Efficient water-saving irrigation method based on data processing

Abstract

The invention provides an efficient water-saving irrigation method based on data processing, which comprises the following steps: setting a data acquisition network and an edge computing device in a target irrigation area, acquiring a multi-source monitoring signal of the target irrigation area by using the data acquisition network, and performing adaptive filtering processing on the multi-source monitoring signal by using the edge computing device, filtering data are obtained; obtaining filtering data by using a cloud computing platform, and obtaining a prediction result of an irrigation decision through a built-in irrigation prediction model of the cloud computing platform; generating a control instruction according to the prediction result, and dynamically adjusting an electric valve and a water pump of a preset irrigation system according to the control instruction. Various technologies of real-time monitoring, data processing, intelligent prediction and automatic regulation and control are integrated, the utilization efficiency of water resources can be greatly improved while crop growth conditions are guaranteed, and accurate and efficient irrigation management is achieved. The comprehensive system advantage has a profound influence on agricultural sustainable development and water resource management.

Description

Technical Field

[0001] The present invention relates to the technical field of smart agriculture, and in particular to a high-efficiency water-saving irrigation method based on data processing. Background Art

[0002] With the increase of global population and the acceleration of urbanization, water shortage has become an important factor restricting the sustainable development of agriculture. Agricultural water accounts for about 70% of global freshwater consumption, and traditional irrigation methods often have problems such as serious waste of water resources and low irrigation efficiency. Therefore, the development of efficient and water-saving irrigation technology is particularly important for improving agricultural production efficiency, ensuring food security and achieving sustainable use of water resources.

[0003] The invention patent with application number 202311174370.2 discloses an intelligent and efficient water-saving irrigation method and system, which relates to the field of intelligent irrigation technology and includes the following steps: obtaining the status data of the area to be irrigated, and obtaining the irrigation water demand of the area to be irrigated based on the status data, wherein the status data includes the water shortage image data and water loss data of the vegetation in the irrigation area; obtaining the working data of the irrigation equipment in the area to be irrigated, wherein the working data includes the real-time irrigation water pressure, real-time flow data and the position distance between the irrigation equipment and the water pump of the irrigation equipment in the area to be irrigated; according to the irrigation water demand and working data, the output power of the water pump is obtained, and the water pump is controlled to pump water to the irrigation equipment according to the output power. The patent can avoid the long-term high-power operation of the water pump, which can not only reduce the waste of power resources, but also save water resources and improve the economic benefits of irrigation work. However, the system relies on the acquired status data, including the water shortage image data and water loss data of the vegetation. If there is an error in the sensor or image recognition algorithm, the calculation of the irrigation water demand may be inaccurate, thereby affecting the irrigation effect. And the irrigation demand is affected by many factors, such as weather changes, soil types and humidity changes. If these changes are not reflected in the status data in a timely manner, it will lead to inappropriate irrigation timing and water volume, resulting in waste of resources or plant damage. Summary of the invention

[0004] In order to overcome the deficiencies of the prior art, the object of the present invention is to provide an efficient water-saving irrigation method based on data processing, which can quickly respond to environmental changes and achieve efficient water-saving irrigation without losing accuracy.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] An efficient water-saving irrigation method based on data processing, comprising:

[0007] A data acquisition network and an edge computing device are set up in the target irrigation area, and the data acquisition network is used to collect multi-source monitoring signals of the target irrigation area; the data acquisition network includes a soil moisture sensor, a temperature sensor, a light intensity sensor, and a meteorological sensor;

[0008] Using the edge computing device to perform adaptive filtering on the multi-source monitoring signal to obtain filtered data;

[0009] The filtered data is obtained by using a cloud computing platform, and a prediction result of an irrigation decision is obtained through an irrigation prediction model built into the cloud computing platform; the prediction result includes the required irrigation time and the required irrigation water volume;

[0010] A control instruction is generated according to the prediction result, and the electric valve and water pump of the preset irrigation system are dynamically adjusted according to the control instruction.

[0011] Preferably, the multi-source monitoring signal includes: a soil moisture sub-signal, a temperature sub-signal, a light intensity sub-signal and a meteorological sub-signal; the meteorological sub-signal is used to reflect wind direction, wind speed, air pressure and rainfall.

[0012] Preferably, the edge computing device is used to perform adaptive filtering on the multi-source monitoring signal to obtain filtered data, including:

[0013] Shifting the multi-source monitoring signals to obtain shifted signals;

[0014] Performing wavelet decomposition on the shifted signal to obtain a plurality of wavelet coefficients;

[0015] Determine the filtering threshold according to the decomposition scale and length of the signal;

[0016] Constructing a denoising function according to the filtering threshold;

[0017] The noise of the translated signal is removed by using the denoising function to obtain the filtered data.

[0018] Preferably, the calculation formula of the filtering threshold is: Among them, λ is the filtering threshold, j is the decomposition scale of the translated signal, and d j is the decomposition of the wavelet coefficients with a scale of j, N represents the length of the shifted signal, and median represents the median operation.

[0019] Preferably, removing the noise of the translated signal by using the denoising function to obtain the filtered data comprises:

[0020] The corresponding wavelet coefficients are removed by using the denoising function to obtain a smoothed signal; wherein the denoising function is: Among them, X represents the wavelet coefficient, n represents the adjustment coefficient;

[0021] Performing inverse translation on the smoothed signal to obtain an inverse translated signal;

[0022] Perform analog-to-digital conversion on the inverse-shifted signal to obtain the filtered data.

[0023] Preferably, the method for constructing the irrigation prediction model comprises:

[0024] Acquire a first data set constructed by filtering historical multi-source monitoring signals;

[0025] obtaining a second data set constructed from historical irrigation decision data;

[0026] The first data set and the second data set are input into an initial neural network for training to obtain the trained irrigation prediction model.

[0027] Preferably, the irrigation prediction model includes an input layer, an LSTM layer, a fully connected layer and an output layer connected in sequence; the LSTM layer includes a first sublayer and a second sublayer; the first sublayer includes 128 units, and the second sublayer includes 64 units.

[0028] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0029] The present invention provides an efficient water-saving irrigation method based on data processing, comprising: setting a data acquisition network and an edge computing device in a target irrigation area, and using the data acquisition network to collect multi-source monitoring signals of the target irrigation area; the data acquisition network includes a soil moisture sensor, a temperature sensor, a light intensity sensor and a meteorological sensor; using the edge computing device to perform adaptive filtering processing on the multi-source monitoring signals to obtain filtered data; using a cloud computing platform to obtain the filtered data, and obtaining the prediction results of irrigation decisions through the irrigation prediction model built into the cloud computing platform; the prediction results include the required irrigation time and the required irrigation water volume; generating control instructions according to the prediction results, and dynamically adjusting the electric valves and water pumps of the preset irrigation system according to the control instructions. The data acquisition network of the present invention is responsible for continuously updating the monitoring data, and after preliminary data cleaning and feature extraction through edge computing, it is finally input into the irrigation prediction model for training and prediction, so as to generate corresponding control instructions and realize dynamic irrigation scheme adjustment. In this way, the whole system has a strong adaptability and intelligence level, and can accurately meet the irrigation needs of plant growth. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0031] Figure 1 A flow chart of a method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0034] Figure 1 A flow chart of a method provided by an embodiment of the present invention, such as Figure 1 As shown, the present invention provides a high-efficiency water-saving irrigation method based on data processing, comprising:

[0035] Step 100: Setting a data acquisition network and an edge computing device in a target irrigation area, and using the data acquisition network to collect multi-source monitoring signals of the target irrigation area; the data acquisition network includes a soil moisture sensor, a temperature sensor, a light intensity sensor, and a meteorological sensor;

[0036] Step 200: using the edge computing device to perform adaptive filtering on the multi-source monitoring signal to obtain filtered data;

[0037] Step 300: using a cloud computing platform to obtain the filtered data, and using an irrigation prediction model built into the cloud computing platform to obtain a prediction result of an irrigation decision; the prediction result includes the required irrigation time and the required irrigation water volume;

[0038] Step 400: Generate a control instruction according to the prediction result, and dynamically adjust the electric valve and water pump of the preset irrigation system according to the control instruction.

[0039] Specifically, the setting principle of the data acquisition network in this embodiment is: uniform distribution. In the irrigation area, soil moisture sensors, temperature sensors, light intensity sensors and meteorological sensors should be evenly distributed so that the system can comprehensively monitor environmental changes in the area. In addition, soil type and crop distribution must also be considered. Different soil types and crop types may lead to different irrigation needs, so sensors should be arranged at representative locations of different types of soil and crop distribution areas.

[0040] Furthermore, when on hillsides or in areas with slopes, sensors should be set up at different heights to obtain more accurate information on soil moisture and water dynamics. If there are areas prone to waterlogging, sensors should be set up in these areas to detect waterlogging problems early and avoid soaking of crop roots.

[0041] Furthermore, the sensor of this embodiment is placed as close to the water source as possible, which not only facilitates water volume monitoring but also helps to make quick irrigation decisions.

[0042] Specifically, the consideration principles for the location of the edge computing device in this embodiment include:

[0043] a. Network coverage: Edge computing devices are set up in locations with good network signals to ensure real-time data transmission and processing with sensors and central servers. Edge computing devices need to cover the entire irrigation area to process data from various sensors and perform preliminary calculations.

[0044] b. Equipment accessibility: Edge computing equipment should be set up in an easily accessible location for regular maintenance, firmware updates, and troubleshooting. When setting up, consider the equipment's waterproof, dustproof, and windproof protection measures to ensure its long-term stable operation.

[0045] c. Power supply. Edge computing devices require a stable power supply, so you should choose a location close to a power source, or ensure that it can be powered by alternative energy sources such as solar energy.

[0046] In summary, the location of the data collection network and edge computing equipment in this embodiment needs to comprehensively consider factors such as the geographical characteristics of the region, soil and crop types, water source location, network coverage requirements, equipment maintenance convenience, and environmental protection. Reasonable location can improve the efficiency and reliability of the irrigation system and ensure the healthy growth of crops.

[0047] Preferably, the multi-source monitoring signal includes: a soil moisture sub-signal, a temperature sub-signal, a light intensity sub-signal and a meteorological sub-signal; the meteorological sub-signal is used to reflect wind direction, wind speed, air pressure and rainfall.

[0048] Preferably, the edge computing device is used to perform adaptive filtering on the multi-source monitoring signal to obtain filtered data, including:

[0049] Shifting the multi-source monitoring signals to obtain shifted signals;

[0050] Performing wavelet decomposition on the shifted signal to obtain a plurality of wavelet coefficients;

[0051] Determine the filtering threshold according to the decomposition scale and length of the signal;

[0052] Constructing a denoising function according to the filtering threshold;

[0053] The noise of the translated signal is removed by using the denoising function to obtain the filtered data.

[0054] Preferably, the calculation formula of the filtering threshold is: Among them, λ is the filtering threshold, j is the decomposition scale of the translated signal, and d j is the decomposition of the wavelet coefficients with a scale of j, N represents the length of the shifted signal, and median represents the median operation.

[0055] Specifically, in the wavelet threshold denoising process, the quality of threshold selection directly affects the effect of denoising. Since the wavelet coefficient amplitude of noise has the characteristic of decreasing with the increase of decomposition scale, this embodiment introduces the decomposition scale to adaptively adjust the filtering threshold.

[0056] Preferably, removing the noise of the translated signal by using the denoising function to obtain the filtered data comprises:

[0057] The corresponding wavelet coefficients are removed by using the denoising function to obtain a smoothed signal; wherein the denoising function is: Among them, X represents the wavelet coefficient, n represents the adjustment coefficient;

[0058] Performing inverse translation on the smoothed signal to obtain an inverse translated signal;

[0059] Perform analog-to-digital conversion on the inverse-shifted signal to obtain the filtered data.

[0060] Specifically, this embodiment uses wavelet coefficients to measure the attenuation deviation, so that the attenuation of wavelet coefficients with larger absolute values ​​decreases as their absolute values ​​increase, thereby effectively avoiding the loss of high-frequency information and improving the signal-to-noise ratio of the signal.

[0061] Preferably, the method for constructing the irrigation prediction model comprises:

[0062] Acquire a first data set constructed by filtering historical multi-source monitoring signals;

[0063] obtaining a second data set constructed from historical irrigation decision data;

[0064] The first data set and the second data set are input into an initial neural network for training to obtain the trained irrigation prediction model.

[0065] Preferably, the irrigation prediction model includes an input layer, an LSTM layer, a fully connected layer and an output layer connected in sequence; the LSTM layer includes a first sublayer and a second sublayer; the first sublayer includes 128 units, and the second sublayer includes 64 units.

[0066] Specifically, this embodiment considers time series data (such as changes in soil moisture and meteorological variables), and the LSTM network is more suitable. LSTM is a special recursive neural network that is suitable for processing time series data and has the ability to remember long-term dependencies.

[0067] Furthermore, the network structure of the irrigation prediction model includes:

[0068] Input layer: The input data format is usually a three-dimensional array (number of samples, time steps, number of features).

[0069] LSTM layer, the first layer LSTM includes 128 units and returns a sequence (returns the output of each time step to facilitate processing by the next layer); the second layer LSTM includes 64 units and does not return a sequence (only returns the output of the last time step)

[0070] Fully connected layer: The fully connected layer is connected after the LSTM layer, and the output types and features are the same (i.e., the time and amount of water required for irrigation).

[0071] Output layer: It has 2 nodes (corresponding to the amount of water and time required for irrigation), and usually uses a linear activation function.

[0072] Optionally, the output of the first LSTM layer (the state of each time step) is passed to the second LSTM layer, forming a strong time dependency. The final state of the LSTM layer (the output of the last time step) is connected to the fully connected layer. Exemplarily, the model structure of this embodiment is: input layer (number of samples, time step, n features) --> LSTM layer 1 (128 units, return sequence) --> LSTM layer 2 (64 units) --> fully connected layer (2 nodes) --> output layer (2 nodes).

[0073] Furthermore, the steps of processing the prediction results of this embodiment are as follows:

[0074] 1. This embodiment analyzes the output results obtained from the irrigation prediction model. The results include:

[0075] Amount of water required for irrigation: expressed in volume units (such as cubic meters).

[0076] Time required for irrigation: expressed in time units (such as minutes).

[0077] 2. Based on the prediction results, the main steps for generating control instructions are as follows:

[0078] a. Define the control logic according to the specific irrigation system and crop needs:

[0079] If the amount of water required for irrigation is greater than a certain threshold, the water pump is turned on and a certain flow rate is maintained.

[0080] If the irrigation time required exceeds the preset value, irrigation is carried out in batches to avoid over-irrigation.

[0081] b. Calculate valve and pump settings

[0082] The predicted water volume and time are used to calculate the valve opening ratio and the operating parameters of the pump. The control instructions include:

[0083] Electric valve opening, the proportion of the valve opening according to demand (such as 80% open).

[0084] Water pump flow setting, set the flow rate of the water pump (such as how many cubic meters flow out per minute).

[0085] c. Generate specific instruction format

[0086] Generate control instructions according to the above rules to form an executable command format.

[0087] The beneficial effects of the present invention are as follows:

[0088] (1) By setting up soil moisture sensors, temperature sensors, light intensity sensors and meteorological sensors, the system can obtain detailed environmental data to ensure a comprehensive understanding of irrigation conditions. The data acquisition network can monitor environmental changes in real time, promptly reflect changes in soil and climate conditions, and facilitate rapid response.

[0089] (2) The edge computing device of the present invention performs adaptive filtering on multi-source monitoring signals, which can effectively reduce noise and improve data quality. The present invention can obtain more accurate input data, thereby improving the accuracy of model prediction. Preliminary data processing on the edge computing device reduces the need to transmit raw data to the cloud platform, reduces latency, and improves the response speed of the system.

[0090] (3) The present invention further analyzes the filtered data through a cloud computing platform, and can apply more complex machine learning models to make irrigation predictions and generate efficient irrigation decisions. Based on real-time data, the system can automatically adjust irrigation strategies to adapt to different weather conditions and crop needs, thereby achieving precision irrigation.

[0091] (4) The irrigation plan generated by the present invention according to the specific needs of plants avoids over-irrigation and waste of water resources, ensuring that water is effectively used. In addition, it reduces unnecessary water consumption and avoids environmental problems such as soil salinization caused by excessive irrigation, which helps protect the ecological environment.

[0092] (5) The modular design of the present invention allows new sensors and computing devices to be easily added and expanded to new irrigation areas. As technology continues to advance, cloud computing platforms and models can be regularly updated, such as by introducing more efficient algorithms, thereby improving the intelligence level of the entire system.

[0093] (6) The present invention combines historical data with real-time data to make irrigation decisions more scientific and reliable, reducing the errors caused by empiricism. The present invention can also continuously collect and store data, conduct long-term analysis and optimization, and further improve the effect of irrigation strategies.

[0094] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0095] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. An efficient water-saving irrigation method based on data processing, characterized in that: include: A data acquisition network and an edge computing device are set up in the target irrigation area, and the data acquisition network is used to collect multi-source monitoring signals of the target irrigation area; the data acquisition network includes a soil moisture sensor, a temperature sensor, a light intensity sensor, and a meteorological sensor; Using the edge computing device to perform adaptive filtering on the multi-source monitoring signal to obtain filtered data; The filtered data is obtained by using a cloud computing platform, and a prediction result of an irrigation decision is obtained through an irrigation prediction model built into the cloud computing platform; the prediction result includes the required irrigation time and the required irrigation water volume; A control instruction is generated according to the prediction result, and the electric valve and water pump of the preset irrigation system are dynamically adjusted according to the control instruction.

2. The efficient water-saving irrigation method based on data processing according to claim 1, characterized in that: The multi-source monitoring signal includes: a soil moisture sub-signal, a temperature sub-signal, a light intensity sub-signal and a meteorological sub-signal; the meteorological sub-signal is used to reflect wind direction, wind speed, air pressure and rainfall.

3. The efficient water-saving irrigation method based on data processing according to claim 1, characterized in that: The edge computing device is used to perform adaptive filtering on the multi-source monitoring signal to obtain filtering data, including: Shifting the multi-source monitoring signals to obtain shifted signals; Performing wavelet decomposition on the shifted signal to obtain a plurality of wavelet coefficients; Determine the filtering threshold according to the decomposition scale and length of the signal; Constructing a denoising function according to the filtering threshold; The noise of the translated signal is removed by using the denoising function to obtain the filtered data.

4. The high-efficiency water-saving irrigation method based on data processing according to claim 3 is characterized in that: The calculation formula of the filtering threshold is: Among them, λ is the filtering threshold, j is the decomposition scale of the translated signal, and d j is the decomposition of the wavelet coefficients with a scale of j, N represents the length of the shifted signal, and median represents the median operation.

5. The efficient water-saving irrigation method based on data processing according to claim 4 is characterized in that: Using the denoising function to remove the noise of the translated signal to obtain the filtered data includes: The corresponding wavelet coefficients are removed by using the denoising function to obtain a smoothed signal; wherein the denoising function is: Among them, X represents the wavelet coefficient, n represents the adjustment coefficient; Performing inverse translation on the smoothed signal to obtain an inverse translated signal; Perform analog-to-digital conversion on the inverse-shifted signal to obtain the filtered data.

6. The efficient water-saving irrigation method based on data processing according to claim 1, characterized in that: The method for constructing the irrigation prediction model comprises: Acquire a first data set constructed by filtering historical multi-source monitoring signals; obtaining a second data set constructed from historical irrigation decision data; The first data set and the second data set are input into an initial neural network for training to obtain the trained irrigation prediction model.

7. The high-efficiency water-saving irrigation method based on data processing according to claim 1, characterized in that: The irrigation prediction model includes an input layer, an LSTM layer, a fully connected layer and an output layer which are connected in sequence; the LSTM layer includes a first sublayer and a second sublayer; the first sublayer includes 128 units, and the second sublayer includes 64 units.

Images