Water and fertilizer integrated irrigation control method for sand culture celery planting

Abstract

The invention relates to the technical field of irrigation control, in particular to a sand culture celery planting-oriented water and fertilizer integrated irrigation control method, which comprises the following steps: acquiring a humidity sequence and a flow sequence at each moment; when the humidity data at the current moment is larger than a preset prediction threshold value, humidity data prediction is started, and the prediction process comprises the steps that according to the difference between the discrete degrees of the humidity data before and after all the mutation points and the difference between all the mutation points and the next mutation point, the time lag confidence coefficient of all the mutation points is obtained, and then the lag duration is obtained; acquiring a rising sequence at the current moment; and according to the oscillation characteristic of the rising sequence at the current moment, the number difference and distance of sudden change points between the humidity sequence and the rising sequence, and the dispersion degree of the flow sequence, a smoothing coefficient at the current moment is obtained, then humidity data under the lagging duration after the current moment is predicted, and whether irrigation needs to be stopped or not is judged. According to the method, the irrigation control effect is improved by more accurately predicting the future soil humidity.

Description

Technical Field

[0001] The present application relates to the field of irrigation control technology, and in particular to a water-fertilizer integrated irrigation control method for sand-cultured celery cultivation. Background Art

[0002] Celery, a typical plant with both medicinal and edible properties, has important nutritional and health benefits. As a shallow-rooted crop that thrives on fertilizer, celery is extremely sensitive to water and fertilizer management. If the amount of water and fertilizer used in alternating irrigation does not meet the celery's required levels, it will affect the celery's growth rate and reduce yields. If the amount of water and fertilizer used is much higher than the celery's required levels, it will cause problems such as water and fertilizer waste, reduced plant resistance to pests and diseases, soil salinization, and reduced celery quality. Therefore, it is necessary to construct an integrated water and fertilizer irrigation method based on the irrigation characteristics of celery to ensure that the irrigation flow of water and fertilizer meets the required levels for sand-grown celery, ensuring celery quality and growth rate while saving water and fertilizer.

[0003] When controlling the water and fertilizer irrigation of sand-grown celery in the existing technology, although the nonlinear interference problem of the existing water and fertilizer irrigation system has been taken into account, and the irrigation system has been precisely controlled through control algorithms such as PID control; however, the existing technology has not fully taken into account the large inertia and hysteresis characteristics of the soil. When the sensor detects that the water and fertilizer content in the soil reaches the set threshold, after the irrigation valve is closed, the remaining water and fertilizer in the pipeline will still be irrigated. If the amount of water and fertilizer accumulates, it will lead to excessive irrigation water and fertilizer content in the celery, affecting the quality of the celery; and if the valve is closed before the water and fertilizer content in the soil reaches the threshold, the problem of insufficient irrigation water and fertilizer content may occur, thereby affecting the yield of celery and the effect of irrigation control is poor. Summary of the Invention

[0004] In order to solve the above technical problems, the present application provides a water-fertilizer integrated irrigation control method for sand-cultured celery cultivation to solve the existing problems.

[0005] The water-fertilizer integrated irrigation control method for sand-cultured celery cultivation in this application adopts the following technical solutions:

[0006] One embodiment of the present application provides a water-fertilizer integrated irrigation control method for sand-cultured celery cultivation, the method comprising the following steps:

[0007] Real-time acquisition of humidity data and flow data at each moment in the drip irrigation area, and then obtaining the humidity sequence and flow sequence at the current moment;

[0008] When the humidity data at the current moment is greater than the preset prediction threshold, humidity data prediction begins. The prediction process is as follows:

[0009] S1: Obtain all mutation points in the humidity sequence at the current moment, and obtain the time lag confidence of each mutation point based on the difference between the discrete degrees of all humidity data before each mutation point and the discrete degrees of all humidity data after each mutation point, as well as the difference between each mutation point and the next mutation point; obtain the lag duration based on the maximum value of the time lag confidence of all mutation points;

[0010] S2: Obtain the current rising sequence based on the mutation point with the largest time lag confidence in the humidity sequence at the current moment, and divide the rising sequence into multiple subsequences based on the mutation points in the rising sequence; obtain the rising oscillation coefficient at the current moment based on the discrete degree of the shortest distance between all data in the rising sequence and its fitted straight line at the current moment, the discrete degree of the slope of the fitted straight line corresponding to all subsequences in the rising sequence, and the difference in the number of mutation points between the humidity sequence at the current moment and the rising sequence;

[0011] S3: Based on the degree of discreteness of the flow sequence at the current moment and the distance between the flow sequence and the rising sequence, the irrigation stability coefficient at the current moment is obtained. Combined with the rising oscillation coefficient at the current moment, the smoothing coefficient at the current moment is obtained, and then the humidity data with a lag time after the current moment is predicted, and it is determined whether irrigation needs to be stopped.

[0012] Preferably, the humidity sequence and flow sequence at the current moment refer to sequences composed of humidity data and flow data from the start of irrigation to the current moment in chronological order.

[0013] Preferably, the calculation formula for the time lag confidence of each mutation point is: u =B u ×exp(C u );where A u is the time lag confidence of the u-th mutation point; B u is the first difference of the u-th mutation point; C u is the difference between the u+1th mutation point and the uth mutation point; exp() is an exponential function with the natural constant e as the base; wherein, the first difference of each mutation point is obtained by: calculating the discrete degree of all humidity data before each mutation point in the humidity sequence and the discrete degree of all humidity data after each mutation point respectively; and recording the absolute difference between the two discrete degrees as the first difference of each mutation point.

[0014] Preferably, the lag time is the time between the moment corresponding to the maximum value of the lag confidence of all mutation points in the humidity sequence at the current moment and the moment when irrigation starts.

[0015] Preferably, the rising sequence at the current moment refers to a sequence consisting of all data after the mutation point with the maximum time lag confidence in the humidity sequence at the current moment.

[0016] Preferably, the calculation formula of the rising oscillation coefficient at the current moment is: Wherein, D is the rising oscillation coefficient at the current moment, F is the first discrete degree at the current moment, G is the second discrete degree at the current moment, and H is the difference between the number of mutation points in the humidity sequence at the current moment and the number of mutation points in the rising sequence; wherein, the first discrete degree at the current moment refers to the variance of the shortest distance between all data in the rising sequence at the current moment and its fitted straight line; the second discrete degree at the current moment refers to the variance of the slope of the fitted straight line corresponding to all subsequences in the rising sequence at the current moment.

[0017] Preferably, the calculation formula of the irrigation stability coefficient at the current moment is: Where K is the irrigation stability coefficient at the current moment; L is the variance of all data in the flow sequence at the current moment; W is the DTW distance between the flow sequence at the current moment and the rising sequence; τ is a preset constant.

[0018] Preferably, the calculation formula of the smoothing coefficient at the current moment is: Where α is the smoothing coefficient at the current moment, D is the rising oscillation coefficient at the current moment, K is the irrigation stability coefficient at the current moment, and exp() is an exponential function with the natural constant e as the base.

[0019] Preferably, the specific process of predicting the humidity data at the lag time after the current moment is: using the rising sequence and smoothing coefficient at the current moment as input of the cubic exponential smoothing algorithm, and outputting the humidity prediction value at the lag time after the current moment.

[0020] Preferably, the specific process of determining whether irrigation needs to be stopped is: the lag time is recorded as t seconds, and when the humidity prediction value t seconds after the current moment is greater than or equal to the preset humidity threshold, irrigation needs to be stopped; otherwise, irrigation does not need to be stopped.

[0021] This application has at least the following beneficial effects:

[0022] This application addresses the problem that the existing technology does not fully consider the influence of soil hysteresis and residual water and fertilizer in the pipeline, which leads to excessive or insufficient irrigation. By constructing a time lag confidence, the credibility of the lag time of soil moisture changes compared to the irrigation flow rate can be reflected, providing a time benchmark for subsequent predictions; by constructing an upward oscillation coefficient, the irregularity of the fluctuation amplitude and direction of soil moisture in the rising process can be reflected, and the degree of dependence of the prediction model on historical data can be analyzed; by constructing an irrigation stability coefficient, the stability of irrigation flow and its associated influence on soil moisture changes can be reflected, and the weight adjustment of the construction of the smoothing coefficient can be performed; by constructing a smoothing coefficient to accurately predict soil moisture data, it is possible to accurately predict the irrigation stop time, avoid excessive or insufficient soil moisture, improve the effect of irrigation control, and thus save water and fertilizer and improve celery growth. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0024] Figure 1 A flowchart of the steps of the water-fertilizer integrated irrigation control method for sand-cultured celery cultivation provided in this application;

[0025] Figure 2 This is a flowchart for predicting humidity data after the current moment provided by this application. DETAILED DESCRIPTION

[0026] To further illustrate the technical means and effects employed by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effects of the water-fertilizer integrated irrigation control method for sand-cultured celery cultivation proposed in this application. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0027] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0028] The specific scheme of the water-fertilizer integrated irrigation control method for sand-cultured celery cultivation provided by this application is described in detail below with reference to the accompanying drawings.

[0029] An embodiment of the present application provides a water-fertilizer integrated irrigation control method for sand-cultured celery cultivation. Specifically, the following water-fertilizer integrated irrigation control method for sand-cultured celery cultivation is provided. Figure 1 , the method comprises the following steps:

[0030] Step 1: Obtain the humidity data and flow data at each moment in the drip irrigation area in real time, and then obtain the humidity sequence and flow sequence at the current moment.

[0031] The integrated water and fertilizer machine uses water-soluble fertilizer technology to dissolve fertilizer in water and use drip irrigation to apply fertilizer. Since the planting area is large, the growth effect of celery and the water and fertilizer requirements of each group of planting areas may be inconsistent. Therefore, this application uses alternating irrigation for each group to adjust the water and fertilizer concentration ratio and perform regional alternating irrigation. Since each group of drip irrigation areas has already been proportioned with the required water and fertilizer concentration ratio before drip irrigation, this application uses a soil temperature and humidity sensor to collect the water content in the soil to reflect the irrigation effect of the integrated water and fertilizer machine.

[0032] This application analyzes the drip irrigation area currently being irrigated during alternating irrigation in a sand-grown celery planting area. An electromagnetic flowmeter is used to collect the overall irrigation flow rate of the integrated water and fertilizer machine during the irrigation process. Drip irrigation pipes are neatly arranged in rows. N (5-10 values, 5 in this example) soil temperature and humidity sensors are buried at equal intervals in the soil below each row of drip irrigation pipes to collect soil moisture content. The burial depth is close to the root position of the sand-grown celery. In this example, the burial depth is 10 cm. The implementer can set an appropriate value based on actual conditions.

[0033] The average of the moisture content data collected by all soil temperature and humidity sensors at the same time in the drip irrigation area currently being irrigated is recorded as the humidity data at that moment; the irrigation flow collected at each moment in the drip irrigation area currently being irrigated is recorded as the flow data at each moment.

[0034] All data is collected synchronously and in real time, with a one-second interval, starting when the fertigation machine begins drip irrigation. The humidity and flow data from the start of irrigation to the current moment are sequentially sequenced and recorded as the current humidity and flow sequences.

[0035] In order to eliminate the dimensional influence between the data, the humidity data and the flow data are normalized respectively. The normalization methods include Z-score, maximum value normalization, maximum and minimum value normalization, etc. This embodiment adopts maximum value normalization.

[0036] Since the predicted humidity needs to be compared with the preset humidity threshold later, this application uses the preset maximum humidity value (from the range of 293-305mm, 300mm in this embodiment) as the maximum value for normalization when normalizing the humidity sequence.

[0037] Step 2: When the humidity data at the current moment is greater than the preset prediction threshold, humidity data prediction is started to obtain the humidity prediction value after the current moment.

[0038] When existing growers perform multiple alternating irrigations on sand-grown celery, they typically monitor the temperature and humidity data in each soil group and compare the data with a preset threshold. When the threshold is reached, the solenoid valve is used to stop the irrigation of that group. However, due to the large inertia, nonlinearity, and time lag of soil, irrigation control based solely on the temperature and humidity data in the soil combined with a preset threshold will result in time errors, which in turn affect the growth of sand-grown celery. To achieve better irrigation control, predictions can be made based on the data variation characteristics of soil moisture. By analyzing the lag time between the change in soil moisture and the irrigation moment, the soil moisture in the future can be predicted. When the predicted humidity reaches the preset threshold, the irrigation of that group can be stopped, thereby achieving precise alternating irrigation control and ensuring the growth rate and quality of sand-grown celery.

[0039] Since the data is collected in real time, when the normalized humidity data collected at the current moment is greater than the preset prediction threshold, real-time humidity data prediction begins. In this embodiment, the preset prediction threshold is 0.85, and the implementer can select the value according to the actual situation. Figure 2 This is a flowchart for predicting humidity data after the current moment. The specific prediction process is as follows:

[0040] S1: Obtain all mutation points in the humidity sequence at the current moment, and obtain the time lag confidence of each mutation point based on the difference between the discrete degree of all humidity data before each mutation point and the discrete degree of all humidity data after each mutation point, as well as the difference between each mutation point and the next mutation point; obtain the lag duration based on the maximum value of the time lag confidence of all mutation points.

[0041] Because soil moisture has a lag, when drip irrigation begins, the moisture content at the root of sand-grown celery remains relatively stable, with no change in the water supply. However, as the drip irrigation continues, soil moisture gradually increases. However, because the flow of water and fertilizer within the sand layer is uncertain, soil moisture also fluctuates. Therefore, in-depth analysis of the moisture series is necessary to determine the lag in soil moisture relative to the drip irrigation process.

[0042] The humidity sequence at the current moment is used as the input of the mutation point detection algorithm to obtain all mutation points in the humidity sequence. The mutation point detection algorithm is not limited to the PELT algorithm, the Pettitt algorithm, and the MK algorithm. This embodiment adopts the PELT algorithm.

[0043] All mutation points are sorted according to their position in the humidity sequence. Taking the uth mutation point as an example, the degree of dispersion of all humidity data before and after the uth mutation point in the humidity sequence is calculated. The absolute difference between the two degrees of dispersion is recorded as the first difference at the uth mutation point. The first difference reflects the difference in the change of humidity sequence elements before and after the uth mutation point. The larger the value, the greater the possibility that the humidity data will no longer be stable after the uth mutation point. The degree of dispersion can be calculated using variance, coefficient of variation, standard deviation, and other methods. This embodiment uses variance calculation.

[0044] As a preferred embodiment, the time lag confidence of each mutation point is obtained based on the first difference of each mutation point and the difference between the next mutation point and each mutation point, which is used to characterize whether a significant humidity increase occurs at each mutation point.

[0045] In this embodiment, the time lag confidence of the u-th mutation point is recorded as A u , its specific expression is: A u =B u ×exp(C u );where A u is the time lag confidence of the u-th mutation point; B u is the first difference of the u-th mutation point; C u is the difference between the u+1th mutation point and the uth mutation point; exp() is an exponential function with the natural constant e as its base. The exponential function is used to increase the parameter weight, considering that the difference between the normalized values ​​is small, while also avoiding the problem of a negative second difference. It should be noted that if the uth mutation point is the last mutation point, the difference between the uth mutation point and the previous mutation point is calculated as the second difference of the uth mutation point.

[0046] C u It can reflect whether the humidity data at the u+1 mutation point moment has a large upward change compared to the u mutation point moment. The larger the value, the higher the soil humidity at the u+1 mutation point moment is compared to the u mutation point moment, thus reflecting the greater the possibility of a rapid increase in soil humidity in the current drip irrigation area. uIt can reflect whether the soil moisture in the current drip irrigation area has undergone a significant upward change near the u-th mutation point. The larger the value, the greater the possibility that the soil moisture begins to gradually increase starting from the u-th mutation point.

[0047] The time between the maximum value of the time lag confidence of all mutation points in the humidity sequence at the current moment and the start of irrigation is recorded as the lag time t, thereby representing the lag time of soil moisture at the root position of sand-cultured celery in the drip irrigation area compared with the irrigation of the water-fertilizer integrated machine.

[0048] S2: Obtain the rising sequence at the current moment based on the mutation point with the largest time lag confidence in the humidity sequence at the current moment, and divide the rising sequence into multiple subsequences based on the mutation points in the rising sequence; obtain the rising oscillation coefficient at the current moment based on the discrete degree of the shortest distance between all data in the rising sequence at the current moment and its fitting straight line, the discrete degree of the slope of the fitting line corresponding to all subsequences in the rising sequence, and the difference in the number of mutation points between the humidity sequence at the current moment and the rising sequence.

[0049] Furthermore, after obtaining the lag time, the soil moisture fluctuation at the root position of the sand-cultured celery can be analyzed, and the soil moisture data after the lag time of t seconds can be predicted by combining the data of the humidity rising period and the rising fluctuation. This can determine whether the current drip irrigation area needs to stop irrigation, thereby ensuring the growth of celery while saving water and fertilizer, and realizing precise control of alternating water and fertilizer irrigation.

[0050] Since the flow direction of water and fertilizer in the soil is not fixed and water and fertilizer will penetrate into deeper levels, there will inevitably be fluctuations in the process of rising soil moisture.

[0051] The sequence consisting of all elements following the mutation point with the highest confidence level in the current humidity sequence is recorded as the current rising sequence. A straight line fitting algorithm is used to fit the rising sequence. The shortest distance between each element in the rising sequence and the fitted line is calculated. The variance of all shortest distances is then calculated and recorded as the first degree of dispersion at the current moment. The first degree of dispersion reflects the severity of soil moisture fluctuations during the rising process; larger values ​​indicate greater fluctuations in soil moisture.

[0052] Furthermore, although the first degree of dispersion can reflect the degree of fluctuation of soil moisture during the rising process, it does not reflect the direction of fluctuation. Therefore, it is possible that the soil moisture fluctuates frequently but the degree of dispersion between the shortest distances is small, which requires further analysis.

[0053] The ascending sequence is divided into multiple subsequences based on all the mutation points in the sequence at the current moment. A straight line is fitted to each subsequence using a straight-line fitting algorithm. The variance of the slopes of the fitted lines for all subsequences is then calculated and recorded as the second degree of dispersion at the current moment. The second degree of dispersion reflects the irregularity of the fluctuation direction in the ascending sequence. A larger value indicates a greater frequency of sudden increases in soil moisture, or even a downward trend. This indicates that predictions based on the overall data are less likely to work, and that recent data is more important.

[0054] As a preferred embodiment, the rising oscillation coefficient at the current moment is obtained based on the discrete degree of the shortest distance between all data in the rising sequence at the current moment and its fitting straight line, the discrete degree of the slope of the fitting straight line corresponding to all subsequences in the rising sequence, and the difference in the number of mutation points between the humidity sequence at the current moment and the rising sequence, which is used to characterize the oscillation degree of the soil moisture data in the rising process before the current moment.

[0055] In this embodiment, the rising oscillation coefficient at the current moment is recorded as D, and its specific expression is: Where D is the current rising oscillation coefficient, F is the first discrete degree at the current moment, G is the second discrete degree at the current moment, and H is the difference between the number of mutation points in the humidity sequence at the current moment and the number of mutation points in the rising sequence. It should be noted that the minimum value of H is 1 and cannot be 0.

[0056] H reflects the number of mutation points during a period of stable soil moisture. Smaller values ​​indicate more accurate sensor data collection, thus avoiding the risk of misjudgment due to environmental interference. The rising oscillation coefficient reflects the degree of oscillation in soil moisture data during its rise prior to the current moment. Larger values ​​indicate greater fluctuations in soil moisture during its rise due to uncertainty in the flow of water and fertilizer in the soil. This, in turn, reduces the regularity of rising soil moisture data and makes it less suitable to use long-term historical data for prediction.

[0057] S3: Based on the degree of discreteness of the flow sequence at the current moment and the distance between the flow sequence and the rising sequence, the irrigation stability coefficient at the current moment is obtained. Combined with the rising oscillation coefficient at the current moment, the smoothing coefficient at the current moment is obtained, and then the humidity data with a lag time after the current moment is predicted, and it is determined whether irrigation needs to be stopped.

[0058] Furthermore, integrated fertigation systems first dissolve soluble fertilizers in water before applying the fertilizers. However, since most fertilizers undergo a thermal reaction when dissolved, and different fertilizers have different dissolution rates, mixing them can cause sediment buildup, impacting drip irrigation stability. Therefore, it is necessary to analyze the fluctuation characteristics of irrigation flow and study its impact on the fluctuation during soil moisture rise to achieve more precise alternating irrigation control.

[0059] As a preferred embodiment, the irrigation stability coefficient at the current moment is obtained according to the discrete degree of the flow sequence at the current moment and the distance between the flow sequence and the rising sequence, which is used to characterize the stability of the flow data in the current drip irrigation area at the current moment.

[0060] In this embodiment, the irrigation stability coefficient at the current moment is recorded as K, and its specific expression is: Where K is the irrigation stability coefficient at the current moment; L is the variance of all data in the flow sequence at the current moment; W is the DTW distance between the flow sequence at the current moment and the rising sequence; τ is a preset constant. To avoid the denominator being 0, the rounding range is [0.001, 0.01]. The value has little effect on the calculation and can be ignored. In this embodiment, τ is set to 0.005, and the implementer can choose the value at his own discretion.

[0061] A smaller L indicates a more stable irrigation flow rate during the irrigation process. W reflects whether irrigation flow significantly affects moisture changes; larger values ​​indicate a smaller impact on soil moisture. K reflects the stability of flow data in the current drip irrigation area; smaller values ​​indicate a more unstable irrigation flow rate. The irrigation stability coefficient can be used to supplement the upward oscillation coefficient of soil moisture, further determining whether predictions are based on historical or recent data.

[0062] Furthermore, although soil moisture fluctuates during the irrigation process, it still has a clear upward trend overall. Therefore, this application uses the classic triple exponential smoothing algorithm to predict soil moisture data. In the triple exponential smoothing algorithm, the selection of the smoothing coefficient is key. The larger the smoothing coefficient, the faster the prediction algorithm reacts to the data and the more it focuses on recent data for prediction; the smaller the smoothing coefficient, the more the prediction algorithm focuses on making predictions based on the overall data. If the smoothing coefficient is not selected properly, it will directly affect the prediction accuracy.

[0063] As a preferred embodiment, the smoothing coefficient at the current moment is obtained based on the rising oscillation coefficient and the irrigation temperature coefficient at the current moment. In this embodiment, the smoothing coefficient at the current moment is recorded as α, and its specific expression is: Where α is the smoothing coefficient at the current moment, D is the rising oscillation coefficient at the current moment, K is the irrigation stability coefficient at the current moment, and exp() is an exponential function with the natural constant e as the base.

[0064] When (D / K) is smaller, it means that the irrigation flow at the current moment is more stable, the rising process of soil moisture is more stable, and the water and fertilizer flow are more uniform, so it is easier to predict humidity based on the overall humidity data; conversely, it means that the irrigation flow data is more unstable, the rising process of soil moisture fluctuates more, and it is more necessary to predict humidity based on recent data.

[0065] The current rising sequence and smoothing coefficient are used as inputs to a cubic exponential smoothing algorithm, which outputs a predicted humidity value for the next t seconds after the current moment. When the predicted humidity value t seconds after the current moment is greater than or equal to a preset humidity threshold, the integrated water and fertilizer machine stops irrigation, and the remaining water and fertilizer in the pipeline completes final irrigation of the sand-grown celery in the current drip irrigation area. The preset humidity threshold is the soil humidity data when the soil moisture reaches the irrigation requirement for sand-grown celery. Because this embodiment normalizes the humidity data, the preset humidity threshold in this embodiment is 1.

[0066] Furthermore, for the next group that needs irrigation, the water-fertilizer ratio is readjusted, and then irrigation control is performed in the same way, thereby achieving precise alternating irrigation control of water and fertilizer integration for sand-cultured celery.

[0067] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0068] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0069] The above-described embodiments are only used to illustrate the technical solutions of the present application, and not to limit them. Modifications to the technical solutions described in the aforementioned embodiments, or equivalent replacements of some of the technical features therein, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A water-fertilizer integrated irrigation control method for sand-cultured celery, characterized in that: The method comprises the following steps: Real-time acquisition of humidity data and flow data at each moment in the drip irrigation area, and then obtaining the humidity sequence and flow sequence at the current moment; When the humidity data at the current moment is greater than the preset prediction threshold, humidity data prediction begins. The prediction process is as follows: S1: Obtain all mutation points in the humidity sequence at the current moment, and obtain the time lag confidence of each mutation point based on the difference between the discrete degrees of all humidity data before each mutation point and the discrete degrees of all humidity data after each mutation point, as well as the difference between each mutation point and the next mutation point; obtain the lag duration based on the maximum value of the time lag confidence of all mutation points; S2: Obtain the current rising sequence based on the mutation point with the largest time lag confidence in the humidity sequence at the current moment, and divide the rising sequence into multiple subsequences based on the mutation points in the rising sequence; obtain the rising oscillation coefficient at the current moment based on the discrete degree of the shortest distance between all data in the rising sequence and its fitted straight line at the current moment, the discrete degree of the slope of the fitted straight line corresponding to all subsequences in the rising sequence, and the difference in the number of mutation points between the humidity sequence at the current moment and the rising sequence; S3: Based on the degree of discreteness of the flow sequence at the current moment and the distance between the flow sequence and the rising sequence, the irrigation stability coefficient at the current moment is obtained. Combined with the rising oscillation coefficient at the current moment, the smoothing coefficient at the current moment is obtained, and then the humidity data with a lag time after the current moment is predicted, and it is determined whether irrigation needs to be stopped.

2. The water-fertilizer integrated irrigation control method for sand culture celery planting according to claim 1, characterized in that: The humidity sequence and flow sequence at the current moment refer to sequences composed of humidity data and flow data from the start of irrigation to the current moment in a chronological order.

3. The water-fertilizer integrated irrigation control method for sand culture celery planting according to claim 1, characterized in that: The calculation formula of the time lag confidence of each mutation point is: u =B u ×exp(C u );where A u is the time lag confidence of the u-th mutation point; B u is the first difference of the u-th mutation point; C u is the difference between the u+1th mutation point and the uth mutation point; exp() is an exponential function with the natural constant e as the base; wherein, the first difference of each mutation point is obtained by: calculating the discrete degree of all humidity data before each mutation point in the humidity sequence and the discrete degree of all humidity data after each mutation point respectively; and recording the absolute difference between the two discrete degrees as the first difference of each mutation point.

4. The water-fertilizer integrated irrigation control method for sand culture celery planting according to claim 1, characterized in that: The lag time is the time between the moment corresponding to the maximum value of the lag confidence of all mutation points in the humidity sequence at the current moment and the moment of starting irrigation.

5. The water-fertilizer integrated irrigation control method for sand culture celery planting according to claim 1, characterized in that: The rising sequence at the current moment refers to a sequence consisting of all data after the mutation point with the largest time lag confidence in the humidity sequence at the current moment.

6. The water-fertilizer integrated irrigation control method for sand culture celery planting according to claim 1, characterized in that: The calculation formula of the rising oscillation coefficient at the current moment is: Wherein, D is the rising oscillation coefficient at the current moment, F is the first discrete degree at the current moment, G is the second discrete degree at the current moment, and H is the difference between the number of mutation points in the humidity sequence at the current moment and the number of mutation points in the rising sequence; wherein, the first discrete degree at the current moment refers to the variance of the shortest distance between all data in the rising sequence at the current moment and its fitted straight line; the second discrete degree at the current moment refers to the variance of the slope of the fitted straight line corresponding to all subsequences in the rising sequence at the current moment.

7. The water-fertilizer integrated irrigation control method for sand culture celery planting according to claim 1, characterized in that: The calculation formula of the irrigation stability coefficient at the current moment is: Where K is the irrigation stability coefficient at the current moment; L is the variance of all data in the flow sequence at the current moment; W is the DTW distance between the flow sequence at the current moment and the rising sequence; τ is a preset constant.

8. The water-fertilizer integrated irrigation control method for sand culture celery planting according to claim 1, characterized in that: The calculation formula of the smoothing coefficient at the current moment is: Where α is the smoothing coefficient at the current moment, D is the rising oscillation coefficient at the current moment, K is the irrigation stability coefficient at the current moment, and exp() is an exponential function with the natural constant e as the base.

9. The water-fertilizer integrated irrigation control method for sand culture celery planting according to claim 1, characterized in that: The specific process of predicting the humidity data at the lag time after the current moment is: using the rising sequence and smoothing coefficient at the current moment as input of the triple exponential smoothing algorithm, and outputting the humidity prediction value at the lag time after the current moment.

10. The water-fertilizer integrated irrigation control method for sand culture celery planting according to claim 1, characterized in that: The specific process of determining whether irrigation needs to be stopped is as follows: the lag time is recorded as t seconds. When the humidity prediction value at the tth second after the current moment is greater than or equal to the preset humidity threshold, irrigation needs to be stopped; otherwise, irrigation does not need to be stopped.

Images