Intelligent remote control method and system for irrigation electric valve

Abstract

The invention belongs to the technical field of control, and particularly relates to an intelligent remote control method and system for an irrigation electric valve, and the method comprises the following steps: S1, obtaining a target soil humidity set value and a current soil humidity actual value of an irrigation region; based on the network communication quality parameters between the remote control terminal and the electric valve, the pipe network pressure sensor data and the historical instruction execution data, the predicted execution lag time of the current control period is obtained through calculation; s2, calculating a theoretical prediction value of the soil humidity by using a preset water and soil model according to the valve opening degree and the environmental parameters of the previous period, and obtaining a model deviation value between the theoretical prediction value and the current actual soil humidity value; by comprehensively analyzing the network quality, the pipe network pressure and the historical data, execution lag in the control process can be recognized and compensated in advance, excessive or insufficient irrigation caused by delay is avoided, and the control stability is improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of control, and particularly relates to an intelligent remote control method and system for an irrigation electric valve. BACKGROUND

[0002] The remote intelligent irrigation system can not only improve the utilization rate of water resources, but also improve the crop yield. The existing intelligent irrigation system usually obtains real-time data through soil humidity sensors deployed in the field, and sends opening and closing instructions to the electric valve of the irrigation area according to the preset humidity threshold value, so as to realize automatic irrigation. From the control center issuing an instruction, to the instruction being transmitted to the on-site controller through a wireless or wired network, to the electric valve completing mechanical action, the whole process has a lag, which often causes control overshoot, resulting in unnecessary excessive irrigation or insufficient irrigation, causing waste of water resources and adverse effects on crop roots.

[0003] The PID control is applied to the irrigation system, although the process of control is considered, but the network delay fluctuation of the remote irrigation system will affect the arrival time of the control instruction, and the traditional PID controller is difficult to cope with the random change of the pure lag time. The response of the irrigation system is comprehensively affected by many factors such as pipe network water pressure, soil texture, environmental temperature and humidity, wind speed, etc. When the actual situation does not match the preset, for example, the valve fails to completely execute the opening degree instruction due to a fault, a deviation will be generated, resulting in a decrease in control accuracy. The current technical solution generally lacks a mechanism for online evaluation and compensation of the model deviation, and also fails to fuse the network quality, historical execution effect and other information into the control decision to realize the adjustment of the control strategy, thereby limiting the stability of the irrigation control. SUMMARY

[0004] The present application provides an intelligent remote control method and system for an irrigation electric valve, to solve the technical problems of remote irrigation control lag fluctuation, model deviation without compensation, and multi-source information not fused in the prior art, resulting in insufficient precision and stability.

[0005] In a first aspect, the present application provides an intelligent remote control method for an irrigation electric valve, comprising the following steps: S1, obtaining a target soil humidity set value and a current soil humidity actual value of an irrigation area; based on network communication quality parameters between a remote control terminal and an electric valve, pipe network pressure sensor data and historical instruction execution data, a predicted execution lag time of the current control period is calculated; S2, using a preset soil-water model, a theoretical prediction value of soil humidity is calculated according to the valve opening degree and environmental parameters of the last control cycle, and a model deviation between the theoretical prediction value and the actual value of the current soil humidity is obtained; when the model deviation exceeds a first preset threshold, the actual running state of the electric valve is inquired, and the soil-water model is corrected based on the inquiry result; S3, based on a current target difference between the actual value of the current soil humidity and the target soil humidity set value, the target difference of the last control cycle, the predicted execution lag time and the model deviation, a valve opening degree adjustment amount of the next control cycle is calculated; the predicted execution lag time is used to determine a control gain attenuation coefficient, and the model deviation is used to determine an adjustment amplitude weight; S4, a control instruction with a target opening degree being a sum of the current valve opening degree and the valve opening degree adjustment amount is generated, and after the expected execution time of the control instruction is set in combination with the predicted execution lag time, the control instruction is sent to the electric valve; whether the absolute value of the current target difference is less than a second preset threshold and whether the moving average value of the target difference change rate is less than a change rate threshold in a continuous preset number of control cycles are judged, and when the two conditions are met at the same time, the current irrigation control is ended.

[0006] Further, based on the network communication quality parameters between the remote control terminal and the electric valve, the pipe network pressure sensor data and the historical instruction execution data, a predicted execution lag time of the current control cycle is calculated, including: Based on the historical instruction issue time stamp, the instruction receiving time stamp and the instruction execution completion time stamp fed back by the electric valve, and the pipe network pressure data during execution, a relationship model between the instruction execution delay and the pipe network pressure is established; The network data packet round trip time and the packet loss rate between the terminal and the electric valve at the current time are obtained, and a predicted transmission delay under the current network state is calculated; The current pipe network pressure sensor data is obtained, and the current instruction execution delay is predicted using the relationship model; The predicted transmission delay and the predicted current instruction execution delay are added to obtain the predicted execution lag time of the current control cycle.

[0007] Further, when the model deviation exceeds the first preset threshold, the actual running state of the electric valve is inquired, and the soil-water model is corrected based on the inquiry result, including: The first preset threshold is set to 5% of the target soil humidity set value; The actual average opening degree and the cumulative opening time of the electric valve in the last control cycle are inquired; According to the inquired actual average opening degree and the cumulative opening time, the actual irrigation water quantity of the last cycle is recalculated in combination with the pipe network pressure; The actual irrigation water amount is taken as an input, and the soil infiltration rate parameter in the water-soil model is adjusted by an optimization algorithm, so that the error between the humidity prediction value of the water-soil model under the input and the actual value of the current soil humidity is minimized.

[0008] Further, the prediction execution lag time is used to determine a control gain attenuation coefficient, including: The control gain attenuation coefficient is calculated by the following formula : , wherein, is the prediction execution lag time; is a preset attenuation constant; and exp is a natural exponential function.

[0009] Further, the model deviation amount is used to determine an adjustment amplitude weight, including: The adjustment amplitude weight W is calculated by the following formula: , wherein, is the absolute value of the model deviation amount; is a preset weight influence factor.

[0010] Further, based on a current target difference value between the actual value of the current soil humidity and the set value of the target soil humidity, a target difference value of the last period, the prediction execution lag time, and the model deviation amount, an adjustment amount of the valve opening degree of the next control period is calculated, including: The adjustment amount of the valve opening degree is calculated by the following improved incremental proportional-integral control algorithm : , wherein W is the adjustment amplitude weight; is a proportional gain; is an integral gain; is the current target difference value; is the target difference value of the last period; is the control gain attenuation coefficient.

[0011] Further, after the expected execution time of the control instruction is set in combination with the prediction execution lag time, the control instruction is sent to the electric valve, including: An current system timestamp of the control terminal is acquired ; The prediction execution lag time is added to the current system timestamp to obtain the expected execution time ; A control instruction data packet containing the target opening degree and the expected execution time is encapsulated, and is sent to the electric valve through a wireless communication module.

[0012] Further, in S4, the change rate threshold is set according to the prediction execution lag time, including: The change rate threshold is calculated using the following linear function : wherein, is a base change rate threshold; is a hysteresis influence coefficient; is a prediction execution hysteresis time.

[0013] Further, the control instruction data packet is sent to the electric valve through the NB-IoT module.

[0014] In a second aspect, the present application provides an intelligent remote control system for an irrigation electric valve, comprising a memory and a processor, wherein the memory stores computer program instructions, and the computer program instructions, when executed by the processor, implement the intelligent remote control method for the irrigation electric valve.

[0015] The beneficial effects are: the present application can pre-identify and compensate the execution hysteresis in the control process by comprehensively analyzing the network quality, pipe network pressure and historical data, avoids the irrigation over or under caused by delay, and improves the stability of the control. By comparing the theoretical prediction value of the water-soil model with the actual value of the sensor, the model deviation can be found and used to correct the subsequent control decision, even when the actual operating state of the valve does not match the instruction or the environmental factors change suddenly. In the method, the predicted hysteresis time and the model deviation are used as key adjustment factors when calculating the valve opening adjustment amount and setting the irrigation end condition, so that the control output can quickly respond to the target deviation and suppress system oscillation, ensuring that the system can stop smoothly after reaching the target humidity, achieving more accurate soil moisture management, saving water resources and ensuring irrigation quality. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 The flowchart of the intelligent remote control method for the irrigation electric valve. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of the present application.

[0018] The embodiments of the intelligent remote control method for the irrigation electric valve provided by the present application are as follows: As shown in the drawings, Figure 1 The intelligent remote control method for the irrigation electric valve comprises the following steps: S1, obtain the target soil moisture set value of the irrigation area and the current soil moisture actual value collected by the sensor; based on the network communication quality parameter between the remote control terminal and the electric valve, the pipe network pressure sensor data and the historical instruction execution data, the predicted execution lag time of the current control cycle is calculated.

[0019] The remote control terminal such as a mobile phone APP or a webpage terminal pre-sets and stores the target soil moisture set value, for example, 65% relative humidity; the soil moisture sensor deployed in the irrigation area periodically collects the current soil moisture actual value, for example, 50% relative humidity, through a wireless module such as LoRa, and uploads the data to the central control server. The network communication delay is obtained by performing a ping operation on the electric valve controller or recording the round-trip time of the data packet, for example, the one-way delay is 300 ms; according to the pipe network pressure sensor reading, the hydraulic delay is found from the pre-set pressure-flow delivery time relationship table, for example, 0.2 MPa pressure corresponds to 180 s hydraulic delay; the actual time record from the sending of the irrigation instruction to the response of the humidity sensor in the latest several times is recalled, the weighted average value is calculated, and the above-mentioned delay is combined to comprehensively obtain the predicted execution lag time of the current control cycle, for example, the total is 210 s.

[0020] In an optional embodiment, based on the network communication quality parameter between the remote control terminal and the electric valve, the pipe network pressure sensor data and the historical instruction execution data, the predicted execution lag time of the current control cycle is calculated, including: Based on the historical instruction issue timestamp, the electric valve feedback instruction receiving timestamp, the instruction execution completion timestamp and the pipe network pressure data during execution, a relationship model between the instruction execution delay and the pipe network pressure is established; Obtain the network data packet round-trip time and packet loss rate between the terminal and the electric valve at the current time, and calculate the predicted transmission delay under the current network state; Obtain the current pipe network pressure sensor data, and predict the current instruction execution delay by using the relationship model; Add the predicted transmission delay and the predicted current instruction execution delay to obtain the predicted execution lag time of the current control cycle.

[0021] Specifically, historical data, such as a database containing 1000 instruction execution records, each record having instruction issuing time, valve receiving time, execution completion time and average pipe network pressure during the period, are analyzed. Through regression analysis of the data, it is found that when the pipe network pressure increases from 1.5 MPa to 3.0 MPa, the execution time delay of the valve from full closing to half opening increases from 25 s to 35 s. Accordingly, a relationship model is established, such as the instruction execution time delay being equal to 15 plus 6.67 times the pipe network pressure value. At the beginning of the current control cycle, a test data packet is sent to the valve and a response is received, and the network data packet round trip time RTT is measured to be 200 ms, and the packet loss rate is 1%. According to the data, the predicted transmission time delay under the current network state is calculated to be 202 ms, i.e. 0.202 s. The pipe network pressure sensor is read to obtain the current pressure value of 2.5 MPa. The pressure value is substituted into the previously established relationship model to predict the current instruction execution time delay of 31.68 s. The predicted transmission time delay and the predicted current instruction execution time delay are added to obtain the predicted execution lag time of the current control cycle of 31.882 s.

[0022] S2, using a preset irrigation area water-soil model, according to the valve opening degree and environmental parameters of the last cycle, calculating the theoretical predicted value of the soil moisture, and obtaining the model deviation between the theoretical predicted value and the actual value of the current soil moisture; when the model deviation exceeds the first preset threshold, querying the actual running state of the electric valve, and correcting the water-soil model based on the query result.

[0023] The water-soil model is a simplified water balance equation, which calculates the theoretical water inflow according to the valve opening degree of the last cycle, such as 60%, and the current pipe network pressure, and estimates the evaporation amount combined with environmental parameters such as temperature and air humidity, to predict the theoretical value of the current soil moisture, such as 58%; comparing the theoretical value with the actual value collected by the sensor, 55%, to obtain a model deviation of -3%; if the absolute value of the deviation, 3%, exceeds the first preset threshold, such as 2%, an immediate state query instruction is sent to the electric valve; the electric valve returns the actual opening degree through the built-in angle sensor, which may be only 52%, instead of the instruction 60%; using the actual opening degree of 52% to recalculate the theoretical value, if the recalculated theoretical value is close to the actual value, the deviation is attributed to valve execution failure; if there is still a large deviation, the soil infiltration coefficient and other parameters in the water-soil model are adjusted accordingly to be closer to the actual situation.

[0024] In an optional embodiment, when the model deviation exceeds the first preset threshold, the actual running state of the electric valve is queried, and the water-soil model is corrected based on the query result, including: The first preset threshold is set to 5% of the target soil moisture set value; The actual average opening degree and cumulative opening time of the electric valve in the last control cycle are queried. Based on the actual average opening degree and cumulative opening time obtained from the query, and combined with the pipeline pressure, the actual irrigation water volume of the previous cycle is recalculated; Using the actual irrigation water volume as input, the soil infiltration rate parameter in the soil and water model is adjusted through an optimization algorithm to minimize the error between the moisture prediction value of the soil and water model under the input and the actual current soil moisture value.

[0025] Assuming the target soil moisture setting is 40%, the first preset threshold is 2%. At the end of a certain control cycle, the soil moisture predicted by the water and soil model is 37.5%, while the actual value measured by the sensor is 40.2%, a deviation of -2.7%. Since the absolute value of this deviation exceeds the preset threshold of 2%, a correction process will be initiated. A query command is sent to the electric valve, which reports an actual average opening degree of 45% and a cumulative opening time of 600 seconds in the previous control cycle. Combining this with the recorded average pipeline pressure of 2.2 MPa and the valve flow characteristic curve, the actual irrigation water volume for the previous cycle is recalculated to be 5.5. Instead of the 5.0 originally estimated by the model. At 5.5 Using verified actual irrigation water volume as input, optimization algorithms such as gradient descent are applied to adjust the soil and water model. The algorithm iteratively modifies the soil infiltration rate parameter in the model, for example, adjusting it from 0.012 to 0.011, until the model uses the new parameter and 5.5. The humidity prediction calculated from the irrigation amount is closest to the actual measured value of 40.2%, thus completing the calibration of the soil and water model.

[0026] S3, based on the current target difference between the current actual soil moisture value and the target soil moisture setpoint, the target difference of the previous cycle, the predicted execution lag time, and the model deviation, calculate the valve opening adjustment amount for the next control cycle; the predicted execution lag time is used to determine the control gain attenuation coefficient, the model deviation amount is used to determine the adjustment magnitude weight, and the control gain attenuation coefficient and the adjustment magnitude weight together determine the valve opening adjustment amount.

[0027] Specifically, the current target difference is 10%, and the target difference in the previous cycle was 12%. The predicted execution lag time is 210 seconds. A small control gain attenuation coefficient, such as 0.7, is calculated based on a preset function to avoid excessive adjustment due to long lag. A model deviation of -3% indicates that the actual water volume is less than expected. Therefore, an adjustment magnitude weight greater than 1, such as 1.2, is calculated to strengthen the adjustment. The valve opening adjustment is calculated using a PD-like algorithm. The base value is determined by the current target difference and the change in the difference. Then, it is multiplied by the control gain attenuation coefficient of 0.7 and the adjustment magnitude weight of 1.2 to calculate a specific opening adjustment value, such as increasing the opening by 6%.

[0028] In an optional embodiment, the predicted execution lag time is used to determine a control gain attenuation coefficient, comprising: The control gain attenuation coefficient is calculated by the following formula : , wherein, is the predicted execution lag time, in s; is a preset attenuation constant; and exp is a natural exponential function.

[0029] Specifically, in a control cycle, the predicted execution lag time is calculated to be 31.882 s. A preset attenuation constant = 0.15 s.

[0030] According to the given formula, the calculation result of the control gain attenuation coefficient is about 0.0084. The control gain attenuation coefficient far less than 1 will be used in the subsequent control algorithm, which will weaken the effect of the integral term, thereby avoiding excessive irrigation water overshoot caused by too aggressive control action when there is a large delay in system response.

[0031] In an optional embodiment, the model deviation amount is used to determine an adjustment amplitude weight, comprising: The adjustment amplitude weight W is calculated by the following formula: , wherein, is the absolute value of the model deviation amount, in %; is a preset weight influence factor.

[0032] Suppose the soil moisture predicted by the water-soil model is 38%, and the actual measured value of the sensor is 40.7%, then the model deviation amount is -2.7%. The absolute value is 2.7. The value of the preset weight influence factor is 0.5. The numerical value is substituted into the formula for calculation, and the calculation result of the adjustment amplitude weight W is about 0.4255, indicating that due to the current model prediction is not accurate enough, the subsequently calculated valve opening adjustment amount will be multiplied by the weight for reduction, so that the adjustment action of the control system is more gentle and conservative, and the risk of false adjustment caused by inaccurate model is reduced.

[0033] In an optional embodiment, based on the current target difference between the current soil moisture actual value and the target soil moisture set value, the target difference of the last cycle, the predicted execution lag time, and the model deviation amount, the valve opening adjustment amount of the next control cycle is calculated, comprising: The valve opening adjustment amount is calculated by the following improved incremental proportional-integral control algorithm: : wherein W is an adjustment amplitude weight; Kp is a proportional gain; Ki is an integral gain; is a current target difference; is a last period target difference; Kc is a control gain attenuation coefficient.

[0034] In one control instance, the target soil humidity is 45%, the current actual humidity is 40.7%, thus the current target difference is 4.3. The actual humidity of the last period is 40%, thus the last period target difference is 5.0. According to the above calculation, the adjustment amplitude weight W is 0.4255, and the control gain attenuation coefficient Kc is 0.0084. The set proportional gain Kp is 1.2, and the integral gain Ki is 0.3. Substituting all the above values into the improved incremental PI control algorithm formula, the proportional part is calculated to be -0.84, and the integral part is calculated to be about 0.0108. Adding the two parts together, the result is -0.8292, and multiplying the result by the adjustment amplitude weight 0.4255, the calculated valve opening adjustment amount is about -0.35. It indicates that in the next control period, the valve opening needs to be reduced by 0.35% than the current opening.

[0035] S4, generating a control instruction with the target opening being the sum of the current valve opening and the valve opening adjustment amount, and combining the predicted execution lag time to set the expected execution time of the control instruction, and sending it to the electric valve; judging whether the absolute value of the current target difference is less than the second preset threshold value, and whether the moving average value of the target difference change rate in the continuous preset number of control periods is less than the change rate threshold value, the change rate threshold value being set according to the predicted execution lag time, when both conditions are met, ending the current irrigation control.

[0036] Assuming the current valve opening is 60%, the above calculated adjustment is 6%, and the new target opening is 66%; generate a control instruction containing the target opening of 66%; according to the predicted execution lag time of 210s, set the expected execution time of the instruction as the current time plus 210s; package the complete instruction containing the target opening and the future execution timestamp, and send it to the on-site controller of the electric valve through the network; the controller will wait for the local time to reach the timestamp before executing the opening adjustment action. Check the current target difference, for example, the difference between the target 65% and the actual 64.5% is 0.5%, which is less than the second preset threshold, such as 1%; record the change rate of the target difference in the last 5 control periods, and calculate the moving average; the change rate threshold will be adjusted according to the predicted execution lag time, the longer the lag time, the smaller the change rate threshold set, the higher the stability requirement; for example, a long lag of 210s corresponds to a very small change rate threshold, such as 0.02% / min; when the calculated moving average is less than the threshold, it indicates that the soil humidity has entered a very stable convergence state; at this time, both conditions are met, and it is determined that the irrigation is complete, and an instruction is sent to completely close the electric valve.

[0037] In an optional embodiment, after setting the expected execution time of the control instruction in combination with the predicted execution lag time, it is sent to the electric valve, including: Obtain the current system timestamp of the control terminal ; Add the predicted execution lag time to the current system timestamp to obtain the expected execution time ; Package the control instruction data packet containing the target opening and the expected execution time , and send it to the electric valve through the wireless communication module.

[0038] Specifically, the control terminal obtains the current system timestamp , for example, the time is 14:30:05. The predicted execution lag time of this control instruction has been calculated as 31.882s. Add the two times to calculate the expected execution time of the instruction, which is 14:30:36.882. Package the calculated target valve opening, for example, 49.65%, and the expected execution time into a control instruction data packet. The control instruction data packet is sent to the remote electric valve through the wireless communication module, such as the NB-IoT module. After receiving the instruction, the electric valve will not execute immediately, but will wait for the internal clock to reach 14:30:36.882 before starting to accurately execute the opening adjustment action.

[0039] In an optional embodiment, the change rate threshold is set according to a predicted execution lag time, comprising: The change rate threshold is calculated by using the following linear function : Wherein, The unit of is humidity% / min; is a basic change rate threshold; is a lag influence coefficient; is a predicted execution lag time, with the unit of s.

[0040] Specifically, assuming that the set basic change rate threshold is 0.05% / min, and the lag influence coefficient is 0.00033s per minute per percentage point. In the current control period, the calculated predicted execution lag time is 31.882s.

[0041] According to the formula, the basic change rate threshold is added to the product of the lag influence coefficient and the predicted execution lag time. The result is about 0.06052. The change rate threshold of the current control period is set to 0.06052% / min. It is illustrated that when the system delay is large, the tolerance to the sudden change of soil humidity data will be improved. Only when the humidity change rate exceeds a higher threshold, it is determined to be abnormal and triggers the corresponding processing, thereby avoiding overreaction to normal but slightly intense humidity changes due to information delay.

[0042] The embodiment of the intelligent remote control system of the irrigation electric valve provided by the application comprises: The intelligent remote control system of the irrigation electric valve comprises a processor and a memory, and the memory stores computer program instructions. When the computer program instructions are executed by the processor, the intelligent remote control method of the irrigation electric valve is realized.

[0043] The intelligent remote control system of the irrigation electric valve further comprises a communication interface and other components known to those skilled in the art. The settings and functions of the components are known in the art, and thus will not be described here.

[0044] In this disclosure, a "storage medium" or "computer readable medium" can be any available medium that can be accessed by a general purpose or special purpose computer system. By way of example, and not limitation, such computer readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code means in the form of computer readable instructions or data structures and that can be accessed by a general purpose or special purpose computer system, or a combination thereof. When information is transferred or provided over a network or another communications connection (either hardwired, wireless, or combination thereof) to a computer, the computer properly views the connection as a computer readable medium. Thus, any such connection is properly termed a computer readable medium. Combinations of the above should also be included within the scope of the computer readable media.

[0045] The above are preferred embodiments of the present application, not to limit the protection scope of the present application, therefore: any equivalent changes made according to the structure, shape, principle of the present application, should be covered within the protection scope of the present application.

Claims

1. A method of intelligent remote control of an irrigation electric valve, characterized in that, The method comprises the following steps: S1, obtaining a target soil moisture setting value and a current soil moisture actual value of an irrigation area; Based on the network communication quality parameters between the remote control terminal and the electric valve, the pipe network pressure sensor data and the historical instruction execution data, the predicted execution lag time of the current control period is calculated; S2, using a preset water-soil model, the theoretical prediction value of the soil moisture is calculated according to the valve opening degree and the environmental parameters of the last period, and the model deviation between the theoretical prediction value and the current soil moisture actual value is obtained; when the model deviation exceeds the first preset threshold, the actual running state of the electric valve is queried, and the water-soil model is corrected based on the query result; S3, based on the current target difference between the current soil moisture actual value and the target soil moisture setting value, the target difference of the last period, the predicted execution lag time and the model deviation, the valve opening degree adjustment amount of the next control period is calculated; The predicted execution lag time is used to determine the control gain attenuation coefficient, and the model deviation is used to determine the adjustment amplitude weight; S4, a control instruction with the target opening degree being the sum of the current valve opening degree and the valve opening degree adjustment amount is generated, and after the expected execution time of the control instruction is set in combination with the predicted execution lag time, the control instruction is sent to the electric valve; whether the absolute value of the current target difference is less than the second preset threshold and whether the moving average value of the target difference change rate in the continuous preset number of control periods is less than the change rate threshold are judged, and when the two conditions are met at the same time, the current irrigation control is ended.

2. The method of claim 1, wherein, Based on the network communication quality parameters between the remote control terminal and the electric valve, the pipe network pressure sensor data and the historical instruction execution data, the predicted execution lag time of the current control period is calculated, which comprises: Based on the historical instruction issue time stamp, the instruction receiving time stamp and the instruction execution completion time stamp fed back by the electric valve, and the pipe network pressure data during execution, a relationship model between the instruction execution time delay and the pipe network pressure is established; The network data packet round trip time and the packet loss rate between the terminal and the electric valve at the current time are obtained, and the predicted transmission time delay under the current network state is calculated; The current pipe network pressure sensor data is obtained, and the current instruction execution time delay is predicted by using the relationship model; The predicted transmission time delay and the predicted current instruction execution time delay are added to obtain the predicted execution lag time of the current control period.

3. The method of claim 1, wherein, When the model deviation exceeds the first preset threshold, the actual running state of the electric valve is queried, and the water-soil model is corrected based on the query result, which comprises: The first preset threshold is set to 5% of the target soil moisture setting value; The actual average opening degree and the cumulative opening time of the electric valve in the last control period are queried; According to the queried actual average opening degree and the cumulative opening time, the actual irrigation water quantity of the last period is recalculated in combination with the pipe network pressure; The actual irrigation water quantity is taken as input, the soil infiltration rate parameter in the water-soil model is adjusted by using an optimization algorithm, so that the error between the humidity prediction value of the water-soil model under the input and the current soil moisture actual value is minimized.

4. The method of claim 1, wherein, The predicted execution lag time is used to determine the control gain attenuation coefficient, which comprises: The control gain attenuation coefficient is calculated by the following formula : , wherein, is a predicted execution lag time; is a preset attenuation constant; and exp is a natural exponential function.

5. The method of intelligent remote control of irrigation motorized valves according to claim 1, characterized in that, The model deviation is used to determine the adjustment amplitude weight, which comprises: The adjustment amplitude weight W is calculated by using the following formula: wherein, is an absolute value of the model bias amount; is a preset weight influence factor.

6. The method of intelligent remote control of irrigation motorized valves according to claim 1, characterized in that, Based on the current target difference between the current soil moisture actual value and the target soil moisture set value, the last period target difference, the predicted execution lag time and the model deviation, the valve opening adjustment amount of the next control period is calculated, including: The valve opening adjustment amount is calculated by the following improved incremental proportional-integral control algorithm : wherein W is an adjustment amplitude weight; is a proportional gain; is an integral gain; is a current target difference value; is a target difference value of the last period; is a control gain attenuation coefficient.

7. The method of intelligent remote control of irrigation motorized valves according to claim 1, characterized in that, After the expected execution time of the control instruction is set according to the predicted execution lag time, the control instruction data packet is sent to the electric valve, including: Acquiring a current system timestamp of the control terminal ; Predict execution lag time Compared with the current system timestamp Add them together to get the expected execution time. ; This will include the target opening size and the expected execution time. The control command data packet is encapsulated and sent to the electric valve through the wireless communication module.

8. The method of intelligent remote control of irrigation motorized valves according to claim 1, characterized in that, In S4, the change rate threshold is set according to the predicted execution lag time, including: The rate of change threshold is calculated using the following linear function : wherein is the base rate of change threshold; is a hysteresis influence coefficient; is a prediction execution hysteresis time.

9. The method of claim 7, wherein the method further comprises: The control instruction data packet is sent to the electric valve through the NB-IoT module.

10. An intelligent remote control system for irrigation electrically operated valves, characterized in that, The irrigation electric valve intelligent remote control method comprises a memory and a processor, and computer program instructions are stored in the memory. When the computer program instructions are executed by the processor, the irrigation electric valve intelligent remote control method of any one of claims 1-9 is realized.

Images