Farmland water level environment monitoring and alarming device

Abstract

The invention relates to the technical field of water level, and particularly discloses a farmland water level environment monitoring and alarming device, which comprises a water level acquisition module for acquiring a water level value in real time, comparing a water level range value, sending an early warning prompt when the range is exceeded, and an early warning analysis module for obtaining a water level early warning value through numerical calculation based on the water level value, an alarm signal is sent if the water level early-warning value exceeds a threshold value, the water level early-warning value needs to be predicted and analyzed if the water level early-warning value does not exceed the threshold value, a trend prediction module carries out linear regression on the multi-cycle water level early-warning value, water level over-limit time is obtained through calculation, an AI analysis module combines historical water level values, the early-warning value, rainfall and farmland displacement, and the water level early-warning value is obtained. The device can comprehensively monitor the water level of the farmland, early warns and predicts the change of the water level in time, and provides an effective decision basis for the management of the water level of the farmland.

Description

Technical Field

[0001] The invention relates to the technical field of water level monitoring, and in particular to a farmland water level environment monitoring alarm device. Background Art

[0002] In modern agricultural production, the rational control of farmland water levels has a vital impact on the growth, development and yield quality of crops. Therefore, monitoring and effectively managing farmland water levels has become a key task in the agricultural field.

[0003] In the existing technology, although automated monitoring equipment can automatically obtain water level data, its capabilities in data processing and analysis are relatively weak. For example, it can only simply record water level values ​​and lacks the function of in-depth mining and comprehensive analysis of water level data. If the water level value can be automatically obtained in real time and the data is pre-processed, analyzed and calculated through algorithms, it can not only accurately determine whether the water level is abnormal, but also issue early warning signals in a timely manner.

[0004] The existing technology is lacking in the prediction of water level change trends, early warning of water level abnormalities, and research on the relationship between water level and rainfall, drainage, etc. As a result, in practical applications, farmers or agricultural managers can only obtain limited water level information and cannot fully understand the dynamic changes of farmland water levels. If the water level change trend is used for prediction and the water level drop time is predicted in combination with rainfall and drainage, this will help farmers and managers take effective countermeasures in advance, regulate farmland water levels, and optimize farmland water management.

[0005] To this end, the present invention provides a farmland water level environment monitoring alarm device. Summary of the invention

[0006] The purpose of the present invention is to provide a farmland water level environment monitoring alarm device to solve the above-mentioned background problems.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] In a first aspect, a farmland water level environment monitoring alarm device comprises:

[0009] Water level acquisition module: obtains the water level value within the monitoring period in real time, compares and analyzes the water level value with the water level range value, and sends a water level warning prompt to the environmental monitoring alarm device if the water level value exceeds the water level range value; if the water level value is within the water level range value, generates a water level analysis signal;

[0010] Early warning analysis module: Based on the water level analysis signal, the water level value and the water level range value within the monitoring period are numerically calculated to obtain the water level early warning value Yj. Based on the water level early warning value Yj, it is determined whether a water level alarm signal needs to be generated, or whether a predictive analysis needs to be performed on the water level early warning value Yj;

[0011] Trend prediction module: if it is necessary to make a prediction analysis on the water level warning value Yj, obtain the water level warning value Yj of multiple monitoring cycles, analyze and calculate the trend of the water level warning value Yj, obtain the water level exceeding limit time under the worst case and linear trend, and send the water level exceeding limit time to the environmental monitoring alarm device;

[0012] AI analysis module: Combine historical water level values, historical water level warning values, rainfall, and farmland drainage as data to build an AI prediction model to predict and analyze the water level and obtain the time when the water level will fall.

[0013] In a second aspect, a farmland water level environment monitoring alarm method comprises:

[0014] Step 1: Real-time acquisition of the water level value within the monitoring period, comparison and analysis of the water level value and the water level range value, if the water level value exceeds the water level range value, send a water level warning prompt to the environmental monitoring alarm device, if the water level value is within the water level range value, generate a water level analysis signal;

[0015] Step 2: Based on the water level analysis signal, the water level value and the water level range value within the monitoring period are numerically calculated to obtain the water level warning value Yj. Based on the water level warning value Yj, it is determined whether a water level alarm signal needs to be generated, or whether a predictive analysis needs to be performed on the water level warning value Yj;

[0016] Step 3: If it is necessary to make a forecast analysis on the water level warning value Yj, obtain the water level warning value Yj of multiple monitoring cycles, analyze and calculate the trend of the water level warning value Yj, obtain the water level exceeding limit time under the worst case and the linear trend, and send the water level exceeding limit time to the environmental monitoring alarm device;

[0017] Step 4: Combine historical water level values, historical water level warning values, rainfall, and farmland drainage as data, use AI to predict and analyze water levels, and obtain the time when the water level will fall.

[0018] Beneficial effects of the present invention:

[0019] (1) By obtaining the water level value in real time and comparing and analyzing it with the precisely set water level range value in a timely manner, whether it is the water level warning prompt sent when the water level exceeds the maximum endpoint value, or the water level analysis signal generated when the water level is within the risk range, it provides a key basis for subsequent processing. This enables farmers or managers to grasp the water level status of farmland in a timely manner, effectively reducing the adverse effects on crops caused by excessively high or low water levels, such as waterlogging causing root hypoxia and rot or drought affecting the normal growth and development of crops, thereby ensuring the healthy growth of crops.

[0020] (2) Through the coordinated work of the early warning analysis module, trend prediction module and AI analysis module, a comprehensive analysis of water level data is achieved. By calculating the water level warning value, analyzing the linear relationship between water level and time, predicting the time when the water level exceeds the limit, and predicting the time when the water level falls in combination with environmental factors, more decision-making support is provided for farmland water level management, which helps to cope with sudden water level changes and enhance the risk resistance of agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The present invention will be further described below in conjunction with the accompanying drawings.

[0022] Figure 1 It is a flow chart of water level data and analysis in a farmland water level environment monitoring alarm device of the present invention;

[0023] Figure 2 It is a flow chart of a method for predicting a water level trend in a farmland water level environment monitoring alarm device in the present invention;

[0024] Figure 3 It is a module diagram of a farmland water level environment monitoring alarm device in the present invention. DETAILED DESCRIPTION

[0025] 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.

[0026] Example 1

[0027] See also Figure 1 As shown, the present invention is a farmland water level environment monitoring alarm device, comprising:

[0028] Water level acquisition module: obtains the water level value within the monitoring period in real time, compares and analyzes the water level value with the water level range value, and sends a water level warning prompt to the environmental monitoring alarm device if the water level value exceeds the water level range value; if the water level value is within the water level range value, generates a water level analysis signal;

[0029] In some embodiments, a pressure-type water level sensor is used to obtain water level values ​​in a monitoring period in real time, and the obtained water level values ​​are subjected to data preprocessing to remove abnormal water level values;

[0030] For example, if the sensor records a data value with a large difference between the previous and next water level values ​​in a short period of time, for example, a water level value of 2m suddenly appears in a farmland with a normal water level in the range of 0.25m-0.45m, then the water level value of 2m is an abnormal value; a statistical method is used to identify abnormal values, and the 3 times standard deviation (3σ) criterion is used to calculate the mean and standard deviation of multiple water level values ​​in the monitoring period, and the data of water level values ​​exceeding the mean ± 3 times the standard deviation are eliminated;

[0031] It should be noted that the pressure sensor is based on the relationship between liquid pressure and depth, through the formula: The water level value h is obtained, where ρ is the density of water, g is the acceleration of gravity, and P is the pressure of water in the farmland. The pressure-type water level sensor is installed at the bottom of the water. When the water level rises, the pressure exerted by the water on the sensor probe gradually increases. The piezoresistors inside the pressure-type water level sensor will change their resistance value due to the change in pressure, and then the pressure can be measured.

[0032] Compare the water level value with the water level range value;

[0033] If the water level value is higher than the maximum endpoint value of the water level range value, it indicates that the water level value of the farmland exceeds the expectation, and a water level warning prompt is sent to the environmental monitoring alarm device;

[0034] If the water level value is higher than the minimum endpoint value of the water level range value and lower than the maximum endpoint value of the water level range value, although the water level value is within the expected range, there is a risk in the water level value, and the water level value needs to be further analyzed to generate a water level analysis signal;

[0035] If the water level value is lower than the minimum endpoint value of the water level range, no processing is performed;

[0036] It should be noted that the water level range value is set by professionals in this field based on experience;

[0037] Early warning analysis module: Based on the water level analysis signal, the water level value and the water level range value within the monitoring period are numerically calculated to obtain the water level early warning value Yj. Based on the water level early warning value Yj, it is determined whether it is necessary to perform a forecast analysis on the water level early warning value Yj;

[0038] The water level value of the farmland at each monitoring moment in the monitoring period is processed with the maximum endpoint value of the water level range to obtain the water level difference;

[0039] It should be noted that there are multiple monitoring moments in the monitoring cycle, and the water level value corresponds to the time of each monitoring moment one by one;

[0040] The water level difference is processed by ratio with the length of the water level range value to obtain the water level proximity ratio;

[0041] Obtain the water level proximity ratio at multiple monitoring moments within the monitoring period;

[0042] The water level proximity ratios in multiple monitoring cycles are taken as data elements and put into a water level data group;

[0043] Calculate the mean and variance of the data elements in the water level data set, mark the variance of the water level proximity ratio in the water level data set as Sf, and mark the mean of the water level proximity ratio in the water level data set as Sj;

[0044] Based on the variance Sf of the water level approach ratio and the mean Sj of the water level approach ratio, the water level warning value Yj is calculated;

[0045] By formula: Get the water level warning value Yj, where a is 0.787 and b is 0.213;

[0046] Compare the water level warning value Yj with the water level warning threshold;

[0047] If the water level warning value Yj ≥ the water level warning threshold, it indicates that the water level warning value of the farmland in the current monitoring period exceeds the expected range, and the water level of the farmland is in a high-risk state. There is no need to make a forecast analysis of the water level warning value Yj, and it is necessary to send a water level alarm signal to the environmental monitoring alarm device;

[0048] If the water level warning value Yj is less than the water level warning threshold, it indicates that the water level warning value of the farmland in the current monitoring period is within the expected range, but the water level of the farmland is at great risk. It is not necessary to send a water level alarm signal to the environmental monitoring alarm device, and it is necessary to make a forecast analysis of the water level warning value Yj of the farmland;

[0049] Exemplary: if the water level range value is [0.25m, 0.45m], a single monitoring cycle is 3 days, monitoring is performed 3 times a day, and the water level values ​​obtained are 0.28, 0.32, 0.35, 0.38, 0.40, 0.42, 0.43, 0.41, 0.36, the water level data group is [0.85, 0.65, 0.5, 0.35, 0.25, 0.15, 0.1, 0.2, 0.45], the mean and variance of the water level proximity ratio are 0.35 and 0.06 respectively, and the water level warning value Yj is calculated to be 0.34. If the water level warning threshold is 0.3, it indicates that the water level warning value Yj of the current paddy field is ≥ the water level warning threshold, and a water level alarm signal is sent to the environmental monitoring alarm device;

[0050] The technical solution of this embodiment is: real-time acquisition of the water level value within the monitoring period, comparison and analysis of the water level value and the water level range value, if the water level value exceeds the water level range value, sending a water level warning prompt to the environmental monitoring alarm device, if the water level value is within the water level range value, generating a water level analysis signal, based on the water level analysis signal, the water level value within the monitoring period and the water level range value are numerically calculated to obtain a water level warning value Yj, based on the water level warning value Yj, determine whether it is necessary to perform a prediction analysis on the water level warning value Yj;

[0051] Example 2

[0052] Based on Example 1, a farmland water level environment monitoring alarm device further includes:

[0053] Trend prediction module: if it is necessary to make a prediction analysis on the water level warning value Yj, obtain the water level warning value Yj of multiple monitoring cycles, analyze and calculate the trend of the water level warning value Yj, obtain the water level exceeding limit time under the worst case and linear trend, and send the water level exceeding limit time to the environmental monitoring alarm device;

[0054] like Figure 2 As shown in Figure 2, the process of water level trend prediction is as follows:

[0055] If it is necessary to make a forecast analysis on the water level warning value Yj, obtain the water level warning values ​​Yj of multiple monitoring periods and the time of the corresponding monitoring periods, and draw a time-water level warning value change curve in a two-dimensional rectangular coordinate system;

[0056] It should be noted that the water level warning value Yj is used as the ordinate and the monitoring period time is used as the abscissa to draw a time-water level warning value change curve. By analyzing the data in the change curve, it is determined whether the time and the water level warning value Yj show a linear correlation trend;

[0057] By calculating the Pearson correlation coefficient r, the trend of the water level warning value Yj is analyzed and judged;

[0058] Specifically, through the formula: Get the Pearson correlation coefficient r, where x i is the time series corresponding to the monitoring period, i = 1, 2, ..., n, y i is the data series of the water level warning value corresponding to the water level warning value Yj, i = 1, 2, ..., n, They are the average value of the time series, the average value of the water level warning value data series, and the number of data n;

[0059] It should be noted that the value range of r is [-1,1]; r = 1 indicates a complete positive linear correlation, r = -1 indicates a complete negative linear correlation, and r = 0 indicates no linear correlation. For example, r = 0.8 indicates that there is a positive linear correlation between the water level warning value Yj and time.

[0060] If the Pearson correlation coefficient r of the time-water level warning value Yj is in the range of [-1,0), it indicates that time and the water level warning value Yj show a negative linear correlation trend, that is, as time gradually moves backward, the water level warning value Yj shows a downward trend, then the water level warning value Yj gradually decreases, and the water level risk of the farmland gradually decreases. When the water level value is lower than the minimum endpoint value of the water level range value, the water level alarm signal is released;

[0061] If the Pearson correlation coefficient r of the time-water level warning value Yj is in the range of (0,1], it indicates that time and the water level warning value Yj show a positive linear correlation trend, that is, as time gradually moves backward, the water level warning value Yj also shows an upward trend. The least squares method is used to fit the time-water level warning value;

[0062] If the Pearson correlation coefficient r of the time-water level warning value Yj is 0, it means that the time and the water level warning value do not show a linear trend. The maximum slope value in the time-water level warning value change curve is obtained, and the maximum slope value is combined with the water level warning value Yj for calculation to obtain the water level exceeding limit time in the worst case.

[0063] If time and water level warning value Yj show a positive linear correlation trend, the time-water level warning value is fitted using the least squares method based on the linear regression equation: Y=kX+b, where Y is the water level warning value, x is time, k is the slope of the fitting line, and b is the intercept of the fitting line on the Y axis;

[0064] In the two-dimensional rectangular coordinate system, draw a fitting straight line, select the water level warning threshold as the reference point, and draw a straight line parallel to the X-axis as the warning reference line;

[0065] Obtain the horizontal coordinate of the intersection of the fitting straight line and the warning reference line to obtain the water level exceeding limit time under the linear trend;

[0066] Send the water level exceedance time under the worst case and linear trend to the environmental monitoring alarm device;

[0067] Example 3

[0068] A farmland water level environment monitoring alarm device, further comprising:

[0069] AI analysis module: Based on the water level alarm signal, combined with historical water level values, historical water level warning values, rainfall, and farmland drainage as data, AI is used to predict and analyze the water level to obtain the time when the water level will fall;

[0070] Based on the water level alarm signal, historical water level values ​​and historical water level warning values ​​are obtained, combined with rainfall and farmland drainage as water level data sets;

[0071] It should be noted that the drainage volume of the farmland was measured by an electromagnetic flow sensor;

[0072] Based on the water level dataset, an AI prediction model was built using a long short-term memory network (LSTM);

[0073] The specific AI prediction model construction method is:

[0074] S1. Preprocess the water level dataset, including removing outliers and missing data, converting data of different magnitudes to the same scale, such as normalizing rainfall data to a value range between 0 and 1; align data time to ensure that all data are arranged at the same time interval and time axis for subsequent analysis and model training;

[0075] S2. Use the long short-term memory network (LSTM) method to build an AI prediction model. The input layer of the model receives environmental factor data, such as rainfall, historical water level values, historical water level warning values, and drainage. The output layer of the model outputs the predicted water level drop time.

[0076] S3. Divide the preprocessed data set into a training set and a test set in a ratio of 7:3, use the training set to train the constructed AI prediction model, set the number of training rounds of the model to 100, and the batch size to 32. After each round of training, use the validation set data to evaluate the model;

[0077] It should be noted that the batch size refers to the number of data samples used to update the model weights each time when training the model. For example, if a training data set contains 1,000 samples and the batch size is set to 100, the data will be divided into 10 batches during the training process, and 100 samples will be used for each update when training the model;

[0078] Based on the AI ​​prediction model, the water level drop time is output and sent to the environmental monitoring alarm device;

[0079] Example 4

[0080] A farmland water level environment monitoring alarm method, comprising:

[0081] Step 1: Real-time acquisition of the water level value within the monitoring period, comparison and analysis of the water level value and the water level range value, if the water level value exceeds the water level range value, send a water level warning prompt to the environmental monitoring alarm device, if the water level value is within the water level range value, generate a water level analysis signal;

[0082] Step 2: Based on the water level analysis signal, the water level value and the water level range value within the monitoring period are numerically calculated to obtain the water level warning value Yj. Based on the water level warning value Yj, it is determined whether a water level alarm signal needs to be generated, or whether a predictive analysis needs to be performed on the water level warning value Yj;

[0083] Step 3: If it is necessary to make a forecast analysis on the water level warning value Yj, obtain the water level warning value Yj of multiple monitoring cycles, analyze and calculate the trend of the water level warning value Yj, obtain the water level exceeding limit time under the worst case and the linear trend, and send the water level exceeding limit time to the environmental monitoring alarm device;

[0084] Step 4: Combine historical water level values, historical water level warning values, rainfall, and farmland drainage as data, use AI to predict and analyze water levels, and obtain the time when the water level will fall.

[0085] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. A farmland water level environment monitoring alarm device, characterized in that: include: Water level acquisition module: obtains the water level value within the monitoring period in real time, compares and analyzes the water level value with the water level range value, and sends a water level warning prompt to the environmental monitoring alarm device if the water level value exceeds the water level range value; if the water level value is within the water level range value, generates a water level analysis signal; Early warning analysis module: Based on the water level analysis signal, the water level value and the water level range value within the monitoring period are numerically calculated to obtain the water level early warning value Yj. Based on the water level early warning value Yj, it is determined whether a water level alarm signal needs to be generated, or whether a predictive analysis needs to be performed on the water level early warning value Yj; Trend prediction module: if it is necessary to make a prediction analysis on the water level warning value Yj, obtain the water level warning value Yj of multiple monitoring cycles, analyze and calculate the trend of the water level warning value Yj, obtain the water level exceeding limit time under the worst case and linear trend, and send the water level exceeding limit time to the environmental monitoring alarm device; AI analysis module: Combine historical water level values, historical water level warning values, rainfall, and farmland drainage as data to build an AI prediction model to predict and analyze the water level and obtain the time when the water level will fall.

2. The farmland water level environment monitoring alarm device according to claim 1, characterized in that: The steps to determine whether a water level alarm signal needs to be generated are as follows: If the water level warning value Yj ≥ the water level warning threshold, there is no need to make a forecast analysis on the water level warning value Yj, but it is necessary to send a water level alarm signal to the environmental monitoring alarm device; The steps to determine whether it is necessary to make a forecast analysis of the water level warning value Yj are as follows: If the water level warning value Yj is less than the water level warning threshold, there is no need to send a water level warning signal to the environmental monitoring alarm device, but a predictive analysis of the water level warning value Yj of the farmland is required.

3. The farmland water level environment monitoring alarm device according to claim 2 is characterized in that: The water level warning value Yj is obtained as follows: Based on the variance Sf of the water level approach ratio and the mean Sj of the water level approach ratio, the water level warning value Yj is calculated; By formula: Get the water level warning value Yj, where a is 0.787 and b is 0.

213.

4. The farmland water level environment monitoring alarm device according to claim 3 is characterized by: The variance Sf of the water level approach ratio is obtained as follows: The water level value of the farmland at each monitoring moment in the monitoring period is processed with the maximum endpoint value of the water level range to obtain the water level difference; The water level difference is processed by ratio with the length of the water level range value to obtain the water level proximity ratio; Obtain the water level proximity ratio at multiple monitoring moments within the monitoring period; The water level proximity ratios in multiple monitoring cycles are taken as data elements and put into a water level data group; The variance of the data elements in the water level data set is calculated, and the variance of the water level proximity ratio in the water level data set is marked as Sf.

5. The farmland water level environment monitoring alarm device according to claim 3 is characterized by: The mean value Sj of the water level approach ratio is obtained as follows: The mean of the data elements in the water level data set is calculated, and the mean of the water level proximity ratios in the water level data set is marked as Sj.

6. The farmland water level environment monitoring alarm device according to claim 1, characterized in that: The method for obtaining the water level exceeding limit time in the worst case is: If it is necessary to make a forecast analysis on the water level warning value Yj, obtain the water level warning values ​​Yj of multiple monitoring periods and the time of the corresponding monitoring periods, and draw a time-water level warning value change curve in a two-dimensional rectangular coordinate system; By calculating the Pearson correlation coefficient r, the trend of the water level warning value Yj is analyzed and judged; If the Pearson correlation coefficient r of the time-water level warning value Yj is 0, the maximum slope value in the change curve of the time-water level warning value is obtained, and the maximum slope value is combined with the water level warning value Yj for calculation to obtain the water level exceeding time in the worst case.

7. The farmland water level environment monitoring alarm device according to claim 6, characterized in that: The Pearson correlation coefficient is obtained as follows: By formula: Get the Pearson correlation coefficient r, where x i is the time series corresponding to the monitoring period, i = 1, 2, ..., n, y i is the data series of the water level warning value corresponding to the water level warning value Yj, i = 1, 2, ..., n, They are the average value of the time series, the average value of the data series of water level warning values, and n is the number of data.

8. The farmland water level environment monitoring alarm device according to claim 1, characterized in that: The method for obtaining the water level exceeding limit time under the linear trend is: If the time and water level warning value Yj show a positive linear correlation trend, the time-water level warning value is fitted using the least squares method based on the linear regression equation: Y=kX+b; In the two-dimensional rectangular coordinate system, draw a fitting straight line, select the water level warning threshold as the reference point, and draw a straight line parallel to the X-axis as the warning reference line; The horizontal coordinate of the intersection of the fitting straight line and the warning reference line is obtained to obtain the water level exceeding limit time under the linear trend.

9. The farmland water level environment monitoring alarm device according to claim 8, characterized in that: The judgment method of the trend of the time and the water level warning value Yj showing a positive linear correlation is: If the Pearson correlation coefficient r of the time-water level warning value Yj is in the range of (0,1], it indicates that the time and the water level warning value Yj show a positive linear correlation trend, and the least squares method is used to fit the time-water level warning value.

10. The farmland water level environment monitoring alarm device according to claim 1, characterized in that: The water level drop time is obtained as follows: Based on the water level alarm signal, historical water level values ​​and historical water level warning values ​​are obtained, combined with rainfall and farmland drainage as water level data sets; Based on the water level dataset, an AI prediction model was built using a long short-term memory network; The specific AI prediction model construction method is: S1. Preprocess the water level data set, including removing outliers and missing data, converting data of different magnitudes to the same scale, such as normalizing rainfall data to make its value range between 0 and 1; align data time to ensure that all data are arranged according to the same time interval and time axis; S2. Use the long short-term memory network method to build an AI prediction model. The input layer of the model receives environmental factor data, such as rainfall, historical water level values, historical water level warning values, and drainage. The output layer of the model outputs the predicted water level drop time. S3. Divide the preprocessed data set into a training set and a test set in a ratio of 7:3, use the training set to train the constructed AI prediction model, set the number of training rounds of the model to 100, and the batch size to 32. After each round of training, use the validation set data to evaluate the model; Based on the AI ​​prediction model, the water level drop time is output and sent to the environmental monitoring alarm device.

Images