Agricultural irrigation power consumption management system and method

Abstract

The invention discloses an agricultural irrigation power consumption management method, and belongs to the technical field of agriculture, and the method comprises the following steps: S1, constructing an agricultural planting region three-dimensional model, and carrying out the visualization; s2, acquiring a color histogram of an agricultural planting area, setting a threshold value based on color features of soil and crops, segmenting the soil and the crops, and counting the number of the segmented crops; s3, judging the growth stage of crops in the agricultural planting area; s4, predicting the irrigation electricity consumption of the agricultural planting area; s5, collecting the irrigation electricity consumption of the agricultural planting area in real time, comparing the irrigation electricity consumption with the predicted irrigation electricity consumption, judging the reasonability of the irrigation electricity consumption of the agricultural planting area, reporting the irrigation electricity consumption if the irrigation electricity consumption is reasonable, and otherwise, performing self-inspection, correction and reporting; the invention also discloses an agricultural irrigation power consumption management system. According to the invention, the agricultural planting area can be visualized, so that workers can clearly know the crop planting condition of the agricultural planting area, and the agricultural irrigation electricity consumption can be efficiently managed in real time.

Description

Technical Field

[0001] The present invention belongs to the technical field of agriculture, and in particular relates to an agricultural irrigation electricity management system and method. Background Art

[0002] Agricultural irrigation is an important link in agricultural production, and its electricity management is directly related to agricultural production efficiency, energy utilization and the safety of farmers' lives and property.

[0003] Chinese patent application No. 201910244303.0 discloses an agricultural irrigation electricity management system and method, including obtaining card data of an agricultural irrigation card, and controlling the start of a water pump according to the card data; remotely reading the electricity consumption of the water pump during the start-up process, obtaining and storing electricity consumption information; settling fees based on the electricity consumption information, obtaining and storing settlement information; establishing a reporting network, and reporting both the electricity consumption information and the settlement information to the power supply system through the reporting network.

[0004] The applicant has found that the above agricultural irrigation electricity management system and method only have a recording function, but do not have a self-checking and correction function, and thus have low practicality.

[0005] In view of this, an agricultural irrigation electricity management system and method are designed to solve the above problems. Summary of the invention

[0006] To solve the problems raised in the above background technology. The present invention provides an agricultural irrigation electricity management system and method, which can not only visualize the agricultural planting area so that the staff can clearly understand the crop planting situation in the agricultural planting area, but also realize real-time and efficient management of agricultural irrigation electricity consumption, and can realize self-checking, correction and reporting, thereby improving practical performance.

[0007] Another object of the present invention is to provide an agricultural irrigation electricity management system.

[0008] To achieve the above object, the present invention provides the following technical solution: a method for managing electricity consumption for agricultural irrigation, comprising the following steps:

[0009] S1: Take images of agricultural planting areas through oblique photography, build a 3D model of the agricultural planting area based on 3D reconstruction, and perform visualization;

[0010] S2: Obtain the color histogram of the agricultural planting area based on the three-dimensional model of the agricultural planting area, set thresholds based on the color characteristics of the soil and crops, segment the soil and crops, and count the number of segmented crops;

[0011] S3: Determine the growth stage of the crops segmented from the agricultural planting area based on the constructed and trained crop growth stage determination model;

[0012] S4: Predict the irrigation electricity consumption in the agricultural planting area based on the constructed and trained long short-term memory network model according to the number of crops in the agricultural planting area, the growth stage of crops, and the relationship between each growth stage of crops and the irrigation electricity consumption;

[0013] S5: Collect the irrigation electricity consumption in the agricultural planting area in real time, and determine the rationality of the irrigation electricity consumption in the agricultural planting area based on the comparison with the predicted irrigation electricity consumption. If it is reasonable, report it; if it is unreasonable, self-check, correct and report it.

[0014] Furthermore, the specific steps of step S1 include:

[0015] S101: planning a shooting route and shooting parameters according to the agricultural planting area;

[0016] S102: The drone equipped with the oblique camera performs flight shooting according to a predetermined shooting route and shooting parameters;

[0017] S103: importing the captured images into 3D modeling software for preprocessing, identifying the same feature points in multiple images for image matching, and reconstructing point cloud data of the agricultural planting area using the successfully matched feature points;

[0018] S104: Generate a three-dimensional mesh model based on the point cloud data;

[0019] S105: Mapping the original image texture onto the three-dimensional mesh model to generate a three-dimensional model of the agricultural planting area;

[0020] S106: Export the model file and import it into the platform for visualization.

[0021] Furthermore, the specific steps of step S2 include:

[0022] S201: Rendering the three-dimensional model of the agricultural planting area into an RGB color image;

[0023] S202: Convert the RGB color image into an HSV color image;

[0024] S203: Count the number of times each color value appears in the HSV color image to obtain a color histogram;

[0025] S204: setting a threshold value according to the color characteristics of the soil and the crops, segmenting the soil and the crops based on the set threshold value, and counting the number of the segmented crops.

[0026] Furthermore, the specific steps of step S3 include:

[0027] S301: constructing an initial model for determining the growth stage of crops for learning to determine the growth stage of crops;

[0028] S302: Collect historical data, including crop growth stages;

[0029] S303: Divide the crop growth stage data into a training set and a test set with a ratio of 7:3;

[0030] S304: training and optimizing the initial model for determining the growth stage of crops based on the training set, and testing the initial model for determining the growth stage of crops based on the test set until convergence, which is the model for determining the growth stage of crops;

[0031] S305: The crop growth stage determination model determines the crop growth stage based on the crops segmented from the agricultural planting area.

[0032] Furthermore, in step S305, if the crop growth stage determination model determines that the crop growth stage is a static state, that is, a dead state, feedback is given to step S204, and the counted number of crops is reduced by one.

[0033] Furthermore, the specific steps of step S4 include:

[0034] S401: Constructing the initial model of long short-term memory network to learn the relationship between each growth stage of crops and irrigation electricity consumption;

[0035] S402: Collect historical data, including electricity consumption for irrigation at various growth stages of crops;

[0036] S403: Divide the irrigation electricity consumption data of crops at different growth stages into a training set and a test set, with a ratio of 7:3;

[0037] S404: training and optimizing the initial model of the long short-term memory network based on the training set, and testing the initial model of the long short-term memory network based on the test set until convergence, which is the long short-term memory network model;

[0038] S405: The long short-term memory network model predicts the irrigation electricity consumption in the agricultural planting area based on the number of crops, the growth stage of crops, and the relationship between each growth stage of crops and the irrigation electricity consumption.

[0039] Furthermore, in step S5, the specific steps of the self-check include:

[0040] Check whether the irrigation power consumption at the location where the crop quantity changes has changed. If not, the location is an abnormal irrigation power consumption location. Otherwise, check the irrigation power consumption and the predicted irrigation power consumption one by one until a different location of irrigation power consumption is found. The location is an abnormal irrigation power consumption location.

[0041] Furthermore, in step S5, the specific steps of correction include:

[0042] If the result of the self-detection is that the irrigation power consumption at the location where the crop quantity changes is abnormal, the switch status of the irrigation equipment at that location is corrected;

[0043] If the self-check result shows that the irrigation power consumption in other locations is abnormal, an alarm will be triggered for manual inspection and correction.

[0044] Furthermore, steps S1-S5 are repeated periodically according to the irrigation conditions of the agricultural planting area.

[0045] The agricultural irrigation electricity management system comprises:

[0046] The module for constructing a 3D model of an agricultural planting area uses oblique photography to take images of the agricultural planting area, constructs a 3D model of the agricultural planting area based on 3D reconstruction, and performs visualization;

[0047] The agricultural planting area crop quantity statistics module obtains the agricultural planting area color histogram based on the agricultural planting area three-dimensional model, sets the threshold based on the color characteristics of the soil and crops, segments the soil and crops, and counts the number of the segmented crops;

[0048] The crop growth stage determination module in the agricultural planting area determines the growth stage of the crops segmented in the agricultural planting area based on the constructed and trained crop growth stage determination model;

[0049] The irrigation electricity consumption prediction module for agricultural planting areas predicts the irrigation electricity consumption in agricultural planting areas based on the constructed and trained long short-term memory network model according to the number of crops in the agricultural planting areas, the growth stage of crops, and the relationship between each growth stage of crops and irrigation electricity consumption;

[0050] The agricultural crop irrigation electricity consumption management module collects the irrigation electricity consumption of the agricultural planting area in real time, and determines the rationality of the irrigation electricity consumption in the agricultural planting area based on comparison with the predicted irrigation electricity consumption. If it is reasonable, it will be reported; if it is unreasonable, it will be self-checked, corrected and reported.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] 1. The present invention first constructs and visualizes a three-dimensional model of an agricultural planting area through oblique photography and three-dimensional reconstruction, then segments soil and crops through the three-dimensional model, counts the number of crops, and then determines the growth stage of crops based on the segmented crops through a constructed crop growth stage determination model. Then, the irrigation power consumption of the agricultural planting area is predicted based on the number of crops, the growth stage of crops, and the relationship between each growth stage of crops and the irrigation power consumption through a constructed long short-term memory network model. Afterwards, irrigation power consumption management is achieved by comparing actual and predicted irrigation power consumption. This method can not only visualize the agricultural planting area so that the staff can clearly understand the crop planting situation in the agricultural planting area, but also achieve real-time and efficient management of agricultural irrigation power consumption.

[0053] 2. The present invention can realize self-checking, correction and reporting based on the comparison between actual and predicted irrigation electricity consumption, which can improve practical performance compared with the prior art.

[0054] 3. The present invention is fully automated and does not require human intervention, thereby improving intelligence and further improving practical performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a flow chart of the agricultural irrigation electricity management method of the present invention;

[0056] Figure 2 This is a framework diagram of the agricultural irrigation electricity management system of the present invention;

[0057] In the figure: 1. Three-dimensional model construction module of agricultural planting area; 2. Crop quantity statistics module of agricultural planting area; 3. Crop growth stage determination module of agricultural planting area; 4. Irrigation electricity consumption prediction module of agricultural planting area; 5. Crop irrigation electricity consumption management module of agricultural planting area. DETAILED DESCRIPTION

[0058] 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.

[0059] See also Figure 1-2 The present invention provides the following technical solution: a method for managing electricity consumption for agricultural irrigation, comprising the following steps:

[0060] S1: Take images of agricultural planting areas through oblique photography, build a 3D model of the agricultural planting area based on 3D reconstruction, and perform visualization;

[0061] S2: Obtain the color histogram of the agricultural planting area based on the three-dimensional model of the agricultural planting area, set thresholds based on the color characteristics of the soil and crops, segment the soil and crops, and count the number of segmented crops;

[0062] S3: Determine the growth stage of the crops segmented from the agricultural planting area based on the constructed and trained crop growth stage determination model;

[0063] S4: Predict the irrigation electricity consumption in the agricultural planting area based on the constructed and trained long short-term memory network model according to the number of crops in the agricultural planting area, the growth stage of crops, and the relationship between each growth stage of crops and the irrigation electricity consumption;

[0064] S5: Collect the irrigation electricity consumption in the agricultural planting area in real time, and determine the rationality of the irrigation electricity consumption in the agricultural planting area based on the comparison with the predicted irrigation electricity consumption. If it is reasonable, report it; if it is unreasonable, self-check, correct and report it.

[0065] Specifically, the specific steps of step S1 include:

[0066] S101: planning a shooting route and shooting parameters according to the agricultural planting area;

[0067] The shooting route and shooting parameters are planned by oblique photography professionals;

[0068] S102: The drone equipped with the oblique camera performs flight shooting according to a predetermined shooting route and shooting parameters;

[0069] S103: importing the captured images into 3D modeling software for preprocessing, identifying the same feature points in multiple images for image matching, and reconstructing point cloud data of the agricultural planting area using the successfully matched feature points;

[0070] 3D modeling software includes CloudCompare, MeshLab, and Geomagic Studio;

[0071] S104: Generate a three-dimensional mesh model based on the point cloud data;

[0072] S105: Mapping the original image texture onto the three-dimensional mesh model to generate a three-dimensional model of the agricultural planting area;

[0073] S106: Export the model file and import it into the platform for visualization.

[0074] Specifically, the specific steps of step S2 include:

[0075] S201: Rendering the three-dimensional model of the agricultural planting area into an RGB color image;

[0076] S202: Convert the RGB color image into an HSV color image;

[0077] S203: Count the number of times each color value appears in the HSV color image to obtain a color histogram;

[0078] S204: setting a threshold value according to the color characteristics of the soil and the crops, segmenting the soil and the crops based on the set threshold value, and counting the number of the segmented crops;

[0079] Binarization can be used to convert it into an image containing only black and white, where black represents soil and white represents crops, and the white area can be counted.

[0080] Specifically, the specific steps of step S3 include:

[0081] S301: constructing an initial model for determining the growth stage of crops for learning to determine the growth stage of crops;

[0082] S302: Collect historical data, including crop growth stages;

[0083] S303: Divide the crop growth stage data into a training set and a test set with a ratio of 7:3;

[0084] S304: training and optimizing the initial model for determining the growth stage of crops based on the training set, and testing the initial model for determining the growth stage of crops based on the test set until convergence, which is the model for determining the growth stage of crops;

[0085] S305: The crop growth stage determination model determines the crop growth stage based on the crops segmented from the agricultural planting area.

[0086] Specifically, in step S305, if the crop growth stage determination model determines that the crop growth stage is a static state, that is, a dead state, feedback is returned to step S204, and the counted number of crops is reduced by one.

[0087] Specifically, the specific steps of step S4 include:

[0088] S401: Constructing the initial model of long short-term memory network to learn the relationship between each growth stage of crops and irrigation electricity consumption;

[0089] S402: Collect historical data, including electricity consumption for irrigation at various growth stages of crops;

[0090] S403: Divide the irrigation electricity consumption data of crops at different growth stages into a training set and a test set, with a ratio of 7:3;

[0091] S404: training and optimizing the initial model of the long short-term memory network based on the training set, and testing the initial model of the long short-term memory network based on the test set until convergence, which is the long short-term memory network model;

[0092] S405: The long short-term memory network model predicts the irrigation electricity consumption in the agricultural planting area based on the number of crops, the growth stage of crops, and the relationship between each growth stage of crops and the irrigation electricity consumption.

[0093] Specifically, in step S5, the specific steps of self-checking include:

[0094] Check whether the irrigation power consumption at the location where the number of crops has changed has changed. If not, the location is an abnormal irrigation power consumption location. Otherwise, check the irrigation power consumption and the predicted irrigation power consumption one by one until a different location of irrigation power consumption is found. The location is an abnormal irrigation power consumption location.

[0095] The irrigation power consumption of the agricultural planting area predicted by the long short-term memory network model includes the irrigation power consumption of each crop in the agricultural planting area and the total irrigation power consumption. During self-checking, it is first checked whether the irrigation power consumption at the location where the crop quantity changes is zero. If it is not zero, then the location is an abnormal irrigation power consumption location. Otherwise, the irrigation power consumption of each crop is compared until the abnormal irrigation power consumption of a crop is detected and stopped.

[0096] Specifically, in step S5, the specific steps of correction include:

[0097] If the result of the self-detection is that the irrigation power consumption at the location where the crop quantity changes is abnormal, the switch status of the irrigation equipment at that location is corrected;

[0098] If the self-check result shows that the irrigation power consumption at other locations is abnormal, an alarm will be issued for manual inspection and correction;

[0099] The alarms include sound and light alarms and position alarms. The position alarms are displayed on the platform, including longitude and latitude, so that people can quickly check and correct them.

[0100] Specifically, steps S1-S5 are repeated periodically according to the irrigation conditions of the agricultural planting area.

[0101] An agricultural irrigation electricity management system, comprising:

[0102] The agricultural planting area three-dimensional model construction module 1 takes images of the agricultural planting area through oblique photography, constructs a three-dimensional model of the agricultural planting area based on three-dimensional reconstruction, and performs visualization;

[0103] Agricultural planting area crop quantity statistics module 2, based on the agricultural planting area three-dimensional model to obtain the agricultural planting area color histogram, based on the color characteristics of the soil and crops to set the threshold, segment the soil and crops, and count the number of the segmented crops;

[0104] The agricultural planting area crop growth stage determination module 3 determines the growth stage of the crops segmented from the agricultural planting area based on the constructed and trained crop growth stage determination model;

[0105] The irrigation electricity consumption prediction module 4 of the agricultural planting area predicts the irrigation electricity consumption of the agricultural planting area based on the constructed and trained long short-term memory network model according to the number of crops in the agricultural planting area, the growth stage of crops, and the relationship between each growth stage of crops and the irrigation electricity consumption;

[0106] The agricultural planting area crop irrigation electricity consumption management module 5 collects the irrigation electricity consumption of the agricultural planting area in real time, and determines the rationality of the irrigation electricity consumption of the agricultural planting area based on comparison with the predicted irrigation electricity consumption. If it is reasonable, it will be reported; if it is unreasonable, it will be self-checked, corrected and reported.

[0107] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for managing electricity consumption in agricultural irrigation, characterized in that: The following steps are involved: S1: Take images of agricultural planting areas through oblique photography, build a 3D model of the agricultural planting area based on 3D reconstruction, and perform visualization; S2: Obtain the color histogram of the agricultural planting area based on the three-dimensional model of the agricultural planting area, set thresholds based on the color characteristics of the soil and crops, segment the soil and crops, and count the number of segmented crops; S3: Determine the growth stage of the crops segmented from the agricultural planting area based on the constructed and trained crop growth stage determination model; S4: Predict the irrigation electricity consumption in the agricultural planting area based on the constructed and trained long short-term memory network model according to the number of crops in the agricultural planting area, the growth stage of crops, and the relationship between each growth stage of crops and the irrigation electricity consumption; S5: Collect the irrigation electricity consumption in the agricultural planting area in real time, and determine the rationality of the irrigation electricity consumption in the agricultural planting area based on the comparison with the predicted irrigation electricity consumption. If it is reasonable, report it; if it is unreasonable, self-check, correct and report it.

2. The agricultural irrigation electricity management method according to claim 1, characterized in that: The specific steps of step S1 include: S101: planning a shooting route and shooting parameters according to the agricultural planting area; S102: The drone equipped with the oblique camera performs flight shooting according to a predetermined shooting route and shooting parameters; S103: importing the captured images into 3D modeling software for preprocessing, identifying the same feature points in multiple images for image matching, and reconstructing point cloud data of the agricultural planting area using the successfully matched feature points; S104: Generate a three-dimensional mesh model based on the point cloud data; S105: Mapping the original image texture onto the three-dimensional mesh model to generate a three-dimensional model of the agricultural planting area; S106: Export the model file and import it into the platform for visualization.

3. The agricultural irrigation electricity management method according to claim 1, characterized in that: The specific steps of step S2 include: S201: Rendering the three-dimensional model of the agricultural planting area into an RGB color image; S202: Convert the RGB color image into an HSV color image; S203: Count the number of times each color value appears in the HSV color image to obtain a color histogram; S204: setting a threshold value according to the color characteristics of the soil and the crops, segmenting the soil and the crops based on the set threshold value, and counting the number of the segmented crops.

4. The agricultural irrigation electricity management method according to claim 3, characterized in that: The specific steps of step S3 include: S301: constructing an initial model for determining the growth stage of crops for learning to determine the growth stage of crops; S302: Collect historical data, including crop growth stages; S303: Divide the crop growth stage data into a training set and a test set with a ratio of 7:3; S304: training and optimizing the initial model for determining the growth stage of crops based on the training set, and testing the initial model for determining the growth stage of crops based on the test set until convergence, which is the model for determining the growth stage of crops; S305: The crop growth stage determination model determines the crop growth stage based on the crops segmented from the agricultural planting area.

5. The agricultural irrigation electricity management method according to claim 4, characterized in that: In step S305, if the crop growth stage determination model determines that the crop growth stage is a static state, that is, a dead state, feedback is returned to step S204, and the counted number of crops is reduced by one.

6. The agricultural irrigation electricity management method according to claim 1, characterized in that: The specific steps of step S4 include: S401: Constructing the initial model of long short-term memory network to learn the relationship between each growth stage of crops and irrigation electricity consumption; S402: Collect historical data, including electricity consumption for irrigation at various growth stages of crops; S403: Divide the irrigation electricity consumption data of crops at different growth stages into a training set and a test set, with a ratio of 7:3; S404: training and optimizing the initial model of the long short-term memory network based on the training set, and testing the initial model of the long short-term memory network based on the test set until convergence, which is the long short-term memory network model; S405: The long short-term memory network model predicts the irrigation electricity consumption in the agricultural planting area based on the number of crops, the growth stage of crops, and the relationship between each growth stage of crops and the irrigation electricity consumption.

7. The agricultural irrigation electricity management method according to claim 1, characterized in that: In step S5, the specific steps of self-checking include: Check whether the irrigation power consumption at the location where the crop quantity changes has changed. If not, the location is an abnormal irrigation power consumption location. Otherwise, check the irrigation power consumption and the predicted irrigation power consumption one by one until a different location of irrigation power consumption is found. The location is an abnormal irrigation power consumption location.

8. The agricultural irrigation electricity management method according to claim 1, characterized in that: In step S5, the specific steps of correction include: If the result of the self-detection is that the irrigation power consumption at the location where the crop quantity changes is abnormal, the switch status of the irrigation equipment at that location is corrected; If the self-check result shows that the irrigation power consumption in other locations is abnormal, an alarm will be triggered for manual inspection and correction.

9. The agricultural irrigation electricity management method according to claim 1, characterized in that: The steps S1-S5 are repeated periodically according to the irrigation conditions of the agricultural planting area.

10. An agricultural irrigation electricity management system according to claims 1-9, characterized in that: include: A module (1) for constructing a three-dimensional model of an agricultural planting area, which takes images of the agricultural planting area by oblique photography, constructs a three-dimensional model of the agricultural planting area based on three-dimensional reconstruction, and performs visualization; The agricultural planting area crop quantity counting module (2) obtains the agricultural planting area color histogram based on the agricultural planting area three-dimensional model, sets a threshold based on the color characteristics of the soil and the crops, segments the soil and the crops, and counts the number of the segmented crops; The agricultural planting area crop growth stage determination module (3) determines the growth stage of the crops segmented from the agricultural planting area based on the constructed and trained crop growth stage determination model; An agricultural planting area irrigation electricity consumption prediction module (4) predicts the irrigation electricity consumption of the agricultural planting area based on the constructed and trained long short-term memory network model according to the number of crops in the agricultural planting area, the growth stage of the crops, and the relationship between each growth stage of the crops and the irrigation electricity consumption; The agricultural crop irrigation power consumption management module (5) collects the irrigation power consumption of the agricultural planting area in real time, and determines the rationality of the irrigation power consumption of the agricultural planting area based on comparison with the predicted irrigation power consumption. If it is reasonable, it is reported; if it is unreasonable, it is self-checked, corrected and reported.

Images