Self-adaptive water control method for growth cycle of palm vine seedlings

Abstract

The present application relates to water-saving irrigation technical field, specifically to a palm rattan seedling growth cycle self-adaptive water control method, including the following steps: obtaining the water content scalar and water potential scalar of historical dewing stage and historical hygroscopic stage output by soil moisture sensor, respectively carrying out curve fitting operation, generating dewing characteristic curve and hygroscopic characteristic curve; respectively carrying out second order derivation operation to the dewing characteristic curve and the hygroscopic characteristic curve, extracting the water content scalar corresponding to the inflection point of zero second order derivative. The present application can change the water supply intensity with the degree of water deficit; in the approaching upper limit interval, the intermittent pulse signal is sent with water deficit scalar as independent variable, and the machine is stopped synchronously when the closed instruction is triggered, which can reduce the excessive water supply, deep seepage and invalid operation of water pump when approaching the target water content state, thereby improving the water environment stability in palm rattan seedling growth cycle.

Description

Technical Field

[0001] This invention relates to the field of water-saving irrigation technology, and in particular to an adaptive water control method for the growth cycle of palm vine seedlings. Background Technology

[0002] Water-saving irrigation technology mainly involves the precise supply, dynamic allocation, and efficient utilization of water resources in forestry planting.

[0003] Current water-saving irrigation systems typically focus on the efficiency of water supply, allocation, and utilization. However, they often rely on fixed moisture content thresholds or empirically determined thresholds for setting specific control boundaries. Fixed thresholds fail to reflect the difference in soil water potential between the desiccation and hygroscopic stages, leading to premature irrigation start-up or delayed shut-off. When soil moisture content approaches the threshold, even slight fluctuations in sensor readings can trigger frequent valve switching, increasing solenoid valve wear, pump start-ups and shutdowns, and instability in the water supply process. Furthermore, in assessing water deficit levels, simply supplying water based on whether it falls below the target value results in treating extreme and moderate water shortages the same, potentially leading to excessive water output during mild shortages or insufficient water replenishment during severe shortages. Therefore, improvements are needed. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology and propose an adaptive water control method for the growth cycle of palm vine seedlings.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: an adaptive water control method for the growth cycle of palm vine seedlings, comprising the following steps:

[0006] The soil moisture sensor outputs historical desiccation and hygroscopic phases, along with their corresponding water potential scalars. Curve fitting is then performed on these scalars to generate desiccation and hygroscopic characteristic curves. Second-order derivatives are calculated on both curves to extract the water content scalars corresponding to inflection points where the second derivative is zero. The water content scalars extracted from the desiccation curves are defined as the lower limit scalar for valve opening, and the water content scalars extracted from the hygroscopic curves are defined as the upper limit scalar for valve closing. These two scalars are then combined to construct a hysteresis dual-threshold parameter.

[0007] Obtain the current moisture content scalar, compare the current moisture content scalar with the valve opening lower limit scalar and the valve closing upper limit scalar within the hysteresis dual threshold parameters, and generate a toggle switch command; determine the polarity state of the toggle switch command and map it to the solenoid valve action state to generate a discrete valve control state;

[0008] When the discrete control state of the valve is an open command, the difference between the upper limit scalar of the valve closing and the current moisture content scalar is calculated to establish a moisture deficit scalar; the moisture deficit scalar is compared with the preset first deficit threshold and second deficit threshold respectively to determine the interval to which the current moisture content scalar belongs.

[0009] Preferably, the method further includes:

[0010] Based on the interval assignment, when it is determined that the interval is approaching the upper limit, an intermittent pulse signal is sent to the variable frequency water pump with the water deficit scalar as the independent variable; when the closing command is triggered, the intermittent pulse signal is terminated, and a stop command is simultaneously sent to the variable frequency water pump to obtain an adaptive flow regulation flow.

[0011] Preferably, the steps for obtaining the desiccation characteristic curve and the moisture absorption characteristic curve are as follows:

[0012] The system acquires historical desiccation stage moisture content scalars, historical desiccation stage water potential scalars, historical hygroscopic stage moisture content scalars, and historical hygroscopic stage water potential scalars from the soil moisture sensor. Curve fitting is performed according to the mapping relationship between the historical desiccation stage moisture content scalars and the historical desiccation stage water potential scalars. Curve fitting is also performed according to the mapping relationship between the historical hygroscopic stage moisture content scalars and the historical hygroscopic stage water potential scalars. Abnormal mapping points with repeated moisture content scalars and discontinuous water potential scalar jumps are removed, and fitting segments with continuous mapping directions are retained to generate desiccation characteristic curves and hygroscopic characteristic curves.

[0013] Preferably, the step of obtaining the hysteresis dual threshold parameter is as follows:

[0014] Extract the curvature change corresponding to each moisture content scalar in the dehumidification characteristic curve, and extract the curvature change corresponding to each moisture content scalar in the moisture absorption characteristic curve. Confirm the inflection point where the curvature change changes from positive to negative or from negative to positive by the second derivative being zero. Read the moisture content scalar corresponding to the inflection point of the dehumidification characteristic curve and the moisture content scalar corresponding to the inflection point of the moisture absorption characteristic curve, and generate the lower limit scalar for opening the valve and the upper limit scalar for closing the valve.

[0015] Determine whether the lower limit scalar of valve opening is less than the upper limit scalar of valve closing. If the lower limit scalar of valve opening is less than the upper limit scalar of valve closing, then construct the parameters by combining them in the order of the lower limit scalar of valve opening first and the upper limit scalar of valve closing last. If the lower limit scalar of valve opening is greater than or equal to the upper limit scalar of valve closing, then revisit the dehumidification characteristic curve and the moisture absorption characteristic curve to reconfirm the moisture content scalar corresponding to the inflection point, until the lower limit scalar of valve opening is less than the upper limit scalar of valve closing, and generate a hysteresis double threshold parameter.

[0016] Preferably, the step of obtaining the toggle switch command is as follows:

[0017] Obtain the current moisture content scalar, retrieve the lower limit scalar for opening the valve and the upper limit scalar for closing the valve within the hysteresis dual threshold parameters, align the current moisture content scalar item by item to the threshold range of the lower limit scalar for opening the valve and the upper limit scalar for closing the valve, mark the opening flip condition where the current moisture content scalar is less than or equal to the lower limit scalar for opening the valve, mark the closing flip condition where the current moisture content scalar plus the preset infiltration compensation scalar is greater than or equal to the upper limit scalar for closing the valve, mark the holding condition where the current moisture content scalar is greater than the lower limit scalar for opening the valve and the current moisture content scalar plus the preset infiltration compensation scalar is less than the upper limit scalar for closing the valve, and generate a flip switch command.

[0018] Preferably, the step of obtaining the discrete control state of the valve is as follows:

[0019] Read the polarity states corresponding to the opening flip condition, closing flip condition, and holding condition. When the opening flip condition is met, map the flip switch command to the solenoid valve opening action. When the closing flip condition is met, map the flip switch command to the solenoid valve closing action. When the holding condition is met, read the current solenoid valve action state and retain the original polarity of the current solenoid valve action state to form the solenoid valve action state.

[0020] Verify the solenoid valve's opening action, closing action, and the reserved state of the current solenoid valve's action state. When the solenoid valve's action state corresponds to the solenoid valve's opening action, output an opening command; when the solenoid valve's action state corresponds to the solenoid valve's closing action, output a closing command; when the solenoid valve's action state corresponds to the reserved state of the current solenoid valve's action state, maintain the current solenoid valve's action state unchanged, and generate the valve's discrete control state.

[0021] Preferably, the step of obtaining the water deficit scalar is as follows:

[0022] When the discrete control state of the valve is an open command, the numerical position of the upper limit scalar of valve closure is read, the numerical position of the current moisture content scalar is read, the difference is calculated according to the numerical interval between the upper limit scalar of valve closure and the current moisture content scalar, invalid intervals less than zero in the difference calculation are eliminated, and the effective interval representing the distance between the current moisture content scalar and the upper limit scalar of valve closure is retained. The effective interval is defined as the degree of moisture deficit, and a moisture deficit scalar is established.

[0023] Preferably, the step of obtaining the interval affiliation is as follows:

[0024] Retrieve the preset first and second deficit thresholds, first check the threshold order where the first deficit threshold is greater than the second deficit threshold, then align the water deficit scalar to the numerical boundaries of the first and second deficit thresholds respectively, record the comparison status when the water deficit scalar is greater than the first deficit threshold, record the comparison status when the water deficit scalar is less than or equal to the first deficit threshold and greater than the second deficit threshold, record the comparison status when the water deficit scalar is less than or equal to the second deficit threshold, and generate the deficit threshold comparison result;

[0025] Based on the deficit threshold comparison results, the comparison status corresponding to the moisture deficit scalar is read. When the moisture deficit scalar is greater than the first deficit threshold, the current moisture content scalar is determined to be in the extreme deficit range. When the moisture deficit scalar is less than or equal to the first deficit threshold and greater than the second deficit threshold, the current moisture content scalar is determined to be in the moderate deficit range. When the moisture deficit scalar is less than or equal to the second deficit threshold, the current moisture content scalar is determined to be in the range approaching the upper limit, and the range assignment is generated.

[0026] Preferably, the step of obtaining the adaptive flow regulation stream is as follows:

[0027] Read the extreme deficit interval, moderate deficit interval, and near-upper limit interval corresponding to the water deficit scalar, map the extreme deficit interval to the upper limit output position of the continuous duty cycle in the variable frequency pump pulse width adjustment coordinate system, map the moderate deficit interval to the lower limit output position of the continuous duty cycle in the variable frequency pump pulse width adjustment coordinate system, and map the near-upper limit interval to the intermittent pulse output position in the variable frequency pump pulse width adjustment coordinate system, thereby generating a piecewise mapping relationship between the water deficit scalar and the variable frequency pump pulse width adjustment coordinate system;

[0028] Based on the piecewise mapping relationship between the water deficit scalar and the variable frequency pump pulse width adjustment coordinate system, the interval to which the current water content scalar belongs is read. When the interval belongs to the extreme deficit interval, a continuous duty cycle upper limit signal is sent to the variable frequency pump. When the interval belongs to the moderate deficit interval, a continuous duty cycle lower limit signal is sent to the variable frequency pump. When the interval belongs to the interval approaching the upper limit, the water deficit scalar is used as the basis for pulse interval change. According to the quadratic attenuation law that the pulse duration width gradually shortens as the water deficit scalar gradually decreases, an intermittent pulse signal is sent to the variable frequency pump to form the variable frequency pump pulse width adjustment output state.

[0029] Based on the pulse width modulation output state of the variable frequency water pump, the triggering state of the closing command is continuously checked. When the closing command is not triggered, the selected output state among the continuous duty cycle upper limit signal, the continuous duty cycle lower limit signal, and the intermittent pulse signal is maintained. When the closing command is triggered, the intermittent pulse signal is stopped from being sent, and a stop command is sent to the variable frequency water pump simultaneously to switch the output state of the variable frequency water pump to the no-output state, thereby obtaining the adaptive flow regulation flow.

[0030] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0031] In this invention, by analyzing the curve relationships between historical desiccation stage moisture content scalars, historical desiccation stage water potential scalars, historical moisture absorption stage moisture content scalars, and historical moisture absorption stage water potential scalars, desiccation characteristic curves and moisture absorption characteristic curves are formed respectively. Then, the lower limit scalar for valve opening and the upper limit scalar for valve closing are extracted using the inflection point where the second derivative is zero. This ensures that the irrigation opening and closing boundaries no longer rely on a single fixed empirical value, but are jointly determined by the hysteresis differences during soil desiccation and moisture absorption processes, thus more accurately reflecting the water release and replenishment process in the root zone of palm vine seedlings. By comparing the current moisture content scalar with the hysteresis dual threshold parameters, a flip-switch command is generated, and this flip-switch command is mapped to discrete valve control. This system avoids frequent opening and closing of the solenoid valve when the moisture content fluctuates repeatedly near the critical point, thus improving the stability of irrigation operations. A water deficit scalar is established by the difference between the upper limit scalar of the valve closure and the current moisture content scalar. Based on the first deficit threshold and the second deficit threshold, the system divides the water supply intensity into extreme deficit range, moderate deficit range, and range approaching the upper limit, allowing the water supply intensity to vary in segments according to the degree of water deficit. In the range approaching the upper limit, intermittent pulse signals are issued with the water deficit scalar as the independent variable, and the system stops synchronously when the closing command is triggered. This can reduce excessive water supply, deep seepage, and ineffective operation of the water pump when approaching the target moisture content, thereby improving the stability of the water environment during the growth cycle of palm vine seedlings. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the steps of the present invention;

[0033] Figure 2 This is a schematic diagram of the desiccation characteristic curve and the moisture absorption characteristic curve. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0035] Please see Figure 1-2This invention provides a technical solution: an adaptive water control method for the growth cycle of palm vine seedlings, comprising the following steps:

[0036] The soil moisture sensor outputs historical desiccation and hygroscopic phases, along with their corresponding water potential scalars, are obtained. Curve fitting is then performed on these scalars to generate desiccation and hygroscopic characteristic curves. Second-order derivatives are calculated on both curves to extract the water content scalars corresponding to inflection points where the second derivative is zero. The water content scalar extracted from the desiccation characteristic curve is defined as the lower limit scalar for valve opening, and the water content scalar extracted from the hygroscopic characteristic curve is defined as the upper limit scalar for valve closing. These two scalars are then combined to construct a hysteresis dual-threshold parameter.

[0037] Obtain the current moisture content scalar, compare the current moisture content scalar with the valve opening lower limit scalar and valve closing upper limit scalar within the hysteresis double threshold parameters, and generate a toggle switch command; determine the polarity state of the toggle switch command and map it to the solenoid valve action state to generate the valve discrete control state;

[0038] When the valve discrete control state is an open command, calculate the difference between the upper limit scalar of valve closure and the current moisture content scalar to establish a moisture deficit scalar; compare the moisture deficit scalar with the preset first deficit threshold and second deficit threshold respectively to determine the interval to which the current moisture content scalar belongs.

[0039] Based on the interval assignment, when it is determined that the interval is approaching the upper limit, an intermittent pulse signal is sent to the variable frequency water pump with the water deficit scalar as the independent variable; when the closing command is triggered, the intermittent pulse signal is terminated and a shutdown command is sent to the variable frequency water pump simultaneously to obtain the adaptive flow regulation flow.

[0040] The steps for obtaining the desiccation characteristic curve and the moisture absorption characteristic curve are as follows:

[0041] The system acquires historical desiccation stage moisture content scalars, historical desiccation stage water potential scalars, historical hygroscopic stage moisture content scalars, and historical hygroscopic stage water potential scalars from the soil moisture sensor. Curve fitting is performed according to the mapping relationship between the historical desiccation stage moisture content scalars and the historical desiccation stage water potential scalars. Curve fitting is also performed according to the mapping relationship between the historical hygroscopic stage moisture content scalars and the historical hygroscopic stage water potential scalars. Abnormal mapping points with repeated moisture content scalars and discontinuous water potential scalar jumps are removed, and fitting segments with continuous mapping directions are retained to generate desiccation characteristic curves and hygroscopic characteristic curves.

[0042] Specifically, all data points continuously monitored and recorded by soil moisture sensors during historical irrigation and natural drying cycles were acquired. The datasets were divided into historical desiccation stage datasets and historical moisture absorption stage datasets based on the changing trends of moisture content. The historical desiccation stage dataset contained multiple pairs of historical desiccation stage moisture content scalars and historical desiccation stage water potential scalars, and the historical moisture absorption stage dataset was similarly divided. Independent curve fitting was performed on these two datasets. The Van Genuchten model was used to characterize the nonlinear relationship between moisture content and water potential. The model parameters were estimated using a nonlinear least squares method (such as the Levenberg-Marquardt algorithm). Before fitting, the data was preprocessed to remove outliers. Specifically, the data points were sorted by moisture content scalar, the difference in water potential scalar between adjacent data points was calculated, and a dynamic jump threshold was set. This threshold was determined by calculating the sum of the average of all differences and three times the standard deviation. For example, if the average difference was 0.05 MPa and the standard deviation was 0.02 MPa, then the threshold was 0.05 + 3 * 0.02 = 0.11. MPa. When the absolute value of the difference between the water potential scalar values ​​of two points with the same or very close water content exceeds this threshold, it is determined to be an abnormal mapping point caused by sensor noise or signal interference and is removed. After removing discontinuous jump points, continuous data segments with monotonic changing trends are retained, and these clean data points are used to finally complete the fitting, generating the desiccation characteristic curve representing the soil water loss process and the hygroscopic characteristic curve representing the soil water absorption process, respectively.

[0043] The steps for obtaining the hysteresis double threshold parameter are as follows:

[0044] Extract the curvature change corresponding to each moisture content scalar in the dehumidification characteristic curve, and extract the curvature change corresponding to each moisture content scalar in the moisture absorption characteristic curve. Confirm the inflection point where the curvature change changes from positive to negative or from negative to positive by the second derivative being zero. Read the moisture content scalar corresponding to the inflection point of the dehumidification characteristic curve and the moisture content scalar corresponding to the inflection point of the moisture absorption characteristic curve, and generate the lower limit scalar for opening the valve and the upper limit scalar for closing the valve.

[0045] Determine if the lower limit scalar of valve opening is less than the upper limit scalar of valve closing. If the lower limit scalar of valve opening is less than the upper limit scalar of valve closing, then construct the parameters in the order of lower limit scalar of valve opening first and upper limit scalar of valve closing last. If the lower limit scalar of valve opening is greater than or equal to the upper limit scalar of valve closing, then revisit the dehumidification characteristic curve and the moisture absorption characteristic curve to reconfirm the moisture content scalar corresponding to the inflection point, until the lower limit scalar of valve opening is less than the upper limit scalar of valve closing, and generate hysteresis double threshold parameters.

[0046] Specifically, based on the two continuous functions generated in the previous step—the desiccation characteristic curve and the hygroscopic characteristic curve—we find the points on these two curves where the rate of change of water potential with respect to moisture content is the fastest, i.e., the inflection points. These points physically correspond to the regions in the soil pore structure where the water state changes most drastically. Finding the inflection points is achieved by calculating the second derivative of the curves, thus representing the water potential... Considered moisture content function Calculate its second derivative. Since the characteristic curve is an analytic Van Genuchten function, its second derivative can be directly obtained analytically. Then, numerical methods are used to solve the equation. Specifically, the effective range of moisture content (from residual moisture content to saturated moisture content) is discretized, for example, with a step size of 0.001 cm³ / cm³. The second derivative value at each discrete point is calculated, and the position where the sign of the second derivative value changes is detected. The point where the second derivative is zero is accurately located by linear interpolation. The moisture content scalar corresponding to this point is the inflection point. The moisture content scalar corresponding to the inflection point calculated from the desiccation characteristic curve is defined as the trigger threshold for starting irrigation, i.e., the lower limit scalar of valve opening. The moisture content scalar corresponding to the inflection point calculated from the hygroscopic characteristic curve is defined as the target threshold for stopping irrigation, i.e., the upper limit scalar of valve closing. Finally, these two core control parameters are generated.

[0047] After obtaining the lower limit scalar of valve opening and the upper limit scalar of valve closing, a necessary physical logic check is performed: determining whether the lower limit scalar of valve opening is strictly less than the upper limit scalar of valve closing. This relationship directly reflects the soil moisture hysteresis effect. Under normal circumstances, during water loss, the soil moisture content is always higher than during water absorption at the same water potential. Therefore, the moisture content corresponding to the inflection point should also be lower. If the comparison result shows that the lower limit scalar of valve opening is less than the upper limit scalar of valve closing (e.g., the lower limit of valve opening is 0.25 and the upper limit of valve closing is 0.40), the check passes. At this time, following the fixed order of lower limit scalar of valve opening first and upper limit scalar of valve closing last, these two values ​​are combined into an ordered pair (0.25, 0.40). If the comparison result shows that the lower limit scalar of valve opening is greater than or equal to the upper limit scalar of valve closing, it is determined that there is an error in the previous curve fitting or inflection point extraction process, which may be due to excessive noise in the original data or the fitting algorithm getting stuck in a local optimum. At this time, an automatic callback correction process is initiated. This process first attempts to adjust the Van The initial parameters for fitting the Genuchten model are used and the fitting operation is re-performed. If the problem persists after one retry, a Savitzky-Golay smoothing filter (e.g., setting the window size to 5 data points and the polynomial order to 2) is applied to the original data before refitting to further reduce noise. This correction-verification loop is executed a maximum of 5 times. If the conditions are still not met after 5 times, a set of safe default values ​​preset according to the general growth requirements of palm vine seedlings and the current soil type (e.g., the default value for sandy soil is (0.20, 0.35)) is loaded until a valid threshold combination is obtained in which the lower limit of the valve opening is less than the upper limit of the valve closing, and finally, a hysteretic double threshold parameter is generated.

[0048] The steps to obtain the toggle switch command are as follows:

[0049] Obtain the current moisture content scalar, retrieve the lower limit scalar for opening the valve and the upper limit scalar for closing the valve within the hysteresis dual threshold parameters, align the current moisture content scalar item by item to the threshold range of the lower limit scalar for opening the valve and the upper limit scalar for closing the valve, mark the opening flip condition where the current moisture content scalar is less than or equal to the lower limit scalar for opening the valve, mark the closing flip condition where the current moisture content scalar plus the preset infiltration compensation scalar is greater than or equal to the upper limit scalar for closing the valve, mark the holding condition where the current moisture content scalar is greater than the lower limit scalar for opening the valve and the current moisture content scalar plus the preset infiltration compensation scalar is less than the upper limit scalar for closing the valve, and generate a flip switch command.

[0050] Specifically, the system acquires the current moisture content scalar value from the soil moisture sensor in real time and retrieves pre-defined hysteresis dual-threshold parameters from memory, namely the lower limit scalar value for valve opening and the upper limit scalar value for valve closing. To prevent over-irrigation, a preset infiltration compensation scalar value is introduced. This scalar value compensates for the phenomenon that the sensor reading continues to rise due to gravity-induced downward infiltration of water after the valve is closed. The value of this compensation scalar value is determined through calibration experiments on specific soil plots. For example, in an irrigation test, water supply is stopped when the moisture content reaches a certain level, and the subsequent natural maximum increase in moisture content is recorded. Multiple measurements are taken and the average value is calculated. If the average increase for the current loam soil is 0.015... If the current water content is set to cm³ / cm³, the preset infiltration compensation scalar is set to this value. Then, the current water content scalar is compared with these two thresholds to establish three mutually exclusive logical conditions: First, mark the opening flip condition, which is true if the current water content scalar is less than or equal to the lower valve opening limit scalar; Second, mark the closing flip condition, which is true if the sum of the current water content scalar and the preset infiltration compensation scalar is greater than or equal to the upper valve closing limit scalar; Third, mark the holding condition, which is true if the current water content scalar is greater than the lower valve opening limit scalar and the sum of the current water content scalar and the preset infiltration compensation scalar is less than the upper valve closing limit scalar. Based on which condition is met at the current moment, the corresponding flip switch command is generated. This command is a status identifier, such as "open", "closed" or "hold".

[0051] The steps for obtaining the discrete control state of a valve are as follows:

[0052] Read the polarity states corresponding to the opening flip condition, closing flip condition, and holding condition. When the opening flip condition is met, map the flip switch command to the solenoid valve opening action. When the closing flip condition is met, map the flip switch command to the solenoid valve closing action. When the holding condition is met, read the current solenoid valve action state and retain the original polarity of the current solenoid valve action state to form the solenoid valve action state.

[0053] Verify the solenoid valve's opening action, closing action, and the reserved state of the current solenoid valve's action state. When the solenoid valve's action state corresponds to the solenoid valve's opening action, output an opening command; when the solenoid valve's action state corresponds to the solenoid valve's closing action, output a closing command; when the solenoid valve's action state corresponds to the reserved state of the current solenoid valve's action state, maintain the current solenoid valve's action state unchanged, and generate the valve's discrete control state.

[0054] Specifically, based on the toggle switch command generated in the previous step (whose state is "on", "closed", or "held"), and combined with the current operating state of the solenoid valve, the next target action of the solenoid valve is determined. First, the polarity state of the solenoid valve is defined, for example, using logic "1" to represent on and logic "0" to represent closed. Then, a state variable stored in the controller's memory is read, namely the current solenoid valve operating state, which records the final state (on or closed) of the solenoid valve in the previous control cycle. Next, mapping is performed according to the toggle switch command: if the toggle switch command is "on", then the target action (i.e., the solenoid valve operating state) is set. If the solenoid valve is set to "open" (logic "1"), it means that the next step should be to enter the open state, regardless of whether the valve is currently open or closed. If the toggle switch command is "closed", the solenoid valve action state is set to "closed" (logic "0"). If the toggle switch command is "hold", no changes are made. The current value of the solenoid valve action state is directly read (logic "1" or "0") and used as the target action for this time, that is, its original polarity is retained. This process transforms the three possibilities (open, closed, hold) from the comparison logic into a clear, binary target state (open or closed), forming the solenoid valve action state.

[0055] Based on the determined solenoid valve operating state (i.e., target state), the physical control signal that ultimately drives the solenoid valve hardware is generated. First, the solenoid valve operating state is compared with the actual output state of the previous cycle. When the solenoid valve operating state is determined to be an open action, the controller will output a high-level signal or a specific digital instruction, i.e., an open instruction, to the drive circuit connected to the solenoid valve to drive the valve open. When the solenoid valve operating state is determined to be a closed action, a low-level signal or a corresponding close instruction, i.e., a close instruction, is output to close the valve. When the solenoid valve operating state corresponds to the reserved state of the current solenoid valve operating state, it means that the target state of the system is consistent with the current actual state (e.g., the valve is already open, and the instruction is to keep it open). At this time, the controller will maintain the existing output signal unchanged, for example, continue to output a high level to keep the valve open, instead of sending a new instruction. After generating the corresponding open or close instruction, the state variable recording the current solenoid valve state in memory is updated synchronously to keep it consistent with the newly issued instruction, providing a basis for the judgment of the next control cycle. The stable, jitter-free discrete level signal generated in this way is the discrete control state of the valve.

[0056] The steps to obtain the water deficit scalar are as follows:

[0057] When the valve discrete control state is an open command, read the numerical position of the upper limit scalar of valve closure, read the numerical position of the current moisture content scalar, calculate the difference according to the numerical interval between the upper limit scalar of valve closure and the current moisture content scalar, discard invalid intervals less than zero in the difference calculation, retain the effective interval representing the distance between the current moisture content scalar and the upper limit scalar of valve closure, define the effective interval as the degree of moisture deficit, and establish a moisture deficit scalar.

[0058] Specifically, the calculation of water deficit is initiated only when the valve discrete control state is an open command, i.e., irrigation is in progress. The purpose of this calculation is to quantify the difference between the current soil moisture and the ideal saturation state, providing a basis for subsequent variable frequency pump flow regulation. The calculation process first reads the specific value of the upper limit scalar of valve closure from the hysteresis dual threshold parameter, and at the same time obtains the real-time current moisture content scalar from the sensor. Then, a subtraction operation is performed to calculate the difference between the upper limit scalar of valve closure and the current moisture content scalar. This difference directly represents the amount of water that needs to be replenished. After the calculation, a validity check is performed. Since the current moisture content must be lower than the upper limit of valve closure when irrigation is started, theoretically the difference should be positive. If the calculation result is less than or equal to zero, it may indicate an abnormal instantaneous sensor reading or a system response delay. In this case, the invalid interval is judged as no deficit, and the water deficit level is forcibly set to 0. For all valid differences greater than zero, the value is directly adopted and defined as a physical quantity characterizing the current water deficit level. Finally, a real-time changing water deficit scalar is established.

[0059] The specific implementation of the step to calculate the degree of water deficit is as follows: In the Within each control cycle, the controller reads the discrete control state of the valve. ,in, Indicates the start command. Indicates a closing instruction; when detected At that time, the soil moisture sensor output is continuously read according to the preset sampling cycle. Moisture content sampling value Calculate the current moisture content scalar value ,Right now ,in, To preset the number of consecutive samples, This represents the i-th moisture content sample value; simultaneously, the lower limit scalar value for valve opening is read from the hysteresis double threshold parameter. and the upper limit scalar of the valve Calculate the difference between the current moisture content scalar and the upper limit scalar of the valve closing. The width of the moisture content adjustment range is determined by the hysteresis dual threshold parameter. As a normalization benchmark, the moisture content difference and the width of the moisture content adjustment range are calculated according to... Calculate the degree of water deficit, among which, The dimensionless normalized water deficit degree; when hour, This indicates that the current moisture content is at the lower limit of the valve opening position; when hour, lie in to Between, and gradually decreases as the current moisture content scalar approaches the upper limit scalar of the valve; when At that time, the positive value truncation function is used to cut off the value. Set as The controller updates according to the control cycle. The normalized water deficit level is used as the water deficit scalar.

[0060] The steps to obtain the interval affiliation are as follows:

[0061] Retrieve the preset first and second deficit thresholds. First, check the threshold order where the first deficit threshold is greater than the second deficit threshold. Then, align the water deficit scalar to the numerical boundaries of the first and second deficit thresholds respectively. Record the comparison status when the water deficit scalar is greater than the first deficit threshold. Record the comparison status when the water deficit scalar is less than or equal to the first deficit threshold and greater than the second deficit threshold. Record the comparison status when the water deficit scalar is less than or equal to the second deficit threshold. Generate the deficit threshold comparison result.

[0062] Based on the deficit threshold comparison results, the comparison status corresponding to the moisture deficit scalar is read. When the moisture deficit scalar is greater than the first deficit threshold, the current moisture content scalar is determined to be in the extreme deficit range. When the moisture deficit scalar is less than or equal to the first deficit threshold and greater than the second deficit threshold, the current moisture content scalar is determined to be in the moderate deficit range. When the moisture deficit scalar is less than or equal to the second deficit threshold, the current moisture content scalar is determined to be in the range approaching the upper limit, and the range assignment is generated.

[0063] Specifically, after obtaining the real-time updated water deficit scalar value, to achieve refined hierarchical control of irrigation flow, it needs to be classified into different deficit levels. This is achieved by retrieving two preset deficit thresholds: a first deficit threshold and a second deficit threshold. These two thresholds are set based on the total irrigation regulation range, i.e., the difference between the upper limit of the valve closing scalar value and the lower limit of the valve opening scalar value (hysteresis interval width). The specific setting method is as follows: the first deficit threshold is set to 60% of this interval width, and the second deficit threshold is set to 20% of this interval width. For example, if the upper limit of valve closing is 0.40 and the lower limit of valve opening is 0.25, then the interval width is 0.15. Therefore, the first deficit threshold = 0.15 * 60% = 0.09, and the second deficit threshold = 0.15 * 20%. =0.03. After setting, first verify that the first deficit threshold (0.09) is indeed greater than the second deficit threshold (0.03) to ensure logical correctness. Then, compare the current water deficit scalar with the numerical boundaries of the two thresholds and record three possible comparison states: water deficit scalar is greater than the first deficit threshold; water deficit scalar is less than or equal to the first deficit threshold and greater than the second deficit threshold; water deficit scalar is less than or equal to the second deficit threshold. Record the judgment results (Boolean values) of these three states to generate the deficit threshold comparison result.

[0064] Based on the deficit threshold comparison results generated in the previous step, the degree of water deficit is mapped to a clear, discrete interval classification. This classification result will be directly used to guide the operation strategy of the variable frequency water pump. The mapping rules are as follows: Read the deficit threshold comparison results and check which comparison state is true. When the record shows that the current water deficit scalar is greater than the first deficit threshold (for example, the current deficit is 0.12, which is greater than the threshold of 0.09), the current water content is determined to be in the extreme deficit interval, which means that the soil is very dry and requires a large amount of rapid water replenishment; when the record shows that the water deficit scalar is less than or equal to the first deficit threshold and greater than the second deficit threshold (for example, the current deficit is less than or equal to the first deficit threshold and greater than the second deficit threshold), the water deficit is determined to be in the extreme deficit interval. When the water deficit is 0.07 (between 0.09 and 0.03), the current water content is determined to be in a gradual deficit range, indicating that the soil moisture has been replenished to a certain extent and the water replenishment rate needs to be reduced. When the record shows that the water deficit scalar is less than or equal to the second deficit threshold (for example, the current deficit is 0.02, which is less than the threshold of 0.03), the current water content is determined to be in a range approaching the upper limit, indicating that the soil moisture content is close to the target value and a more refined method of fine-tuning water replenishment is needed. Through this mapping process, the continuously changing water deficit scalar is converted into one of the three discrete range assignment states of "extreme deficit", "gradual deficit" and "approaching the upper limit", generating the range assignment.

[0065] The steps for obtaining the adaptive flow regulation stream are as follows:

[0066] Read the extreme deficit interval, moderate deficit interval, and near-upper limit interval corresponding to the water deficit scalar. Map the extreme deficit interval to the upper limit output position of the continuous duty cycle in the variable frequency pump pulse width adjustment coordinate system, map the moderate deficit interval to the lower limit output position of the continuous duty cycle in the variable frequency pump pulse width adjustment coordinate system, and map the near-upper limit interval to the intermittent pulse output position in the variable frequency pump pulse width adjustment coordinate system to generate a piecewise mapping relationship between the water deficit scalar and the variable frequency pump pulse width adjustment coordinate system.

[0067] Based on the piecewise mapping relationship between the water deficit scalar and the variable frequency pump pulse width adjustment coordinate system, the interval to which the current water content scalar belongs is read. When the interval belongs to the extreme deficit interval, a continuous duty cycle upper limit signal is sent to the variable frequency pump. When the interval belongs to the moderate deficit interval, a continuous duty cycle lower limit signal is sent to the variable frequency pump. When the interval belongs to the interval approaching the upper limit, the water deficit scalar is used as the basis for pulse interval change. According to the quadratic attenuation law of the pulse duration width gradually shortening as the water deficit scalar gradually decreases, an intermittent pulse signal is sent to the variable frequency pump to form the variable frequency pump pulse width adjustment output state.

[0068] Based on the pulse width modulation output state of the variable frequency water pump, the triggering state of the closing command is continuously checked. When the closing command is not triggered, the selected output state among the continuous duty cycle upper limit signal, continuous duty cycle lower limit signal and intermittent pulse signal is maintained. When the closing command is triggered, the intermittent pulse signal is stopped from being sent, and a stop command is sent to the variable frequency water pump at the same time to switch the output state of the variable frequency water pump to the no-output state, thus obtaining the adaptive flow regulation flow.

[0069] Specifically, based on the three defined states—extreme deficit, moderate deficit, and near-maximum deficit—a clear set of control rules is established, from the water deficit state to the pulse width modulation (PWM) output signal of the variable frequency pump. This process predefines a specific pump operating mode for each state, forming a segmented mapping relationship. The mapping is as follows: the extreme deficit state is mapped to a high duty cycle continuous PWM signal, which is the output position at the upper limit of the continuous duty cycle, for example, set to 95% duty cycle, to drive the pump to operate at near full power, achieving rapid water replenishment at maximum flow; the moderate deficit state is mapped to… A continuous PWM signal with a medium duty cycle is output at the lower limit of the continuous duty cycle, for example, a 50% duty cycle, so that the water pump continuously replenishes water at a stable medium flow rate. The interval approaching the upper limit is mapped to an intermittent pulse working mode, i.e., the intermittent pulse output position. In this mode, the water pump no longer runs continuously, but fine-tunes the water supply through a series of switching pulses. This set of fixed rules assigns the three discrete intervals to three distinct water pump control strategies (high continuous flow, medium continuous flow, and intermittent pulse flow), generating a piecewise mapping relationship between the water deficit scalar and the variable frequency water pump pulse width adjustment coordinate system.

[0070] Based on the established segmented mapping relationship and reading the current moisture content interval, specific control signals are sent to the variable frequency pump. When the interval is classified as an extremely deficient interval, the controller continuously sends a preset upper limit signal of the continuous duty cycle (such as a PWM wave with a 95% duty cycle) to the driver of the variable frequency pump. When the interval is classified as a moderately deficient interval, a lower limit signal of the continuous duty cycle (such as a PWM wave with a 50% duty cycle) is sent. When the interval is classified as a range approaching the upper limit, the control method becomes more refined. At this time, the real-time changing moisture deficit scalar is used as the independent variable for dynamically adjusting the pulse parameters, and the pulse duration (i.e., the pump start time) is also considered. ) and water deficit scalar (denoted as It is proportional to the square of ) and follows the law of quadratic decay, specifically the relationship is as follows: Among them, the proportionality coefficient A preset adjustment gain is determined through on-site debugging. For example, if the expected maximum pulse width corresponding to the maximum deficit value in this range (i.e., the second deficit threshold of 0.03) is 1.5 seconds, then... The pump shut-off time within the pulse cycle (e.g., fixed at 5 seconds) Then it is ,along with Gradually decrease, It will shorten the time more quickly, achieving precise "point irrigation" at the end of the irrigation period. In this way, the corresponding control logic is selected and executed according to the different intervals, forming the output state of the variable frequency water pump pulse width regulation.

[0071] While the variable frequency water pump is continuously running according to its pulse width modulation output state, the controller continuously checks the trigger status of the closing command generated by the moisture content comparison logic in a parallel fast cycle. When the closing command has not yet been triggered, the system maintains the current output strategy, that is, according to the interval to which the moisture deficit scalar belongs, it continues to send the corresponding continuous duty cycle upper limit signal, continuous duty cycle lower limit signal, or dynamically adjusted intermittent pulse signal to maintain the normal irrigation operation of the water pump. Once the closing command is detected to be triggered (i.e., the current moisture content plus infiltration compensation has reached the valve closing limit), the system immediately executes the synchronous shutdown procedure. First, regardless of the current state of the water pump... In terms of output status, especially for intermittent pulse signals being sent, immediately stop the generation and transmission of all PWM signals. Secondly, synchronously send a clear stop command to the variable frequency pump driver. This command usually sets the PWM duty cycle to 0% or triggers the stop through a dedicated control pin, reducing the pump's output power to zero, i.e., switching to an empty output state. This action is coordinated with the closing action of the solenoid valve to ensure that when the target moisture content is achieved, the water supply power source and the passage are cut off simultaneously. The complete flow output process of the pump from high speed to low speed to pulse and finally to stop during the entire irrigation cycle constitutes the final adaptive flow regulation flow.

[0072] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. An adaptive water control method for the growth cycle of palm vine seedlings, characterized in that, Includes the following steps: The soil moisture sensor outputs historical desiccation and hygroscopic phases, along with their corresponding water potential scalars. Curve fitting is then performed on these scalars to generate desiccation and hygroscopic characteristic curves. Second-order derivatives are calculated on both curves to extract the water content scalars corresponding to inflection points where the second derivative is zero. The water content scalars extracted from the desiccation curves are defined as the lower limit scalar for valve opening, and the water content scalars extracted from the hygroscopic curves are defined as the upper limit scalar for valve closing. These two scalars are then combined to construct a hysteresis dual-threshold parameter. Obtain the current moisture content scalar, compare the current moisture content scalar with the valve opening lower limit scalar and the valve closing upper limit scalar within the hysteresis dual threshold parameters, and generate a toggle switch command; determine the polarity state of the toggle switch command and map it to the solenoid valve action state to generate a discrete valve control state; When the discrete control state of the valve is an opening command, the difference between the upper limit scalar of the valve closing and the current moisture content scalar is calculated to establish a moisture deficit scalar; the moisture deficit scalar is compared with the preset first deficit threshold and the second deficit threshold respectively to determine the interval to which the current moisture content scalar belongs. According to the interval assignment, when it is determined that it is in the interval approaching the upper limit, an intermittent pulse signal is sent to the variable frequency water pump with the water deficit scalar as the independent variable. When the closing command is triggered, the intermittent pulse signal is terminated, and a stop command is simultaneously sent to the variable frequency water pump to obtain an adaptive flow regulation flow. The steps for obtaining the desiccation characteristic curve and the moisture absorption characteristic curve are as follows: The system acquires the historical desiccation stage moisture content scalar, historical desiccation stage water potential scalar, historical hygroscopic stage moisture content scalar, and historical hygroscopic stage water potential scalar output from the soil moisture sensor. Curve fitting is performed according to the mapping relationship between the historical desiccation stage moisture content scalar and the historical desiccation stage water potential scalar. Curve fitting is also performed according to the mapping relationship between the historical hygroscopic stage moisture content scalar and the historical hygroscopic stage water potential scalar. Abnormal mapping points with repeated moisture content scalar and discontinuous water potential scalar jumps are removed, and fitting segments with continuous mapping directions are retained to generate desiccation characteristic curves and hygroscopic characteristic curves. The steps for obtaining the hysteresis dual threshold parameter are as follows: Extract the curvature change corresponding to each moisture content scalar in the dehumidification characteristic curve, and extract the curvature change corresponding to each moisture content scalar in the moisture absorption characteristic curve. Confirm the inflection point where the curvature change changes from positive to negative or from negative to positive by the second derivative being zero. Read the moisture content scalar corresponding to the inflection point of the dehumidification characteristic curve and the moisture content scalar corresponding to the inflection point of the moisture absorption characteristic curve, and generate the lower limit scalar for opening the valve and the upper limit scalar for closing the valve. Determine whether the lower limit scalar of valve opening is less than the upper limit scalar of valve closing. If the lower limit scalar of valve opening is less than the upper limit scalar of valve closing, then construct the parameters by combining them in the order of the lower limit scalar of valve opening first and the upper limit scalar of valve closing last. If the lower limit scalar of valve opening is greater than or equal to the upper limit scalar of valve closing, then revisit the dehumidification characteristic curve and the moisture absorption characteristic curve to reconfirm the moisture content scalar corresponding to the inflection point, until the lower limit scalar of valve opening is less than the upper limit scalar of valve closing, and generate a hysteresis double threshold parameter.

2. The adaptive water control method for the growth cycle of palm vine seedlings according to claim 1, characterized in that, The steps for obtaining the flip switch command are as follows: Obtain the current moisture content scalar, retrieve the lower limit scalar for opening the valve and the upper limit scalar for closing the valve within the hysteresis dual threshold parameters, align the current moisture content scalar item by item to the threshold range of the lower limit scalar for opening the valve and the upper limit scalar for closing the valve, mark the opening flip condition where the current moisture content scalar is less than or equal to the lower limit scalar for opening the valve, mark the closing flip condition where the current moisture content scalar plus the preset infiltration compensation scalar is greater than or equal to the upper limit scalar for closing the valve, mark the holding condition where the current moisture content scalar is greater than the lower limit scalar for opening the valve and the current moisture content scalar plus the preset infiltration compensation scalar is less than the upper limit scalar for closing the valve, and generate a flip switch command.

3. The adaptive water control method for the growth cycle of palm vine seedlings according to claim 1, characterized in that, The steps for obtaining the discrete control state of the valve are as follows: Read the polarity states corresponding to the opening flip condition, closing flip condition, and holding condition. When the opening flip condition is met, map the flip switch command to the solenoid valve opening action. When the closing flip condition is met, map the flip switch command to the solenoid valve closing action. When the holding condition is met, read the current solenoid valve action state and retain the original polarity of the current solenoid valve action state to form the solenoid valve action state. Verify the solenoid valve's opening action, closing action, and the reserved state of the current solenoid valve's action state. When the solenoid valve's action state corresponds to the solenoid valve's opening action, output an opening command; when the solenoid valve's action state corresponds to the solenoid valve's closing action, output a closing command; when the solenoid valve's action state corresponds to the reserved state of the current solenoid valve's action state, maintain the current solenoid valve's action state unchanged, and generate the valve's discrete control state.

4. The adaptive water control method for the growth cycle of palm vine seedlings according to claim 1, characterized in that, The steps for obtaining the water deficit scalar are as follows: When the discrete control state of the valve is an open command, the numerical position of the upper limit scalar of valve closure is read, the numerical position of the current moisture content scalar is read, the difference is calculated according to the numerical interval between the upper limit scalar of valve closure and the current moisture content scalar, invalid intervals less than zero in the difference calculation are eliminated, and the effective interval representing the distance between the current moisture content scalar and the upper limit scalar of valve closure is retained. The effective interval is defined as the degree of moisture deficit, and a moisture deficit scalar is established.

5. The adaptive water control method for the growth cycle of palm vine seedlings according to claim 1, characterized in that, The steps for obtaining the interval affiliation are as follows: Retrieve the preset first and second deficit thresholds, first check the threshold order where the first deficit threshold is greater than the second deficit threshold, then align the water deficit scalar to the numerical boundaries of the first and second deficit thresholds respectively, record the comparison status when the water deficit scalar is greater than the first deficit threshold, record the comparison status when the water deficit scalar is less than or equal to the first deficit threshold and greater than the second deficit threshold, record the comparison status when the water deficit scalar is less than or equal to the second deficit threshold, and generate the deficit threshold comparison result; Based on the deficit threshold comparison results, the comparison status corresponding to the moisture deficit scalar is read. When the moisture deficit scalar is greater than the first deficit threshold, the current moisture content scalar is determined to be in the extreme deficit range. When the moisture deficit scalar is less than or equal to the first deficit threshold and greater than the second deficit threshold, the current moisture content scalar is determined to be in the moderate deficit range. When the moisture deficit scalar is less than or equal to the second deficit threshold, the current moisture content scalar is determined to be in the range approaching the upper limit, and the range assignment is generated.

6. The adaptive water control method for the growth cycle of palm vine seedlings according to claim 1, characterized in that, The steps for obtaining the adaptive flow regulation stream are as follows: Read the extreme deficit interval, moderate deficit interval, and near-upper limit interval corresponding to the water deficit scalar, map the extreme deficit interval to the upper limit output position of the continuous duty cycle in the variable frequency pump pulse width adjustment coordinate system, map the moderate deficit interval to the lower limit output position of the continuous duty cycle in the variable frequency pump pulse width adjustment coordinate system, and map the near-upper limit interval to the intermittent pulse output position in the variable frequency pump pulse width adjustment coordinate system, thereby generating a piecewise mapping relationship between the water deficit scalar and the variable frequency pump pulse width adjustment coordinate system; Based on the piecewise mapping relationship between the water deficit scalar and the variable frequency pump pulse width adjustment coordinate system, the interval to which the current water content scalar belongs is read. When the interval belongs to the extreme deficit interval, a continuous duty cycle upper limit signal is sent to the variable frequency pump. When the interval belongs to the moderate deficit interval, a continuous duty cycle lower limit signal is sent to the variable frequency pump. When the interval belongs to the interval approaching the upper limit, the water deficit scalar is used as the basis for pulse interval change. According to the quadratic attenuation law that the pulse duration width gradually shortens as the water deficit scalar gradually decreases, an intermittent pulse signal is sent to the variable frequency pump to form the variable frequency pump pulse width adjustment output state. Based on the pulse width modulation output state of the variable frequency water pump, the triggering state of the closing command is continuously checked. When the closing command is not triggered, the selected output state among the continuous duty cycle upper limit signal, the continuous duty cycle lower limit signal, and the intermittent pulse signal is maintained. When the closing command is triggered, the intermittent pulse signal is stopped from being sent, and a stop command is sent to the variable frequency water pump simultaneously to switch the output state of the variable frequency water pump to the no-output state, thereby obtaining the adaptive flow regulation flow.

Images