Self-adaptive irrigation control method and system based on soil permeability

Abstract

The invention discloses a self-adaptive irrigation control method and system based on soil permeability, a model is established based on linear regression and mean square error evaluation indexes to fit the current soil permeability, and then according to current environmental data such as soil depth, soil humidity and the like, an irrigation duration prediction model is used to predict the soil permeability. And the irrigation duration in the current environment is predicted, so that self-adaptive irrigation control is realized.

Description

Technical Field

[0001] The present invention mainly relates to the field of garden design and irrigation technology, and in particular to a self-adaptive irrigation control method and system based on soil permeability. Background Art

[0002] Existing irrigation control methods usually control irrigation based on a preset irrigation plan or manual judgment, both of which have corresponding defects.

[0003] There are several disadvantages to preset irrigation plans:

[0004] 1. Unable to adapt to real-time changes: Preset irrigation plans are often set based on fixed schedules and water volumes, and cannot be adjusted according to real-time changes in weather, soil moisture, plant needs, etc., which may lead to insufficient irrigation during droughts and waste of water resources during rainy seasons.

[0005] 2. Unable to precisely control the amount of irrigation: It is difficult to achieve precise water control with a preset irrigation plan, which may result in over-irrigation in some areas and under-irrigation in other areas, thus affecting the growth and health of plants.

[0006] 3. Lack of flexibility: Preset irrigation plans lack flexibility and cannot cope with emergencies or special circumstances, such as sudden rainfall or plant pests and diseases.

[0007] For manual judgment and control of irrigation, its main disadvantages include:

[0008] 1. Highly subjective: Manual judgment of irrigation amount often relies on the experience and intuition of maintenance personnel, which is highly subjective and may result in inaccurate irrigation amount.

[0009] 2. Inefficiency: Manual judgment of irrigation volume requires maintenance personnel to observe the growth status and water demand of plants on site and manually adjust irrigation equipment. This method is inefficient and difficult to cover large areas of gardens.

[0010] 3. High labor cost: Manual judgment of irrigation volume requires human cooperation, especially in large gardens, where more maintenance personnel need to be hired to complete irrigation work, resulting in high labor costs.

[0011] Therefore, whether it is a preset irrigation plan or manual judgment and control of irrigation methods, there are problems such as uneven irrigation, waste of water resources, high labor costs, and low efficiency.

[0012] More importantly, both the preset irrigation plan and the manual judgment and control irrigation methods do not take into account the actual permeability of the soil. However, changes in soil permeability will directly affect the irrigation effect. Too much irrigation water will lead to waste of water resources and excessive soil moisture, while too little water will cause vegetation water shortage. Therefore, the above irrigation methods cannot perceive the actual water demand. Summary of the invention

[0013] The present invention provides an adaptive irrigation control method based on soil permeability and a system thereof, so as to solve the problem that the existing irrigation method cannot determine the real water requirement of the soil.

[0014] In the first aspect of the present invention, an adaptive irrigation control method based on soil permeability is disclosed, the control method comprising:

[0015] Set the target soil moisture value H required for the irrigation control area 0 , and the early warning humidity threshold H that affects plant growth 1 , collect real-time soil moisture H;

[0016] If the current real-time soil moisture H is lower than the warning moisture threshold H 1 , a soil drought warning is issued, and a control instruction for the set fixed irrigation duration is activated;

[0017] If the current real-time soil moisture H is higher than the warning moisture threshold H 1 , and is lower than the preset target soil moisture value H 0 , combined with the current soil permeability, the irrigation time prediction model is called to calculate the irrigation time t, and the control instruction of the irrigation time t is started;

[0018] If the current real-time soil moisture H is higher than the preset target soil moisture value H 0 , generating a “turn off irrigation” control instruction.

[0019] Furthermore, during the control instruction process of starting irrigation, that is, during the irrigation process, the target soil moisture value is continuously collected in real time, and the control instruction of adjusting the irrigation duration is fed back based on the real-time collected target soil moisture value to adapt to the environmental changes in the irrigation control area.

[0020] Furthermore, the formula for calculating the irrigation time T by calling the irrigation time prediction model is:

[0021]

[0022] Where T represents the irrigation duration; Δθ represents the absolute value of the difference between the target soil moisture value and the current real-time soil moisture; D represents the soil depth, that is, the soil depth that needs to be replenished with water; V represents the irrigation water flow rate; and K represents the current soil permeability.

[0023] Furthermore, the current soil permeability is calculated based on the soil permeability fitting model M k Got it.

[0024] Furthermore, the soil permeability fitting model M k The construction process is:

[0025] S201. Extract input features and target variables, feature variables: data of irrigation control area including soil temperature, irrigation water consumption, soil depth and soil moisture; target variable: soil permeability;

[0026] S202. Standardize the parameters of the feature variables and the target variables;

[0027] S203. Divide the collected soil data into a training set and a test set;

[0028] S204. Create and train a linear regression model using the data in the training set;

[0029] S205. Use the test data in the test set to predict the trained linear regression model and output the predicted target value;

[0030] S206. Calculate the mean square error between the target variable value in the test set and the predicted target value obtained in S205;

[0031] S207. According to the mean square error calculated in S206, iteratively update the model parameters, repeat steps S205 and S206 to minimize the mean square error, until the regression coefficient array β of the linear regression model is outputted = [β 0 , β 1 , β 2 , β 3 ] and the intercept constant α 0 .

[0032] Furthermore, based on the regression coefficient array and intercept constant of the output linear regression model, the soil permeability formula of the irrigation control area is obtained:

[0033] K=α 0 +β 0 * Irrigation water consumption + β 1 *Soil depth + β 2 *Soil moisture + β 3 *Soil temperature

[0034] Where K is soil permeability, β 0 , β 1 , β 2 and β 3 They respectively represent the influence of irrigation water volume, soil depth, soil moisture and soil temperature on soil permeability.

[0035] In a second aspect of the present invention, the present invention also provides an adaptive irrigation control system based on soil permeability, the control system comprising:

[0036] The data collection unit subscribes to the sensor data of the environment and soil installed in the irrigation control area on the IOT platform to detect the soil environment data such as soil temperature, soil moisture, soil depth, soil saturated water content, etc. in the irrigation area in real time;

[0037] The data processing unit is connected to the data processing unit, and is used to receive the soil environment data sent by the data acquisition unit and perform standardized preprocessing on the data; then, based on the collected soil environment data, a soil permeability fitting model is constructed, and combined with the current soil data, the current soil permeability of the irrigation control area is fitted; based on the real-time soil moisture of the irrigation control area, it is determined whether to irrigate, and if irrigation is required, the irrigation duration prediction model is called, and the control instruction of the irrigation duration is obtained in combination with the current soil permeability, and an irrigation plan is formulated;

[0038] A warning unit, connected to the data processing unit, for monitoring and warning based on abnormal data processed by the data processing unit;

[0039] The irrigation unit is connected to the data processing unit, executes the irrigation plan, sends the control instructions of the irrigation duration to the message queue issued by the device instructions of the IOT platform, and transmits it to the irrigation controller to control the solenoid valve to open and irrigate the soil.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] 1. The present invention provides an adaptive irrigation control method and system based on soil permeability. The method and system can determine whether irrigation is needed according to soil moisture by collecting soil parameters in real time. A warning threshold and a preset humidity value are provided to form three irrigation judgment results, which increases the output result types, avoids the waste of water resources and uneven watering caused by manual irrigation or a single judgment criterion, improves the utilization rate of water resources, and is beneficial to improving the growth trend of plants.

[0042] 2. When the soil is dry and needs irrigation, the present invention predicts the irrigation time under the current environment based on the soil permeability, which is more in line with the actual water demand of the soil. It optimizes the irrigation process and reduces the impact of excessive or insufficient watering on plants.

[0043] 3. In addition, the adaptive irrigation control method and system based on soil permeability provided by the present invention can train and construct a soil permeability fitting model according to different soil types and environmental conditions, adjust parameter settings, and have stronger adaptability.

[0044] The present invention will be explained in detail below with reference to the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A flowchart of an adaptive irrigation control method based on soil permeability in an embodiment of the present invention;

[0046] Figure 2 The figure is a structural diagram of an adaptive irrigation control system based on soil permeability in an embodiment of the present invention. DETAILED DESCRIPTION

[0047] To facilitate understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings, but the present invention can be implemented in different forms and is not limited to the embodiments described in the text. On the contrary, these embodiments are provided to make the content disclosed in the present invention more thorough and comprehensive.

[0048] Due to the differences in parameters such as soil permeability and soil moisture changes, current irrigation methods such as preset irrigation plans and manual watering cannot clearly determine the actual water demand. Too much or too little irrigation water will affect the healthy growth of plants. In order to solve the problem that irrigation methods such as preset plans and manual watering cannot clearly determine the actual water demand of the soil, the present invention discloses an adaptive irrigation control method based on soil permeability, comprising the following steps:

[0049] S1. Data collection: Real-time monitoring and collection of environmental data in the irrigation control area, including soil parameters such as soil moisture, soil temperature, soil depth, and irrigation water consumption. Soil depth refers to the depth of soil that needs to be replenished with water.

[0050] S2. Constructing a soil permeability fitting model: Collect soil environmental data of the irrigation control area for 3 months or more, including soil temperature, soil moisture, soil depth, irrigation water consumption, etc., and construct a soil permeability fitting model M based on linear regression and mean square error evaluation indicators. k , which is used to fit the soil permeability K.

[0051] Soil permeability fitting model M k The specific construction process is:

[0052] S201. Extract input features and target variables, feature variables: environmental parameters of the irrigation control area including soil temperature, irrigation water consumption, soil depth, soil moisture, etc.; target variable: soil permeability.

[0053] It should be noted that the type and quantity of input characteristic variables can increase with technological development and are not limited to the above-mentioned soil temperature, irrigation water consumption, soil depth and soil moisture, but can also include other parameters such as soil saturated moisture content, air humidity, air temperature, etc.

[0054] S202. Standardize the parameters of the feature variables and the target variables.

[0055] S203. Divide the soil environmental data collected over several months into a training set and a test set.

[0056] S204. Create and train a linear regression model using the data in the training set.

[0057] S205. Use the test data in the test set to predict the trained linear regression model and output the predicted target value.

[0058] S206. Calculate the mean square error between the target variable value in the test set and the predicted target value obtained in S205.

[0059] S207. The predictive ability of the model on unseen data is evaluated by the mean square error. Minimizing the mean square error is the optimization goal of the linear regression model.

[0060] That is, according to the mean square error calculated in S206, the model parameters are iteratively updated by the batch gradient descent (BGD) method, and steps S205 and S206 are repeated to minimize the mean square error until the regression coefficient array β of the linear regression model is outputted as [β 0 , β 1 , β 2 , β 3 ] and the intercept constant α 0 .

[0061] According to the regression coefficient array and intercept constant of the output linear regression model, the soil permeability formula of the irrigation control area is obtained:

[0062] K=α 0 +β 0 * Irrigation water consumption + β 1 *Soil depth + β 2 *Soil moisture + β 3 *Soil temperature

[0063] Where K is soil permeability, β 0 , β 1 , β 2 and β 3 They respectively represent the influence of irrigation water volume, soil depth, soil moisture and soil temperature on soil permeability.

[0064] S3. Fitting the current soil permeability: Every morning (a fixed time can be set as needed, such as 3 a.m.), the soil temperature, soil moisture, soil depth, irrigation water consumption and other data collected by the equipment in the irrigation area the day before are used to fit the soil permeability model M. k , calculate the soil permeability K of the previous day 0 .

[0065] S4. Analysis of irrigation duration prediction model: Based on the irrigation duration prediction model M t , combined with the soil permeability K of the previous day 0 And the soil environment data of the irrigation control area, get the time t required for irrigation when the set soil moisture is reached under the current environment, and then create an irrigation task based on the time t.

[0066] The above irrigation duration prediction model M t The construction process is:

[0067] Irrigation duration prediction model M t It is used to calculate the time T required for irrigation when the set soil moisture is reached in the current environment, and then create an irrigation task based on the time T, so as to achieve adaptive irrigation control. The specific calculation formula is:

[0068]

[0069] Wherein, T represents the irrigation duration (unit: second); Δθ represents the absolute value of the difference between the target soil moisture (θtarget) and the current real-time soil moisture (θcurrent). For example, if the target humidity is 0.30 and the current humidity is 0.20, then Δθ = 0.30-0.20 = 0.10; D represents the soil depth, that is, the soil depth that needs to be replenished with water (unit: meter or centimeter); V represents the irrigation water flow rate (unit: cubic meter / second); K represents the current soil permeability (unit: meter / second or centimeter / second), which reflects the penetration rate of water in the soil.

[0070] S5. Configure humidity standard value and abnormal threshold: according to the soil moisture requirements of different plants, set the target soil humidity value H required for the irrigation control area. 0 , and the early warning humidity threshold H that affects plant growth 1 .

[0071] S6. Real-time data collection: Real-time data collection of soil moisture H in the irrigation control area and upload to the irrigation duration prediction model M t .

[0072] S7. Calculate irrigation time: according to the set target soil moisture value H 0 , collected real-time soil moisture H, soil depth D 0, soil permeability K on the previous day 0 and normal irrigation flow V 0 , the irrigation time prediction model is used to calculate the current irrigation task execution time t, the formula is:

[0073] t=(|H 0 –H|)*D 0 / (V 0 *K0)

[0074] S8. Adaptive control: If the current soil moisture H is lower than the warning moisture threshold H 1 , it means that the current soil moisture is lower than the warning humidity threshold that affects plant growth, triggering a soil drought warning, and generating "start irrigation" at the same time. It also starts a control instruction to set a fixed irrigation duration and then continues to judge the soil condition to decide whether to continue irrigation. The default is generally 5 minutes. At this time, the soil is dry, but it does not necessarily require a lot of water, so it is sufficient to irrigate for about 5 minutes first, avoiding the waste of water resources caused by watering once the threshold is lower than the traditional operation method. Of course, 5 minutes can also be set to other values ​​according to plant type or needs.

[0075] If the current soil moisture H is higher than the warning moisture threshold H 1 , and is lower than the preset humidity value H 0 , indicating that the current soil moisture has not reached the drought level, but has not reached the set ideal target soil moisture, so the irrigation duration prediction model M is called. t Calculate the irrigation duration t, and generate a "turn on irrigation" instruction, and the control instruction for turning on irrigation duration t.

[0076] If the current soil humidity H is higher than the preset humidity value H 0 , indicating that the current soil moisture is higher than the set ideal target soil moisture and irrigation cannot continue, a "turn off irrigation" control instruction is generated.

[0077] S9. Execute irrigation: send sprinkler control instructions to the sprinkler controller to execute irrigation.

[0078] S10. Feedback adjustment: Based on the actual irrigation effect, that is, real-time detection of soil moisture value, jump to step S6, loop the above steps, and realize adaptive irrigation control.

[0079] Among them, the current soil moisture H is lower than the warning humidity threshold H 1 When irrigation is started, the irrigation duration is set first and then step S6 is performed to collect the soil moisture H in the irrigation control area in real time. Because the change of soil moisture value is delayed in data reporting during irrigation, each irrigation duration is set to a fixed duration, and a judgment is made after the soil moisture data is reported. Such a cycle iteration can prevent over-irrigation.

[0080] Through the above soil permeability fitting model M k and irrigation duration prediction model M t , and obtain the irrigation time t based on soil infiltration rate.

[0081] Since the soil will become compacted during long-term irrigation, which will reduce the soil permeability, the irrigation time needs to be extended to reach the preset soil moisture value. k To determine and fit the current soil permeability, and combine it with the irrigation duration prediction model M t To calculate the irrigation time in real time, the irrigation time based on soil permeability is obtained. The irrigation time is adjusted with the soil permeability, which makes watering more thorough and beneficial to plant growth. It can also dynamically adjust the irrigation time of the irrigation task to further ensure the irrigation effect.

[0082] Adaptive control irrigation is an intelligent irrigation technology that can dynamically adjust irrigation strategies based on real-time monitored environmental data and plant needs. Compared with traditional fixed-time and fixed-volume irrigation methods, adaptive control irrigation has the following significant advantages and effects:

[0083] 1. Improve water resource utilization efficiency: Adaptive irrigation control can be adjusted according to real-time data such as soil moisture, meteorological data, and plant water requirements to avoid over-irrigation or under-irrigation, thereby greatly improving water resource utilization efficiency. For example, in rainy weather, soil moisture changes, soil permeability and irrigation time change accordingly, shortening irrigation time or stopping irrigation can improve water resource utilization.

[0084] 2. Improve crop yield and quality: by accurately setting the target soil moisture value H at different growth stages of plants 0 and the early warning humidity threshold H that affects plant growth 1 , so that the irrigation time and amount can be accurately controlled, ensuring that plants get enough water during the critical growth period, promoting their growth and development, and ultimately improving crop yield and quality.

[0085] 3. Save energy and reduce labor: The adaptive control irrigation system sets relevant parameters and executes automatically, which can reduce human intervention and manual work, thereby reducing labor costs. At the same time, reasonable irrigation also reduces energy waste caused by excessive irrigation.

[0086] 4. Protect soil structure and prevent soil erosion: By adopting the adaptive control method designed by the present invention and scientifically managing irrigation, it is possible to avoid soil structure damage and soil erosion caused by excessive water and maintain soil health and fertility.

[0087] Based on the above-mentioned adaptive irrigation control method, the present invention also provides an adaptive irrigation control system based on soil permeability, the system comprising:

[0088] The data collection unit subscribes to the IOT platform to install sensors in the environment and soil of the irrigation control area to detect soil environment data such as soil temperature, soil moisture, soil depth, and irrigation water consumption in the irrigation area in real time.

[0089] The data processing unit performs standardized preprocessing on the collected sensor data; then, based on the collected soil environmental data, a soil permeability fitting model is constructed, and combined with the current soil data, the current soil permeability of the irrigation control area is fitted; based on the real-time soil moisture in the irrigation control area, it is determined whether irrigation is required. If irrigation is required, the irrigation duration prediction model is called, and the irrigation duration control instruction is obtained in combination with the current soil permeability to formulate an irrigation plan.

[0090] The warning unit is connected to the data processing unit and is used to monitor and warn based on the abnormal data processed by the data processing unit, such as soil drought warning.

[0091] The irrigation unit is connected to the data processing unit, executes the irrigation plan, sends the control instructions of the irrigation duration to the message queue issued by the device instructions of the IOT platform, and transmits it to the irrigation controller to control the solenoid valve to open and irrigate the soil.

[0092] The present invention provides an adaptive irrigation control method and system based on soil permeability, which constructs a model M based on linear regression and mean square error evaluation index. k The current soil permeability is fitted, and then the irrigation duration prediction model M is used based on the current meteorological data, soil moisture and other environmental data. t , can predict the irrigation duration under the current environment, thereby achieving adaptive irrigation control.

[0093] Adaptively adjusting irrigation plans based on soil permeability and determining irrigation water requirements based on soil permeability can greatly improve water resource utilization compared to fixed irrigation or operations based on manual watering experience.

[0094] The above is an exemplary description of the present invention in combination with the accompanying drawings. It is obvious that the specific implementation of the present invention is not limited to the above-mentioned method. As long as such non-substantial improvements are made by adopting the method concept and technical solution of the present invention, or the concept and technical solution of the present invention are directly applied to other occasions without improvement, they are all within the protection scope of the present invention.

Claims

1. An adaptive irrigation control method based on soil permeability, characterized in that: The control method comprises: Set the target soil moisture value H0 required for the irrigation control area and the early warning moisture threshold H1 that affects plant growth, and collect real-time soil moisture H; If the current real-time soil moisture H is lower than the warning moisture threshold H1, a soil drought warning is issued and a control instruction for the set fixed irrigation duration is started; If the current real-time soil moisture H is higher than the warning moisture threshold H1 and lower than the preset target soil moisture value H0, the irrigation duration prediction model is called in combination with the current soil permeability to calculate the irrigation duration t, and the control instruction for irrigation duration t is started; If the current real-time soil moisture H is higher than the preset target soil moisture value H0, a "turn off irrigation" control instruction is generated.

2. The method of adaptive irrigation control based on soil permeability according to claim 1, characterized in that: During the control instruction process of starting irrigation, that is, during the irrigation process, the target soil moisture value is continuously collected in real time, and the control instruction of adjusting the irrigation duration is fed back based on the real-time collected target soil moisture value to adapt to the environmental changes in the irrigation control area.

3. The method for adaptive irrigation control based on soil permeability according to claim 1, characterized in that: The formula for calling the irrigation duration prediction model to calculate the irrigation duration T is: Where T represents the irrigation duration; Δθ represents the absolute value of the difference between the target soil moisture value and the current real-time soil moisture; D represents the soil depth, that is, the soil depth that needs to be replenished with water; V represents the irrigation water flow rate; and K represents the current soil permeability.

4. The method for adaptive irrigation control based on soil permeability according to any one of claims 1 to 3, characterized in that: The current soil permeability is based on the soil permeability fitting model M k Got it.

5. The method for adaptive irrigation control based on soil permeability according to claim 4, characterized in that: The soil permeability fitting model M k The construction process is: S201. Extract input features and target variables, feature variables: data of irrigation control area including soil temperature, irrigation water consumption, soil depth and soil moisture; target variable: soil permeability; S202. Standardize the parameters of the feature variables and the target variables; S203. Divide the collected soil data into a training set and a test set; S204. Create and train a linear regression model using the data in the training set; S205. Use the test data in the test set to predict the trained linear regression model and output the predicted target value; S206. Calculate the mean square error between the target variable value in the test set and the predicted target value obtained in S205; S207. According to the size of the mean square error calculated in S206, iteratively update the model parameters, repeat steps S205 and S206 to minimize the mean square error, until the regression coefficient array β = [β0, β1, β2, β3] and the intercept constant α0 of the linear regression model are output.

6. The method for adaptive irrigation control based on soil permeability according to claim 5, characterized in that: According to the regression coefficient array and intercept constant of the output linear regression model, the soil permeability formula of the irrigation control area is obtained: K=α0+β0*irrigation water consumption+β1*soil depth+β2*soil moisture+β3*soil temperature Where K is soil permeability, β0, β1, β2 and β3 represent the influence of irrigation water volume, soil depth, soil moisture and soil temperature on soil permeability, respectively.

7. An adaptive irrigation control system based on soil permeability, characterized in that: The control system comprises: The data collection unit subscribes to the sensor data of the environment and soil installed in the irrigation control area on the IOT platform to detect the soil environment data such as soil temperature, soil moisture, soil depth, soil saturated water content, etc. in the irrigation area in real time; The data processing unit is connected to the data processing unit, and is used to receive the soil environment data sent by the data acquisition unit and perform standardized preprocessing on the data; then, based on the collected soil environment data, a soil permeability fitting model is constructed, and combined with the current soil data, the current soil permeability of the irrigation control area is fitted; based on the real-time soil moisture of the irrigation control area, it is determined whether to irrigate, and if irrigation is required, the irrigation duration prediction model is called, and the control instruction of the irrigation duration is obtained in combination with the current soil permeability, and an irrigation plan is formulated; A warning unit, connected to the data processing unit, for monitoring and warning based on abnormal data processed by the data processing unit; The irrigation unit is connected to the data processing unit, executes the irrigation plan, sends the control instructions of the irrigation duration to the message queue issued by the device instructions of the IOT platform, and transmits it to the irrigation controller to control the solenoid valve to open and irrigate the soil.

Images