Seedling raising water and fertilizer accurate management and control method and system based on artificial intelligence

Abstract

The invention belongs to the technical field of data processing, and particularly relates to a seedling raising water and fertilizer accurate management and control method and system based on artificial intelligence, and the method comprises the steps: determining the growth significance of seedlings in each period; according to the difference of the trend change moments between different dimensions, aligning the environment data of each dimension in pairs according to the trend change moments; determining the abnormal degree of each environmental data in the target dimension according to the influence degree of each environmental data in the target dimension on the correlation between the target dimension and other dimensions; determining the correction necessity of each environmental data according to the growth significance of each environmental data in the period in the target dimension and the abnormal degree of each environmental data; according to the correction necessity, screening the to-be-corrected data for correction; and determining the water and fertilizer requirements of the crops according to the corrected environment data of each dimension. The method improves the prediction precision of the water and fertilizer demands, thereby promoting the healthy growth of crops.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing. More specifically, the present invention relates to a precise control method and system for seedling cultivation water and fertilizer based on artificial intelligence. Background Art

[0002] With the continuous growth of the global population and the increasing agricultural demand, traditional agricultural planting models face a series of challenges, including waste of resources, environmental pollution, and uncertainties brought about by climate change. This has prompted agricultural production to develop in a more scientific, refined, and sustainable direction. Especially in the seedling cultivation stage, precise control of water and fertilizer management is crucial for the growth and development of plants.

[0003] Traditional water and fertilizer management methods often rely on experience and extensive management models, resulting in waste of resources and decline in soil fertility. For example, excessive fertilization not only has a negative impact on crops but may also cause environmental problems such as soil acidification and water eutrophication.

[0004] In modern agriculture, sensors and Internet of Things technologies can be used to monitor environmental data in real time during the seedling cultivation process. Based on the changes in environmental data during the seedling cultivation process, the water and fertilizer requirements of seedlings can be predicted to achieve automated irrigation of water and fertilizer. However, during the process of sensor collection of environmental data, affected by factors such as electromagnetic and weather conditions, the collected environmental data may have noise, which affects the accuracy of the prediction results of the water and fertilizer requirements of seedlings, and further affects the growth and yield of crops. Summary of the Invention

[0005] To solve the technical problem that noise in the above environmental data affects the accuracy of predicting the water and fertilizer requirements of seedlings, the present invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides a precise control method for seedling cultivation water and fertilizer based on artificial intelligence, including: Determining the growth significance of each period of seedlings according to the growth conditions of seedlings; collecting environmental data in multiple dimensions; determining multiple trend change moments of environmental data in each dimension; aligning environmental data in each dimension pairwise according to the differences in trend change moments between different dimensions; taking any one dimension as the target dimension, and determining the abnormality degree of each environmental data in the target dimension according to the influence degree of each environmental data in the target dimension on the correlation between the target dimension and the remaining dimensions; determining the necessity of correction of each environmental data according to the growth significance of the period where each environmental data in the target dimension is located and the abnormality degree of each environmental data; screening the data to be corrected according to the size of the correction necessity, and correcting the data to be corrected; determining the water and fertilizer requirements of crops according to the corrected environmental data in each dimension.

[0007] Preferably, determining the growth significance of the seedlings in each period according to the growth conditions of the seedlings includes: obtaining the growth amount of the seedlings in each period, and taking the ratio of the average growth amount of all the seedlings in the same period to the maximum value of the growth amounts of all the seedlings in all periods as the growth significance of the corresponding period.

[0008] Preferably, determining multiple trend change moments of the environmental data in each dimension includes: for any dimension, smoothing the data curve of this dimension, and obtaining the moments corresponding to all the maximum values and all the minimum values in the smoothed data curve, and respectively taking them as a trend change moment of this dimension.

[0009] Preferably, aligning the environmental data in each dimension pairwise according to the differences in the trend change moments between different dimensions includes: for any two dimensions, taking the average time difference between each trend change moment in the first dimension and the nearest trend change moment in the second dimension as the misalignment length, denoted as L; in response to the misalignment length being a positive number, removing the first L environmental data in the environmental data of the first dimension and removing the last L environmental data in the environmental data of the second dimension, and in response to the misalignment length being a negative number, removing the last environmental data in the environmental data of the first dimension and removing the first environmental data in the environmental data of the second dimension to achieve the alignment of the environmental data of these two dimensions.

[0010] Preferably, determining the abnormality degree of each environmental data in the target dimension includes: obtaining the correlation after aligning the environmental data of the target dimension and the other dimensions as the first correlation; taking any environmental data in the target dimension as the target data, removing the target data in the target dimension, removing the corresponding environmental data of the target data in the other dimensions, and calculating the correlation of the environmental data of the target dimension and the other dimensions after removal as the second correlation; taking the difference between the first correlation and the second correlation as the influence degree of the target data on the correlation between the target dimension and the other dimensions; obtaining the weighted average value of the influence degree of the target data on the correlation between the target dimension and the other dimensions as the abnormality degree of the target data.

[0011] Preferably, when obtaining the weighted average of the influence degree of the target data on the correlation between the target dimension and the remaining dimensions, the method for obtaining the weights of the remaining dimensions is as follows: in the data sequence following the target dimension and the remaining dimensions, the absolute value of the difference between the environmental data corresponding to each trend change moment and the environmental data corresponding to the previous trend change moment is used to form the trend change amount sequence of the target dimension; in the data sequence following the remaining dimensions and the target dimension, the absolute value of the difference between the environmental data corresponding to each trend change moment and the environmental data corresponding to the previous trend change moment is used to form the trend change amount sequence of the remaining dimensions; the DTW distance between the trend change amount sequence of the target dimension and the trend change amount sequence of the remaining dimensions is obtained, and the DTW distance is negatively correlated and normalized, and the result of the negative correlation normalization is used as the weights of the remaining dimensions.

[0012] Preferably, the necessity for correction satisfies the expression: ; where represents the necessity for correction of the k-th environmental data in the target dimension; represents the degree of abnormality of the k-th environmental data in the target dimension; represents the growth significance of the period in which the k-th environmental data in the target dimension is located; exp( ) is the exponential function with the natural exponent as the base.

[0013] Preferably, the correction of the data to be corrected includes: performing polynomial fitting on the local data of the data to be corrected, and using the fitting value of the data to be corrected in the fitting result as the corrected value of the data to be corrected; where, when performing polynomial fitting, the fitting weights of the local data are set according to the degree of abnormality of each local data, and the fitting weights of the local data are negatively correlated with the degree of abnormality of each local data.

[0014] Preferably, determining the water and fertilizer requirements of the crop according to the corrected environmental data of each dimension includes: replacing the data to be corrected in the environmental data of each dimension with the corrected environmental data to update the environmental data of each dimension, and inputting the updated environmental data of each dimension into the trained neural network to output the water and fertilizer requirements of the crop.

[0015] In a second aspect, the present invention provides an artificial intelligence-based seedling raising water and fertilizer precise control system, including a processor and a memory, and the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above artificial intelligence-based seedling raising water and fertilizer precise control method is implemented.

[0016] By adopting the above technical solution, the above artificial intelligence-based seedling raising water and fertilizer precise control method is generated into a computer program and stored in the memory to be loaded and executed by the processor, so as to manufacture a terminal device according to the memory and the processor, which is convenient to use.

[0017] The beneficial effects of the present invention are as follows: According to the differences in the moments of trend changes between different dimensions, the environmental data of each dimension are aligned, avoiding the interference of the time-delay changes of the environmental data of each dimension on the correlation analysis between different dimensions, and making the correlation analysis of the trend changes of the environmental data between different dimensions more accurate; The present invention determines the abnormality degree of each environmental data according to the influence degree of each environmental data in the target dimension on the correlation between the target dimension and the remaining dimensions, and can accurately distinguish noise from normal data; The present invention determines the necessity of correction for each environmental data according to the growth significance of the period in which each environmental data in the target dimension is located and the abnormality degree of each environmental data, screens and corrects the correction data according to the necessity of correction, improves the accuracy of the environmental data in the important stage of seedling growth, provides a data basis for predicting the water and fertilizer requirements of crops, improves the prediction accuracy of water and fertilizer requirements, and thus ensures the healthy growth of crops. Description of the Drawings

[0018] Figure 1 is a flowchart schematically showing a method for precise control of seedling raising water and fertilizer based on artificial intelligence in the present invention; Figure 2 is a flowchart schematically showing step S3 of a method for precise control of seedling raising water and fertilizer based on artificial intelligence in the present invention. Detailed Embodiments

[0019] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0020] Next, the detailed embodiments of the present invention will be described in detail in conjunction with the drawings.

[0021] An embodiment of the present invention discloses a method for precise control of seedling raising water and fertilizer based on artificial intelligence. Referring to Figure 1 , it includes steps S1 - S6: S1. Determine the growth significance of each period of the seedling according to the growth situation of the seedling.

[0022] It should be noted that the growth situations of the seedling in different periods are different. For example, affected by various factors such as light, the growth rates of the seedling during the day and at night are different. Therefore, the present invention analyzes the growth situations of the seedling in different periods to determine the growth significance of each period of the seedling.

[0023] Specifically, starting from after the seeds are sown until the end of the seedling stage and entering the growth stage, every T hours is a period. Obtain the growth amount of the seedlings in each period, and take the ratio of the average growth amount of all seedlings in the same period to the maximum value of the growth amounts of all seedlings in all periods as the growth significance of the corresponding period.

[0024] Implementers can set the value of T according to the actual implementation situation. For example, T = 6.

[0025] In one embodiment, the method for obtaining the growth amount of the seedlings in each period is as follows: After the seeds germinate, the staff measures the height of the seedlings every T hours, and takes the difference between the current measurement result and the previous measurement result as the growth amount of the seedlings in the corresponding period.

[0026] In another embodiment, the method for obtaining the growth amount of the seedlings in each period is as follows: Set up a camera on one side of the nursery or hotbed. Starting from after the seeds are sown until the end of the seedling stage and entering the growth stage, use the camera to horizontally take time-lapse photos of the entire growth process of the seedlings. Perform motion filtering on the captured video, and obtain the heights of the seedlings at the beginning and end of each period according to the video after motion filtering. Take the height difference between the end and the beginning of each period as the growth amount of the seedlings in each period.

[0027] It should be noted that the present invention performs motion filtering on the growth video of the seedlings, which can eliminate the short-term jitter of the leaves in the video and highlight the growth process of the seedlings, making the height of the seedlings obtained based on the video more accurate.

[0028] S2. Collect environmental data in multiple dimensions.

[0029] Use the environmental temperature and humidity sensors, light intensity sensors, soil temperature and humidity sensors, etc. deployed in the seedling raising environment to collect environmental data in the seedling raising environment in real time, including environmental temperature, environmental humidity, light intensity, soil humidity, soil temperature, etc.

[0030] In the present invention, the collection frequency is once every 30 minutes, and implementers can set the collection frequency according to the actual implementation situation.

[0031] S3. Determine the abnormality degree of each environmental data in each dimension.

[0032] The flowchart of step S3 refers to Figure 2 , including steps S301 to S303, specifically as follows: S301. Determine multiple trend change moments of the environmental data in each dimension.

[0033] Specifically, for any dimension, with time as the horizontal axis and the environmental data of that dimension as the vertical axis, a data curve of that dimension is plotted. The data curve of that dimension is smoothed, and the times corresponding to all the maximum values and the times corresponding to all the minimum values in the smoothed data curve are obtained, and are respectively used as a trend change time of that dimension.

[0034] S302. According to the differences in the trend change times between different dimensions, align the environmental data of each dimension pairwise according to the trend change times.

[0035] It should be noted that the environmental data of each dimension affect each other. For example, when the light intensity increases, the environmental temperature and the soil temperature increase, and the environmental humidity and the soil humidity decrease. Therefore, the change of the environmental data of one dimension may cause the change of the environmental data of other dimensions, but this change has a time delay, and the time delay degrees of the changes of the environmental data of different dimensions are different. For example, when the light intensity increases, the environmental temperature at the next moment is affected by the light intensity and increases, but after multiple moments, the soil temperature will be affected by the light intensity and increase. The trend change times of each dimension obtained in step S301 are the times when the environmental data of each dimension change due to the influence of the environmental data of other dimensions. The present invention aligns the environmental data of different dimensions according to the differences in the trend change times between different dimensions.

[0036] Specifically, for any two dimensions, the average time difference between each trend change time in the first dimension and the nearest trend change time in the second dimension is used as the misalignment length, denoted as L. For example, when the 1st, 5th, 15th, and 20th times in the first dimension are trend change times, and the 2nd, 6th, 16th, and 21st times in the second dimension are trend change times, the misalignment length is -1.

[0037] When the misalignment length is a positive number, the first L environmental data in the environmental data of the first dimension are removed, and the last L environmental data in the environmental data of the second dimension are removed. On the contrary, when the misalignment length is a negative number, the last environmental data in the environmental data of the first dimension are removed, and the first environmental data in the environmental data of the second dimension are removed, so as to align the environmental data of these two dimensions.

[0038] By performing the above operations pairwise for all dimensions, pairwise alignment of all dimensions can be achieved.

[0039] S303. Take any one dimension as the target dimension, and determine the abnormality degree of each environmental data in the target dimension according to the influence degree of each environmental data in the target dimension on the correlation between the target dimension and the remaining dimensions.

[0040] It should be noted that the delay degrees of environmental data in different dimensions affected by environmental data in the same dimension vary, but their change time lengths are similar, and their change trends are the same or opposite. Therefore, after the environmental data of two dimensions are aligned, the environmental data of these two dimensions are positively or negatively correlated. When there is noise in the environmental data of one dimension, it will cause the correlation between the environmental data of these two dimensions to deteriorate. Therefore, in the present invention, the corresponding environmental data are excluded in two dimensions, and by comparing the change in the correlation between the two dimensions before and after the exclusion, the influence degree of the excluded environmental data on the correlation between the two dimensions is determined, so as to obtain the abnormality degree of the excluded environmental data.

[0041] Specifically, calculate the correlation after aligning the environmental data of the target dimension with the environmental data of the remaining dimensions as the first correlation.

[0042] Take any environmental data in the target dimension as the target data, exclude the target data in the target dimension, exclude the corresponding environmental data of the target data in the remaining dimensions, and calculate the correlation between the target dimension and the environmental data of the remaining dimensions after the exclusion as the second correlation.

[0043] Implementers can set the correlation algorithm according to the actual implementation situation, such as the cosine similarity algorithm.

[0044] Furthermore, the abnormality degree of the target data satisfies the expression: ; where p represents the abnormality degree of the target data in the target dimension; represents the weight of the h-th dimension outside the target dimension; represents the first correlation between the target dimension and the h-th dimension outside the target dimension; represents the second correlation between the target dimension and the h-th dimension outside the target dimension; N represents the number of dimensions; represents the absolute value symbol; max( ) is the maximum value function, which is used to prevent the abnormality degree of the target data from being negative. When is closer to 1 or -1, is larger, and the target dimension and the h-th dimension outside the target dimension show positive or negative correlation. When is greater than 0, it indicates that after excluding the target data and the corresponding data of the target data, the target dimension and the h-th dimension outside the target dimension are more correlated, indicating that the target data affects the correlation between the target dimension and the h-th dimension outside the target dimension. When The greater the influence of the target data on the correlation between the target dimension and the h-th dimension outside the target dimension, the more inconsistent the target data is with the data relationship between these two dimensions, and the more likely the target data is to be noise; when the influence of the target data on the correlation between the target dimension and all other dimensions is greater, the target data is more abnormal.

[0045] Among them, the weight of the h-th dimension outside the target dimension satisfies the expression: ; Among them, represents the weight of the h-th dimension outside the target dimension; D represents the sequence of trend change amounts of the target dimension; represents the sequence of trend change amounts of the h-th dimension outside the target dimension; represents the DTW distance between the sequence of trend change amounts of the target dimension and the sequence of trend change amounts of the h-th dimension outside the target dimension; N represents the number of dimensions; exp( ) is the exponential function with the natural exponent as the base, used to perform a negative correlation mapping on and is used to perform normalization.

[0046] Among them, the method for obtaining the sequence of trend change amounts is: In the data sequence after aligning the target dimension with the other dimensions, the absolute value of the difference between the environmental data corresponding to each trend change moment and the environmental data corresponding to the previous trend change moment forms the sequence of trend change amounts of the target dimension; in the data sequence after aligning the other dimensions with the target dimension, the absolute value of the difference between the environmental data corresponding to each trend change moment and the environmental data corresponding to the previous trend change moment forms the sequence of trend change amounts of the other dimensions.

[0047] It should be noted that the influence of the change in environmental data of the same dimension on the environmental data of different dimensions is different. For example, when the light intensity increases, it may cause a relatively large increase in the environmental temperature. For example, in winter, the environmental temperature rises from -5°C to 7°C, resulting in a relatively small increase in the soil temperature, for example, from 0°C to 5°C. The elements in the sequence of trend change amounts reflect the local change range of the environmental data of the corresponding dimension. When the change ranges of the environments of two dimensions are more similar, it is more accurate to judge whether the target data is abnormal based on the change in the correlation between these two dimensions before and after removing the target data. Therefore, the present invention determines the weight of the h-th dimension outside the target dimension according to the DTW distance between the sequence of trend change amounts of the target dimension and the sequence of trend change amounts of the h-th dimension outside the target dimension. , the smaller the DTW distance between the trend change amount sequence of the target dimension and the trend change amount sequence of the h-th dimension, the more similar the data change ranges of the target dimension and the h-th dimension are, and the greater the weight of the h-th dimension outside the target dimension.

[0048] S4. Determine the correction necessity of each environmental data according to the growth significance of each environmental data in the target dimension at the corresponding period and the abnormality degree of each environmental data.

[0049] It should be noted that the growth significance of each period of the seedling reflects the key points of each growth period of the seedling. When the growth significance is greater, the corresponding period is a more critical stage for the growth of the seedling. At this time, the control of water and fertilizer needs to be more precise to avoid the influence of too little or too much water and fertilizer on the growth of the seedling. Therefore, in the critical stage of seedling growth, the environmental data of each dimension are required to be more accurate. The abnormality degree of the environmental data reflects the accuracy of the environmental data. When the abnormality degree is greater, it indicates that the environmental data is less accurate due to the influence of noise, and at this time, it is more necessary to correct the environmental data. Therefore, the present invention determines the correction necessity of each environmental data according to the growth significance of each environmental data in the target dimension at the corresponding period and the abnormality degree of each environmental data, so as to screen the environmental data that needs to be corrected according to the correction necessity for correction later.

[0050] Specifically, the correction necessity satisfies the expression: ; where represents the correction necessity of the k-th environmental data in the target dimension; represents the abnormality degree of the k-th environmental data in the target dimension; represents the growth significance of the k-th environmental data in the target dimension at the corresponding period; exp( ) is the exponential function with the natural exponential as the base. The correction necessity of the k-th environmental data in the target dimension is related to the abnormality degree of this environmental data. When the growth significance of the k-th environmental data in the target dimension is greater, the higher the accuracy requirement for this environmental data. At this time, in the form of gamma transformation, is used as the exponent of the abnormality degree to expand the abnormality degree. When the growth significance is greater, the exponent is smaller, and the greater the expansion degree of the abnormality degree, making the correction necessity of the -th environmental data in the target dimension greater.

[0051] S5. Screen the data to be corrected according to the size of the correction necessity, and correct the data to be corrected.

[0052] When the necessity for correction is greater than a preset correction threshold, the corresponding environmental data is taken as the data to be corrected. Herein, the correction threshold is set by the implementer according to the actual implementation situation, for example, 0.3.

[0053] For the data to be corrected, a total of S environmental data before and after it are taken as the local data of the data to be corrected. Herein, S is a preset quantity, which is set by the implementer according to the actual implementation situation, for example, S = 6.

[0054] All local data of the data to be corrected are subjected to polynomial fitting using the least squares method. During the fitting process, the loss is: ; where ε is the loss during the fitting process; is the degree of abnormality of the v-th local data of the data to be corrected; represents the v-th local data of the data to be corrected; f(v) represents the fitting value of the v-th local data of the data to be corrected; S represents the preset quantity; exp( ) represents the exponential function with the natural constant as the base; represents the fitting weight of the v-th local data of the data to be corrected. When the degree of abnormality of the v-th local data is greater, the v-th local data is more likely to be noise, and the fitting weight of the v-th local data is smaller. During polynomial fitting, less attention is paid to the v-th local data, thereby avoiding the influence of noise on the fitting result.

[0055] It should be noted that when performing polynomial fitting in the present invention, the highest degree term of the polynomial is cubic, then , where a, b, c, and d are the coefficients of the polynomial. The partial derivatives of a, b, c, and d are obtained using the loss ε, and each partial derivative result is set equal to 0, thereby solving the coefficients a, b, c, and d, and then the specific fitting function can be obtained. This process is a well-known technique in the least squares method and will not be elaborated herein.

[0056] The fitting value of the data to be corrected in its fitting function is taken as the correction value of the data to be corrected. The data to be corrected in the environmental data of each dimension is replaced with its correction value. Thus, the correction of the environmental data of each dimension is achieved.

[0057] S6. Determine the water and fertilizer requirements of the crop based on the corrected environmental data of each dimension.

[0058] The present invention uses a neural network to determine the water and fertilizer requirements of crops. The specific content of the neural network is as follows: The neural network adopts a fully connected neural network. The input of the neural network is the corrected environmental data in each dimension, including the environmental temperature, environmental humidity, light intensity, soil temperature, soil humidity, etc. at each moment within a time period. The output of the neural network is the water and fertilizer requirements of the crops. The data set of the neural network is the environmental data in each dimension in the historical period, and the label is the water and fertilizer requirements, and the label is set by relevant professionals according to professional experience. The loss function of the neural network is the mean square error loss. It should be noted that in the present invention, a time period is one day, and the implementers can set it according to the actual implementation situation.

[0059] Input the corrected environmental data in each dimension within a time period into the trained neural network to output the water and fertilizer requirements, so that the seedling raising water and fertilizer control system can accurately irrigate and fertilize according to the water and fertilizer requirements.

[0060] The embodiment of the present invention also discloses a seedling raising water and fertilizer precise control system based on artificial intelligence, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, a seedling raising water and fertilizer precise control method according to the present invention is realized.

[0061] The above system also includes other components well known to those skilled in the art such as a communication bus and a communication interface, and their settings and functions are known in the art, so they will not be described in detail here.

[0062] In the description of this specification, the meanings of "a plurality" and "several" are at least two, such as two, three or more, etc., unless otherwise specifically defined.

[0063] Although this specification has shown and described multiple embodiments of the present invention, it is obvious to those skilled in the art that such embodiments are provided only by way of example. Those skilled in the art will think of many changes, alterations and alternative ways without departing from the spirit and scope of the present invention. It should be understood that in the process of practicing the present invention, alternative solutions to the embodiments of the present invention described herein can be adopted.

Claims

1. A method for precise control of water and fertilizer for seedling cultivation based on artificial intelligence, characterized in that: include: Determine the growth significance of seedlings at different stages according to the growth of seedlings; Collect environmental data in multiple dimensions; Determine multiple trend change moments of environmental data in each dimension; align the environmental data in each dimension in pairs according to the trend change moments based on the differences in trend change moments between different dimensions; Take any dimension as the target dimension, and determine the abnormality of each environmental data in the target dimension according to the influence of each environmental data in the target dimension on the correlation between the target dimension and other dimensions; determine the necessity of correction of each environmental data according to the growth significance of the period of each environmental data in the target dimension and the abnormality of each environmental data; select the data to be corrected according to the necessity of correction, and correct the data to be corrected; Determine the water and fertilizer requirements of crops based on the corrected environmental data in each dimension.

2. The method for precise control of water and fertilizer for seedling cultivation based on artificial intelligence according to claim 1, characterized in that: Determining the growth significance of the seedlings at various stages according to the growth of the seedlings comprises: The growth of seedlings in each period was obtained, and the ratio of the average growth of all seedlings in the same period to the maximum growth of all seedlings in all periods was taken as the growth significance of the corresponding period.

3. The method for precise control of water and fertilizer for seedling cultivation based on artificial intelligence according to claim 1, characterized in that: The determining of multiple trend change moments of environmental data in each dimension includes: For any dimension, the data curve of the dimension is smoothed, and the time corresponding to all the maximum values ​​and the time corresponding to all the minimum values ​​in the smoothed data curve are obtained, which are respectively used as a trend change moment of the dimension.

4. The method for precise control of water and fertilizer for seedling cultivation based on artificial intelligence according to claim 1, characterized in that: According to the difference in trend change time between different dimensions, aligning the environmental data of each dimension in pairs according to the trend change time includes: For any two dimensions, the average time difference between each trend change moment in the first dimension and the most recent trend change moment in the second dimension is taken as the dislocation length, denoted as L; in response to the dislocation length being a positive number, the first L environmental data in the environmental data of the first dimension are removed, and the last L environmental data in the environmental data of the second dimension are removed; in response to the dislocation length being a negative number, the last L environmental data in the environmental data of the first dimension are removed. The environmental data of the second dimension are removed, and the front The environmental data of each dimension is removed to achieve the alignment of the environmental data in these two dimensions.

5. The method for precise control of water and fertilizer for seedling cultivation based on artificial intelligence according to claim 1, characterized in that: Determining the abnormality degree of each environmental data in the target dimension includes: Obtain the correlation after alignment between the target dimension and the environmental data of the remaining dimensions as the first correlation; take any environmental data in the target dimension as the target data, remove the target data in the target dimension, remove the environmental data corresponding to the target data in the remaining dimensions, and calculate the correlation between the target dimension and the environmental data of the remaining dimensions after removal as the second correlation; take the difference between the first correlation and the second correlation as the influence of the target data on the correlation between the target dimension and the remaining dimensions; The weighted average of the impact of the target data on the correlation between the target dimension and the remaining dimensions is obtained as the abnormality degree of the target data.

6. The method for precise control of water and fertilizer for seedling cultivation based on artificial intelligence according to claim 5, characterized in that: When obtaining the weighted average of the impact of the target data on the correlation between the target dimension and the remaining dimensions, the weights of the remaining dimensions are obtained as follows: In the data sequence after the target dimension is aligned with the remaining dimensions, the absolute value of the difference between the environmental data corresponding to each trend change moment and the environmental data corresponding to the previous trend change moment constitutes the trend change amount sequence of the target dimension; in the data sequence after the remaining dimensions are aligned with the target dimension, the absolute value of the difference between the environmental data corresponding to each trend change moment and the environmental data corresponding to the previous trend change moment constitutes the trend change amount sequence of the remaining dimensions; The DTW distance between the trend change amount sequence of the target dimension and the trend change amount sequences of the remaining dimensions is obtained, the DTW distance is negatively correlated normalized, and the result of the negative correlation normalization is used as the weight of the remaining dimensions.

7. The method for precise control of water and fertilizer for seedling cultivation based on artificial intelligence according to claim 1, characterized in that: The modification necessity satisfies the expression: ； in, Indicates the necessity of correction of the kth environmental data in the target dimension; Indicates the abnormality of the kth environmental data in the target dimension; It represents the growth significance of the period of the kth environmental data in the target dimension; exp( ) is an exponential function with the natural exponential as the base.

8. The method for precise control of water and fertilizer for seedling cultivation based on artificial intelligence according to claim 1, characterized in that: The step of correcting the data to be corrected comprises: A polynomial fitting is performed on the local data of the data to be corrected, and the fitting value of the data to be corrected in the fitting result is used as the correction value of the data to be corrected; wherein, when performing the polynomial fitting, a fitting weight of each local data is set according to the abnormality degree of each local data, and the fitting weight of each local data is negatively correlated with the abnormality degree of each local data.

9. The method for precise control of water and fertilizer for seedling cultivation based on artificial intelligence according to claim 1, characterized in that: Determining the water and fertilizer requirements of crops based on the corrected environmental data in each dimension includes: The data to be corrected in the environmental data of each dimension is replaced with the corrected environmental data to update the environmental data of each dimension. The updated environmental data of each dimension is input into the trained neural network to output the water and fertilizer requirements of the crops.

10. A precise control system of water and fertilizer for seedling cultivation based on artificial intelligence, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, an artificial intelligence-based precise control method of water and fertilizer for seedling cultivation is implemented according to any one of claims 1 to 9.

Images