Claims (5)

a data collection unit that receives environmental information from an environment sensor that senses the cultivation environment of the cultivation unit where crops are grown and is equipped with equipment for controlling the cultivation environment including temperature and humidity;
a control unit controlling the facility equipment by analyzing the received environmental information using a time-series data analysis model and predicting a future cultivation environment;
Including, edge computing device for smart farm. According to claim 1,
The environmental information includes a measurement time and measurement data of a cultivation environment, an edge computing device for a smart farm. According to claim 1,
The control unit,
Obtaining environmental information up to the first point in time;
Estimating the parameters of the time series data analysis model using the obtained environmental information,
Predicting the average and standard deviation of the environmental information after the first time point using a time series data analysis model using the estimated parameters,
Predicting the average and standard deviation of measurement data at a second time point after a certain period from a first time point in which hourly measurement data data exists using an ARIMA model using estimated parameters and parameters,
Calculate a confidence interval (±β) of the measured data at the second time point using the predicted mean and standard deviation,
Calculate an average value of measurement data at a second time point in the future under a preset reliability;
It is determined whether the average value of the measured data of the predicted second time point is within a predetermined range;
An edge computing device for smart farms that controls facility equipment according to the judgment result. According to claim 3,
The first point in time is a current point in time, and the second point in time is a predetermined future point in time from the first point in time, an edge computing device for a smart farm, A cultivation unit including a space in which crops are grown and isolated from the external environment;
Facility equipment for controlling the cultivation environment of the cultivating unit;
an environment sensor for sensing the cultivation environment and generating environment information; and
Claims 1 to 4 of any one of the edge computing device for a smart farm;
Including, smart farm.

Images