Astragalus cultivation control method and system
By sampling the growth data of Astragalus in the Astragalus planting control system, the growth status vector is constructed and the planting environment parameter model is optimized, the problem of uneven growth of Astragalus is solved and the yield and quality of Astragalus is improved.
Patent Information
- Application Number
- CN202410951258.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-16
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-07-16
AI Technical Summary
The prior art cannot accurately control planting environmental parameters based on the real-time growth of Astragalus, resulting in uneven growth and low yield of Astragalus.
By sampling the seedling height, main root length and lateral root number of Astragalus in each control cycle, a growth condition vector is constructed, and the planting environment parameter model needs to be trained based on the vector, and the model is optimized through the loss function to accurately adjust the planting environment.
The accurate assessment of the growth status of Astragalus and the continuous optimization of the planting environment are achieved, the yield and quality of Astragalus are improved, growth abnormalities are discovered in a timely manner, and the accuracy and training efficiency of the planting environment parameter model are improved.
Smart Images

Figure CN118759890B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of planting control, and in particular to an astragalus planting control method and system. Background Art
[0002] In the related art, CN115316212A discloses a method for planting astragalus with high astragaloside content, including the following steps: S1. Planting astragalus normally until the seedling height is 10 cm; S2. Drought stress; S3. Watering and fertilizing. The roots of astragalus begin to grow rapidly in August, and watering and fertilizing begin at this time to provide sufficient nutrients for the growth of astragalus, and watering and fertilizing are carried out 4 times in total; S4. Harvest before the astragalus sprouts in the spring of the second year, and increase the astragaloside content in astragaloside by cold stress. In this scheme, water is controlled in the seedling stage, water and fertilizer are controlled in the middle stage, and the harvesting time is delayed; by simulating astragalus grown in a natural environment, the astragaloside content is increased to achieve the effect of approaching the astragaloside content in wild astragalus.
[0003] CN111602568A relates to a method for planting astragalus in high-cold areas, and belongs to the technical field of Chinese medicinal material planting methods. In order to solve the problems of low autumn sowing greening rate and slow growth of new seedlings in the existing artificial planting technology of astragalus in high-cold areas, this invention provides a method for planting astragalus in high-cold areas, including using compound fertilizer as base fertilizer during deep soil plowing, manually digging astragalus seedlings when the night temperature of the surface in autumn is -4℃±1℃ for at least 3 consecutive days, placing astragalus seedlings in the ditch at a certain angle by using the method of abductor seedlings, and spraying humic acid foliar fertilizer when the seedling height is 10-12cm and during the rapid growth period after the astragalus returns to green. This scheme strictly controls the excavation, processing and transplanting time of astragalus seedlings, ensures the quality and nutritional content of astragalus seedlings, improves the greening rate of astragalus in the following spring, achieves a high greening rate of astragalus, ensures the survival rate and growth rate of seedlings, and improves the yield and quality of astragalus.
[0004] Therefore, the related technology can optimize the planting conditions of Astragalus and improve its yield and efficacy. However, the related technology does not take into account the different requirements of Astragalus for the planting environment under different growth conditions. That is, it is impossible to accurately control the planting environment parameters of Astragalus according to the real-time growth conditions of Astragalus.
[0005] The information disclosed in the background technology section of this application is only intended to deepen the understanding of the general background technology of this application, and should not be regarded as an admission or any form of suggestion that the information constitutes the prior art already known to those skilled in the art. Summary of the invention
[0006] The present invention provides an astragalus planting control method and system, which can solve the technical problem that the related technology cannot accurately control the astragalus planting environment parameters according to the real-time growth status of the astragalus.
[0007] According to a first aspect of the present invention, there is provided a method for controlling astragalus cultivation, comprising: in the ith control period, setting the planting environment of astragalus according to the planting environment parameters determined by the planting environment parameter model of the i-1th control period, i being a positive integer greater than or equal to 1, and in the case of i=1, the planting environment of astragalus is set according to the preset planting environment parameters; sampling all astragalus to obtain sample astragalus, and obtaining the astragalus seedling height, main root length and number of lateral roots of the sample astragalus at multiple moments in the ith control period; determining the astragalus growth status vector according to the astragalus seedling height, the main root length and the number of lateral roots; and determining the astragalus growth status vector according to the astragalus growth status vector of the sample astragalus in the ith control period and the sample astragalus in the i-1th control period. The astragalus growth status vector is used to determine whether the planting environment parameter model of the i-1th control period needs to be trained; if the planting environment parameter model of the i-1th control period needs to be trained, the loss function of the planting environment parameter model of the i-1th control period is determined according to the astragalus growth status vector of the sample astragalus of the i-1th control period and the astragalus growth status vector of the sample astragalus of the i-1th control period; according to the loss function of the planting environment parameter model of the i-1th control period, the planting environment parameter model of the i-1th control period is trained to obtain the planting environment parameter model of the i-1th control period; the planting environment parameters of the i+1th control period are determined through the planting environment parameter model of the i-1th control period.
[0008] According to the present invention, the growth status vector of Astragalus is determined according to the Astragalus seedling height, the main root length and the number of lateral roots, including: fitting the Astragalus seedling height and the moment in the ith control period to obtain the Astragalus seedling height function of the Astragalus seedling height in the ith control period; obtaining the Astragalus seedling height derivative function according to the Astragalus seedling height function; determining the seedling height change rate at multiple moments in the ith control period according to the Astragalus seedling height derivative function; determining the Astragalus seedling height status parameters according to the Astragalus seedling height and the seedling height change rate; determining the Astragalus root status parameters according to the main root length and the number of lateral roots; determining the Astragalus growth status vector according to the Astragalus seedling height status parameters and the Astragalus root status parameters.
[0009] According to the present invention, according to the astragalus seedling height and the seedling height change rate, determining the astragalus seedling height condition parameter includes: according to the formula , determine the astragalus seedling height condition parameter of the sth sample astragalus in the ith control period ,in, is the seedling height of the sth sample Astragalus at the end of the ith control period, To preset the height of Astragalus membranaceus seedlings, is the seedling height change rate of the s-th sample Astragalus membranaceus at the k-th moment in the ith control period, K is the number of moments in the control period, k≤K, and both k and K are positive integers.
[0010] According to the present invention, the root condition parameters of Astragalus membranaceus are determined according to the main root length and the number of lateral roots, including: determining the root vector of Astragalus membranaceus in the i-th control period according to the main root length and the number of lateral roots of the sample Astragalus membranaceus at the end of the i-th control period; determining the standard root vector of Astragalus membranaceus according to the preset main root length and the preset number of lateral roots; determining the standard root vector of Astragalus membranaceus according to the formula , determine the root condition parameters of the sth sample of Astragalus membranaceus in the ith control period ,in, is the taproot length of the sth sample Astragalus at the end of the ith control period, is the number of lateral roots of the sth sample Astragalus at the end of the ith control period, To preset the taproot length, To preset the number of lateral roots, is the root vector of the sth sample of Astragalus membranaceus in the ith control period, is the standard Astragalus root vector.
[0011] According to the present invention, according to the astragalus growth status vector of the astragalus sample of the i-th control period and the astragalus growth status vector of the astragalus sample of the i-1th control period, it is judged whether the planting environment parameter model of the i-1th control period needs to be trained, including: according to the formula , we obtained the astragalus seedling height condition A, astragalus root condition B and balanced growth condition C, among which, and is the preset weight, is the seedling height condition parameter of the sth sample of Astragalus membranaceus in the i-th control period, is the seedling height parameter of the sth sample of Astragalus membranaceus in the i-1th control period, is the root condition parameter of the sth sample of Astragalus membranaceus in the i-th control period, is the root condition parameter of the sth sample of Astragalus membranaceus in the i-1th control period, is the maximum value of the astragalus seedling height condition parameters of multiple samples of astragalus in the i-th control period, is the minimum value of the astragalus seedling height condition parameter of multiple samples of astragalus in the i-th control period, is the maximum value of the root condition parameters of multiple samples of Astragalus membranaceus in the i-th control period, The minimum value of the root condition parameters of multiple samples of Astragalus membranaceus in the i-th control period, To preset a balanced growth condition, S is the number of sample astragalus, s≤S, and s and S are both positive integers; when at least one of the astragalus seedling height condition A, the astragalus root condition B and the balanced growth condition C is met, it is determined that the planting environment parameter model at the i-1th moment needs to be trained.
[0012] According to the present invention, when the planting environment parameter model of the i-1th control period needs to be trained, the loss function of the planting environment parameter model of the i-1th control period is determined according to the astragalus growth status vector of the sample astragalus of the i-1th control period and the astragalus growth status vector of the sample astragalus of the i-1th control period, including: determining the conditions satisfied by the astragalus growth status vector of the sample astragalus of the i-1th control period and the astragalus growth status vector of the sample astragalus of the i-1th control period in the astragalus seedling height condition A, the astragalus root condition B and the balanced growth condition C; determining the loss function of the planting environment parameter model of the i-1th control period according to the conditions satisfied by the astragalus growth status vector of the sample astragalus of the i-1th control period and the astragalus growth status vector of the sample astragalus of the i-1th control period, as well as the astragalus growth status vector of the sample astragalus of the i-1th control period and the astragalus growth status vector of the sample astragalus of the i-1th control period.
[0013] According to the present invention, according to the conditions satisfied by the astragalus growth status vector of the sample astragalus in the i-th control period and the astragalus growth status vector of the sample astragalus in the i-1th control period, and the astragalus growth status vector of the sample astragalus in the i-th control period and the astragalus growth status vector of the sample astragalus in the i-1th control period, the loss function of the planting environment parameter model of the i-1th control period is determined, including: according to the formula , determine the loss function of the planting environment parameter model for the i-1th control period ,in, and is the preset parameter, and , is the maximum value of the astragalus seedling height condition parameter of the sample astragalus in the i-1th control period, is the minimum value of the astragalus seedling height condition parameter of the sample astragalus in the i-1th control period, is the maximum value of the root condition parameter of the Astragalus membranaceus sample in the i-1th control period, is the minimum value of the root condition parameter of the Astragalus membranaceus sample in the i-1th control period, is the threshold value of the Astragalus membranaceus seedling height condition parameter, is the threshold value of the root condition parameter of Astragalus membranaceus, , and is the preset weight, and if is the conditional function.
[0014] According to a second aspect of the present invention, there is provided an astragalus planting control system, comprising: a planting environment parameter module, for setting the planting environment of astragalus in the ith control period according to the planting environment parameters determined by the planting environment parameter model of the i-1th control period, i being a positive integer greater than or equal to 1, and when i=1, the planting environment of astragalus is set according to the preset planting environment parameters; a sample astragalus module, for sampling all astragalus to obtain sample astragalus, and obtaining the astragalus seedling height, main root length and number of lateral roots of the sample astragalus at multiple moments in the ith control period; an astragalus growth status vector module, for determining the astragalus growth status vector according to the astragalus seedling height, the main root length and the number of lateral roots; a judgment module, for determining the astragalus growth status vector according to the astragalus seedling height, the main root length and the number of lateral roots of the sample astragalus in the ith control period and the astragalus growth status vector of the sample astragalus in the i-1th control period. The astragalus growth status vector of the sample astragalus in the i-1 control period is used to judge whether the planting environment parameter model of the i-1 control period needs to be trained; the loss function module is used to determine the loss function of the planting environment parameter model of the i-1 control period according to the astragalus growth status vector of the sample astragalus in the i-1 control period and the astragalus growth status vector of the sample astragalus in the i-1 control period when the planting environment parameter model of the i-1 control period needs to be trained; the training module is used to train the planting environment parameter model of the i-1 control period according to the loss function of the planting environment parameter model of the i-1 control period to obtain the planting environment parameter model of the i-1 control period; the determination module is used to determine the planting environment parameters of the i+1 control period through the planting environment parameter model of the i-1 control period.
[0015] Technical effect: According to the present invention, by sampling and analyzing the growth data of sample astragalus, the growth status vector of astragalus can be obtained, which is helpful to evaluate the growth status of astragalus in each control period and make corresponding adjustments to the planting environment. By determining the loss function based on the growth status of astragalus, and training the planting environment parameter model based on the loss function, continuous improvement and optimization of the planting environment of astragalus can be achieved, thereby improving the yield and quality of astragalus. The planting environment of astragalus is accurately controlled by the planting environment parameter model to better promote the growth and development of astragalus. When determining the astragalus seedling height status parameters, the astragalus seedling height status parameters can be determined by the astragalus seedling height and the seedling height change rate. The astragalus seedling height growth status can be determined from two aspects: the astragalus seedling height and the change in the astragalus seedling height growth rate. The growth trend of astragalus at different growth stages can be monitored, which helps to promptly detect abnormal conditions in the growth of astragalus. When determining the root condition parameters of Astragalus, the root condition parameters of Astragalus can be determined by the cosine similarity between the root vector of Astragalus and the root vector of the standard Astragalus. The two parameters of main root length and lateral root number reflect the overall state of the root structure of Astragalus, and better evaluate the root condition of Astragalus. When judging whether the planting environment parameter model of the i-1 control period needs to be trained, the growth condition vector of Astragalus of the sample Astragalus of the i-1 control period and the growth condition vector of Astragalus of the sample Astragalus of the i-1 control period can be used to judge whether the planting environment parameter model of the i-1 control period needs to be trained. When judging whether the planting environment parameter model needs to be trained, the relationship between the growth condition vectors of Astragalus of the sample Astragalus of two adjacent control periods is referred to, so as to judge whether the planting environment parameter model has reached a stable state and whether the planting environment parameters of the planting environment parameter model are suitable for the growth of Astragalus of the i-1 control period. If the conditions for training are met, the planting environment parameter model needs to be continuously optimized to improve the accuracy of the planting environment parameter model. When determining the loss function of the planting environment parameter model for the i-1th control period, the loss function of the planting environment parameter model for the i-1th control period can be determined by the conditions satisfied by the astragalus growth status vector of each sample astragalus and the astragalus growth status vector of the adjacent control period, so that the planting environment parameter model can quickly reach a stable state, and the overall astragalus seedling height condition parameters and astragalus root condition parameters of the sample astragalus can be maximized, and the overall growth conditions of each sample astragalus can be homogenized, thereby improving training efficiency and improving the accuracy of the planting environment parameter model.
[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only and do not limit the present invention. Other features and aspects of the present invention will become more apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other embodiments can be obtained based on these drawings without creative work.
[0018] Figure 1 A schematic diagram of a process flow of an Astragalus membranaceus planting control method according to an embodiment of the present invention is exemplarily shown;
[0019] Figure 2 A block diagram of an astragalus planting control system according to an embodiment of the present invention is exemplarily shown. DETAILED DESCRIPTION
[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0021] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.
[0022] Figure 1A flow chart of an astragalus planting control method according to an embodiment of the present invention is exemplarily shown, the method comprising: step S101, in the i-th control cycle, setting the planting environment of astragalus according to the planting environment parameters determined by the planting environment parameter model of the i-1-th control cycle, i is a positive integer greater than or equal to 1, and when i=1, the planting environment of astragalus is set according to the preset planting environment parameters; step S102, sampling all astragalus to obtain sample astragalus, and obtaining the astragalus seedling height, main root length and lateral root number of the sample astragalus at multiple moments in the i-th control cycle; step S103, determining the astragalus growth status vector according to the astragalus seedling height, the main root length and the lateral root number; step S104, determining the astragalus growth status vector according to the astragalus growth status vector of the sample astragalus in the i-th control cycle and the i-1-th control cycle. The astragalus growth status vector of the sample astragalus of the i-1 control cycle is used to determine whether the planting environment parameter model of the i-1 control cycle needs to be trained; step S105, when the planting environment parameter model of the i-1 control cycle needs to be trained, the loss function of the planting environment parameter model of the i-1 control cycle is determined according to the astragalus growth status vector of the sample astragalus of the i-1 control cycle and the astragalus growth status vector of the sample astragalus of the i-1 control cycle; step S106, according to the loss function of the planting environment parameter model of the i-1 control cycle, the planting environment parameter model of the i-1 control cycle is trained to obtain the planting environment parameter model of the i-1 control cycle; step S107, the planting environment parameters of the i+1 control cycle are determined through the planting environment parameter model of the i-1 control cycle.
[0023] According to the astragalus planting control method of an embodiment of the present invention, by sampling and analyzing the growth data of the sample astragalus, an astragalus growth status vector can be obtained, which is helpful to evaluate the growth status of the astragalus in each control cycle and make corresponding adjustments to the planting environment. By determining the loss function based on the growth status of the astragalus and training the planting environment parameter model based on the loss function, continuous improvement and optimization of the astragalus planting environment can be achieved, thereby improving the yield and quality of the astragalus. The planting environment parameter model is used to accurately control the planting environment of the astragalus to better promote the growth and development of the astragalus.
[0024] According to one embodiment of the present invention, in step S101, each control cycle can be set to three days, one week, etc., and the present invention does not limit this. In the i-th control cycle, the planting environment of Astragalus is set according to the planting environment parameters determined by the planting environment parameter model of the i-1-th control cycle. For example, in the second control cycle, the planting environment of Astragalus is set according to the planting environment parameters determined by the planting environment parameter model of the first control cycle. In the first control cycle, the planting environment of Astragalus is set according to the preset planting environment parameters. The planting environment parameters include temperature parameters, humidity parameters, light parameters, and the content of various nutrients in the soil of the planting location in the Astragalus planting environment.
[0025] According to one embodiment of the present invention, in step S102, the interval between adjacent moments can be set to 12 hours, 24 hours, etc., and the present invention is not limited to this. The astragalus planting area is divided into several areas, and two astragalus plants are randomly selected in each area as sample astragalus plants, and the growth status of the overall astragalus plants is determined by observing the growth status of the sample astragalus plants. The selected sample astragalus plants are measured and observed at multiple moments in the i-th control cycle, including the astragalus seedling height, main root length and lateral root number of the sample astragalus plants, so as to fully understand the growth status of the astragalus plants in the control cycle.
[0026] According to an embodiment of the present invention, in step S103, based on the values of the height of the astragalus seedling, the length of the main root and the number of lateral roots, the growth status of the astragalus plant can be inferred, such as whether the growth is healthy, whether there are developmental problems or whether it is affected by the planting environment, etc. By comprehensively considering the values of these growth parameters, a comprehensive astragalus growth status vector can be formed, which reflects the overall growth status of the sample astragalus within a certain control period.
[0027] According to one embodiment of the present invention, step S103 includes: fitting the astragalus seedling height and the moment in the i-th control cycle to obtain the astragalus seedling height function of the astragalus seedling height in the i-th control cycle; obtaining the astragalus seedling height derivative function according to the astragalus seedling height function; determining the seedling height change rate at multiple moments in the i-th control cycle according to the astragalus seedling height derivative function; determining the astragalus seedling height status parameters according to the astragalus seedling height and the seedling height change rate; determining the astragalus root status parameters according to the main root length and the number of lateral roots; determining the astragalus growth status vector according to the astragalus seedling height status parameters and the astragalus root status parameters.
[0028] According to one embodiment of the present invention, the height of Astragalus membranaceus seedlings and the moments in the control cycle are fitted to obtain an Astragalus membranaceus seedling height function for describing the change of the height of Astragalus membranaceus seedlings over time in the control cycle. The Astragalus membranaceus seedling height function is differentiated to obtain an Astragalus membranaceus seedling height derivative function. Substituting multiple moments in the control cycle into the Astragalus membranaceus height derivative function to determine the change rate of the seedling height at multiple moments in the control cycle is helpful to evaluate the growth rate and trend of Astragalus membranaceus at different moments. The Astragalus membranaceus seedling height status parameter and the Astragalus membranaceus root status parameter are combined to obtain an Astragalus membranaceus growth status vector, that is, the Astragalus membranaceus seedling height status parameter and the Astragalus membranaceus root status parameter are both elements of the Astragalus membranaceus growth status vector, which integrates the seedling height and root status information of the sample Astragalus membranaceus in the i-th control cycle, and provides a comprehensive growth evaluation.
[0029] According to an embodiment of the present invention, determining the astragalus seedling height condition parameter according to the astragalus seedling height and the seedling height change rate includes: determining the astragalus seedling height condition parameter of the sth sample astragalus in the ith control period according to formula (1): , (1)
[0030] in, is the seedling height of the sth sample Astragalus at the end of the ith control period, To preset the height of Astragalus membranaceus seedlings, is the seedling height change rate of the s-th sample Astragalus membranaceus at the k-th moment in the ith control period, K is the number of moments in the control period, k≤K, and both k and K are positive integers.
[0031] According to one embodiment of the present invention, in formula (1), is the relative difference between the seedling height of the sth sample Astragalus at the end of the ith control period and the preset Astragalus seedling height. The larger the relative difference, the larger the difference between the actual measured Astragalus seedling height and the preset Astragalus seedling height, which indicates that the growth problem of Astragalus seedling height is more serious. The result of 1 minus the relative difference can be obtained. The closer the result is to 1, the closer the actual measured Astragalus seedling height is to the preset Astragalus seedling height, which indicates that the growth condition of Astragalus is better. It is the ratio of the difference between the maximum and minimum values of the seedling height change rate of the sth Astragalus plant in the ith control period to the average value of the seedling height change rate, that is, the relative difference between the maximum and minimum values of the seedling height change rate. The smaller the relative difference, the more uniform the change in the seedling height of the Astragalus. The result of 1 minus the ratio can be obtained. The closer the result is to 1, the more uniform the change in the seedling height of the Astragalus, indicating that the growth condition of the Astragalus is better. By multiplying the above two results, the Astragalus seedling height condition parameter of the sth sample Astragalus in the ith control period can be obtained.
[0032] In this way, the parameters of Astragalus seedling height status can be determined by the Astragalus seedling height and the seedling height change rate. The growth status of Astragalus seedling height can be determined from two aspects: the Astragalus seedling height and the change in the growth rate of Astragalus seedling height. The growth trend of Astragalus at different growth stages can be monitored, which is helpful to timely discover abnormal conditions in the growth of Astragalus.
[0033] According to one embodiment of the present invention, determining the root condition parameters of Astragalus membranaceus according to the main root length and the number of lateral roots includes: determining the root vector of Astragalus membranaceus in the i-th control period according to the main root length and the number of lateral roots of the sample Astragalus membranaceus at the end time of the i-th control period; determining the standard root vector of Astragalus membranaceus according to a preset main root length and a preset number of lateral roots; determining the root condition parameters of Astragalus membranaceus of the s-th sample Astragalus membranaceus in the i-th control period according to formula (2): , (2) Among them, is the taproot length of the sth sample Astragalus at the end of the ith control period, is the number of lateral roots of the sth sample Astragalus at the end of the ith control period, To preset the taproot length, To preset the number of lateral roots, is the root vector of the sth sample of Astragalus membranaceus in the ith control period, is the standard Astragalus root vector.
[0034] According to one embodiment of the present invention, the preset taproot length and the preset number of lateral roots of the Astragalus membranaceus are determined based on the date at which the i-th control cycle ends (for example, the date at which the i-th control cycle ends is the 10th day after the Astragalus membranaceus is planted) and the theoretical growth date of the Astragalus membranaceus, and the standard Astragalus membranaceus root vector is determined based on the preset taproot length and the preset number of lateral roots. In formula (2), It is the cosine similarity between the root vector of the s-th sample Astragalus in the ith control period and the root vector of the standard Astragalus. The closer the cosine similarity is to 1, the closer the root growth condition of the s-th sample Astragalus is to the root growth condition of the standard Astragalus. The root condition parameter of the s-th sample Astragalus in the ith control period is determined according to the cosine similarity. Since too fast or too slow growth rate of Astragalus will lead to the decline of quality and nutritional value of Astragalus, the closer the root condition parameter of Astragalus is to 1, the better the growth condition of Astragalus.
[0035] In this way, the root condition parameters of Astragalus membranaceus can be determined through the cosine similarity between the root vector of Astragalus membranaceus and the root vector of the standard Astragalus membranaceus. The two parameters of main root length and lateral root number reflect the overall state of the root structure of Astragalus membranaceus and better evaluate the root condition of Astragalus membranaceus.
[0036] According to one embodiment of the present invention, the astragalus seedling height condition parameters of the s-th sample astragalus in the i-th control period are combined with the astragalus root condition parameters of the s-th sample astragalus in the i-th control period to obtain the astragalus growth condition vector of the s-th sample astragalus in the i-th control period.
[0037] According to one embodiment of the present invention, in step S104, the astragalus growth status vector of the sample astragalus in the i-th control period is compared with the astragalus growth status vector of the sample astragalus in the i-1th control period, so as to understand the growth change of astragalus in two adjacent periods. According to the growth change of astragalus in two adjacent periods, it is determined whether the planting environment parameter model of the i-1th control period needs to be trained.
[0038] According to an embodiment of the present invention, judging whether the planting environment parameter model of the i-1th control period needs to be trained is performed based on the astragalus growth status vector of the sample astragalus of the i-th control period and the astragalus growth status vector of the sample astragalus of the i-1th control period, including: obtaining the astragalus seedling height condition A, the astragalus root condition B and the balanced growth condition C according to formula (3); (3) Among them, and is the preset weight, is the seedling height condition parameter of the sth sample of Astragalus membranaceus in the i-th control period, is the seedling height parameter of the sth sample of Astragalus membranaceus in the i-1th control period, is the root condition parameter of the sth sample of Astragalus membranaceus in the i-th control period, is the root condition parameter of the sth sample of Astragalus membranaceus in the i-1th control period, is the maximum value of the astragalus seedling height condition parameters of multiple samples of astragalus in the i-th control period, is the minimum value of the astragalus seedling height condition parameter of multiple samples of astragalus in the i-th control period, is the maximum value of the root condition parameters of multiple samples of Astragalus membranaceus in the i-th control period, The minimum value of the root condition parameters of multiple samples of Astragalus membranaceus in the i-th control period, To preset a balanced growth condition, S is the number of sample astragalus, s≤S, and s and S are both positive integers; when at least one of the astragalus seedling height condition A, the astragalus root condition B and the balanced growth condition C is met, it is determined that the planting environment parameter model at the i-1th moment needs to be trained.
[0039] According to one embodiment of the present invention, in the Astragalus membranaceus seedling height condition A, The average value of the difference between the astragalus seedling height condition parameters of the s-th sample astragalus in the i-th control period and the astragalus seedling height condition parameters in the i-1-th control period is less than or equal to 0, indicating that the astragalus seedling height condition parameters in the i-th control period are less than or equal to the astragalus seedling height condition parameters in the i-1-th control period, and the growth rate of the astragalus seedling height slows down, indicating that the planting environment parameters in the i-1-th control period are set incorrectly, or the planting environment parameters in the i-1-th control period are not suitable for the growth of astragalus in the i-th control period, and the planting environment parameter model needs to be trained.
[0040] According to one embodiment of the present invention, in the Astragalus root condition B, The average value of the difference between the root condition parameters of the s-th sample Astragalus in the i-th control period and the root condition parameters of the Astragalus in the i-1th control period is less than or equal to 0, indicating that the root condition parameters of the Astragalus in the i-th control period are less than or equal to the root condition parameters of the Astragalus in the i-1th control period, and the growth rate of the Astragalus root slows down, indicating that the planting environment parameters in the i-1th control period are set incorrectly, or the planting environment parameters in the i-1th control period are not suitable for the growth of Astragalus in the i-th control period, and the planting environment parameter model needs to be trained.
[0041] According to one embodiment of the present invention, in the balanced growth condition C, It is the difference between the maximum and minimum values of the astragalus seedling height condition parameters of multiple sample astragalus plants in the i-th control period, and the ratio of the maximum value of the astragalus seedling height condition parameters of multiple sample astragalus plants, indicating the degree of uneven growth of the astragalus seedling height of multiple sample astragalus plants. It is the ratio of the difference between the maximum and minimum values of the root condition parameters of multiple Astragalus samples in the i-th control period to the maximum value of the root condition parameters of multiple Astragalus samples, indicating the degree of uneven growth of the roots of multiple Astragalus samples. It is a weighted summation based on the differences in seedling height growth and root growth of multiple Astragalus samples, indicating the degree of overall growth unevenness of multiple Astragalus samples. The degree of overall growth unevenness of multiple sample Astragalus is greater than or equal to the preset balanced growth condition, indicating that the degree of growth unevenness of multiple sample Astragalus is too large and not even, and some sample Astragalus are poorly grown, and the planting environment parameter model needs to be trained.
[0042] According to one embodiment of the present invention, when at least one of the above Astragalus seedling height condition A, Astragalus root condition B and balanced growth condition C is met, it is determined that the planting environment parameter model at the i-1th moment needs to be trained.
[0043] In this way, the astragalus growth status vector of the astragalus sample in the i-th control period and the astragalus growth status vector of the astragalus sample in the i-1-th control period can be used to determine whether the planting environment parameter model needs to be trained. When determining whether the planting environment parameter model needs to be trained, the relationship between the astragalus growth status vectors of the astragalus samples in two adjacent control periods is referenced to determine whether the planting environment parameter model has reached a stable state and whether the planting environment parameters of the planting environment parameter model are suitable for the growth of astragalus in the i-th control period. If the conditions for training are met, the planting environment parameter model needs to be continuously optimized to improve the accuracy of the planting environment parameter model.
[0044] According to one embodiment of the present invention, in step S105, the planting environment parameter model of the i-1th control period needs to be retrained to adapt to the change in the growth condition of Astragalus. The loss function is used to measure the difference between the model prediction value and the actual observation value. By minimizing the loss function, according to the Astragalus growth condition vector of the sample Astragalus in the i-th control period and the i-1th control period, the parameters of the planting environment parameter model are adjusted to better fit the growth condition of Astragalus.
[0045] According to one embodiment of the present invention, step S105 includes: determining the conditions satisfied by the astragalus growth status vector of the sample astragalus in the i-th control period and the astragalus growth status vector of the sample astragalus in the i-1th control period in the astragalus seedling height condition A, the astragalus root condition B and the balanced growth condition C; determining the loss function of the planting environment parameter model of the i-1th control period according to the conditions satisfied by the astragalus growth status vector of the sample astragalus in the i-th control period and the astragalus growth status vector of the sample astragalus in the i-1th control period, as well as the astragalus growth status vector of the sample astragalus in the i-th control period and the astragalus growth status vector of the sample astragalus in the i-1th control period.
[0046] According to an embodiment of the present invention, the planting environment parameter model of the i-1th control period is trained and optimized through the loss function. By minimizing the loss function, the model parameters are adjusted so that it can better meet the needs of Astragalus under different growth conditions.
[0047] According to one embodiment of the present invention, based on the conditions satisfied by the astragalus growth status vector of the sample astragalus in the i-th control period and the astragalus growth status vector of the sample astragalus in the i-1th control period, as well as the astragalus growth status vector of the sample astragalus in the i-th control period and the astragalus growth status vector of the sample astragalus in the i-1th control period, the loss function of the planting environment parameter model of the i-1th control period is determined, including: determining the loss function of the planting environment parameter model of the i-1th control period according to formula (4): , (4) Among them, and is the preset parameter, and , is the maximum value of the astragalus seedling height condition parameter of the sample astragalus in the i-1th control period, is the minimum value of the astragalus seedling height condition parameter of the sample astragalus in the i-1th control period, is the maximum value of the root condition parameter of the Astragalus membranaceus sample in the i-1th control period, is the minimum value of the root condition parameter of the Astragalus membranaceus sample in the i-1th control period, is the threshold value of the Astragalus membranaceus seedling height condition parameter, is the threshold value of the root condition parameter of Astragalus membranaceus, , and is the preset weight, and if is the conditional function.
[0048] According to one embodiment of the present invention, in formula (4), The following two situations are expressed in the form of conditional functions. When the astragalus seedling height condition A is met, the value of the above conditional function is That is, the difference between the threshold of the astragalus seedling height condition parameter and the average value of the difference between the astragalus seedling height condition parameters of two adjacent control periods is used as the loss function. In the process of training the planting environment parameter model, the difference is narrowed, so that the average value of the difference between the astragalus seedling height condition parameters of two adjacent control periods is close to the threshold of the astragalus seedling height condition parameter, the astragalus seedling height growth standard is met, and the planting environment parameter model reaches a stable state as soon as possible. When the astragalus seedling height condition A is not met, the value of the above condition function is 0, and the parameters related to the astragalus seedling height in the loss function do not need to be adjusted.
[0049] According to one embodiment of the present invention, in formula (4), The following two situations are expressed in the form of conditional functions. When the root condition B of Astragalus is met, the value of the above conditional function is That is, the difference between the threshold value of the astragalus root condition parameter and the average value of the difference between the astragalus root condition parameters of two adjacent control periods is used as the loss function. In the process of training the planting environment parameter model, the difference is narrowed, so that the average value of the difference between the astragalus root condition parameters of two adjacent control periods is close to the threshold value of the astragalus root condition parameter, the astragalus root growth standard is met, and the planting environment parameter model reaches a stable state as soon as possible. When the astragalus root condition B is not met, the value of the above condition function is 0, and the parameters related to the astragalus root in the loss function do not need to be adjusted.
[0050] According to one embodiment of the present invention, in formula (4), Indicates the following two situations in the form of conditional functions: When , the degree of overall uneven growth of each sample of Astragalus in the i-th control period is less than the degree of overall uneven growth of each sample of Astragalus in the i-1th control period, the value of the conditional function is That is, the average value of the difference between the astragalus seedling height condition parameters, astragalus root condition parameters of each sample astragalus and the average astragalus seedling height condition parameters, and the average astragalus root condition parameters is weighted summed. During the training process, the average value of the difference is reduced, thereby improving the averageness of the overall growth condition of astragalus. When the degree of overall uneven growth of each sample of Astragalus membranaceus in the i-th control period is greater than or equal to the degree of overall uneven growth of each sample of Astragalus membranaceus in the i-1th control period, it means that the planting environment parameter set by the planting environment parameter model is wrong, and the conditional function value at this time is ,because Therefore, when errors occur in the planting environment parameters set by the planting environment parameter model, the training intensity of the planting environment parameter model can be increased through larger preset parameters to quickly correct the setting errors of the planting environment parameter model.
[0051] According to an embodiment of the present invention, in formula (4), the above three items can be weighted summed to obtain the loss function of the planting environment parameter model of the i-1th control cycle.
[0052] In this way, the loss function of the planting environment parameter model of the i-1th control period can be determined by the conditions satisfied by the astragalus growth status vector of each sample astragalus and the astragalus growth status vector of the adjacent control period, so that the planting environment parameter model can quickly reach a stable state, and the overall astragalus seedling height condition parameters and astragalus root condition parameters of the sample astragalus can be maximized, and the overall growth condition of each sample astragalus can be homogenized, thereby improving the training efficiency and the accuracy of the planting environment parameter model.
[0053] According to an embodiment of the present invention, in step S106, after training, the planting environment parameter model of the i-1th control period has been optimized and adjusted. The planting environment parameter model will be used to set the planting environment parameters of the i-th control period, thereby optimizing the planting environment of Astragalus.
[0054] According to one embodiment of the present invention, in step S107, the planting environment parameters of the i+1th control cycle are determined based on the planting environment parameter model of the i-th control cycle, so that the astragalus seedling height condition parameters and the astragalus root condition parameters of the astragalus are maximized and the growth conditions of each astragalus are uniform.
[0055] According to the Astragalus planting control method of an embodiment of the present invention, by sampling and analyzing the growth data of the sample Astragalus, the Astragalus growth status vector can be obtained, which is helpful to evaluate the growth status of Astragalus in each control cycle and make corresponding adjustments to the planting environment. The loss function is determined by the growth status of Astragalus, and the planting environment parameter model is trained according to the loss function, so as to achieve continuous improvement and optimization of the Astragalus planting environment, thereby improving the yield and quality of Astragalus. The planting environment of Astragalus is accurately controlled by the planting environment parameter model to better promote the growth and development of Astragalus. When determining the Astragalus seedling height status parameters, the Astragalus seedling height status parameters can be determined by the Astragalus seedling height and the seedling height change rate. The Astragalus seedling height growth status is determined from two aspects: the Astragalus seedling height and the Astragalus seedling height growth rate change. The growth trend of Astragalus at different growth stages can be monitored, which helps to timely discover abnormal conditions in the growth of Astragalus. When determining the root condition parameters of Astragalus, the root condition parameters of Astragalus can be determined by the cosine similarity between the root vector of Astragalus and the root vector of the standard Astragalus. The two parameters of main root length and lateral root number reflect the overall state of the root structure of Astragalus, and better evaluate the root condition of Astragalus. When judging whether the planting environment parameter model of the i-1 control period needs to be trained, the growth condition vector of Astragalus of the sample Astragalus of the i-1 control period and the growth condition vector of Astragalus of the sample Astragalus of the i-1 control period can be used to judge whether the planting environment parameter model of the i-1 control period needs to be trained. When judging whether the planting environment parameter model needs to be trained, the relationship between the growth condition vectors of Astragalus of the sample Astragalus of two adjacent control periods is referred to, so as to judge whether the planting environment parameter model has reached a stable state and whether the planting environment parameters of the planting environment parameter model are suitable for the growth of Astragalus of the i-1 control period. If the conditions for training are met, the planting environment parameter model needs to be continuously optimized to improve the accuracy of the planting environment parameter model. When determining the loss function of the planting environment parameter model for the i-1th control period, the loss function of the planting environment parameter model for the i-1th control period can be determined by the conditions satisfied by the astragalus growth status vector of each sample astragalus and the astragalus growth status vector of the adjacent control period, so that the planting environment parameter model can quickly reach a stable state, and the overall astragalus seedling height condition parameters and astragalus root condition parameters of the sample astragalus can be maximized, and the overall growth conditions of each sample astragalus can be homogenized, thereby improving training efficiency and improving the accuracy of the planting environment parameter model.
[0056] Figure 2A block diagram of an astragalus planting control system according to an embodiment of the present invention is exemplarily shown, the system comprising: a planting environment parameter module, for setting the planting environment of astragalus in the ith control cycle according to the planting environment parameters determined by the planting environment parameter model of the i-1th control cycle, i being a positive integer greater than or equal to 1, and when i=1, the planting environment of astragalus is set according to the preset planting environment parameters; a sample astragalus module, for sampling all astragalus to obtain sample astragalus, and obtaining the astragalus seedling height, main root length and number of lateral roots of the sample astragalus at multiple moments in the ith control cycle; an astragalus growth status vector module, for determining the astragalus growth status vector based on the astragalus seedling height, the main root length and the number of lateral roots; a judgment module, for determining the astragalus growth status vector based on the astragalus growth status vector of the sample astragalus in the ith control cycle and the ith control cycle. The astragalus growth status vector of the astragalus sample in the i-1 control period is used to determine whether the planting environment parameter model of the i-1 control period needs to be trained; a loss function module is used to determine the loss function of the planting environment parameter model of the i-1 control period according to the astragalus growth status vector of the astragalus sample in the i-1 control period and the astragalus growth status vector of the astragalus sample in the i-1 control period when the planting environment parameter model of the i-1 control period needs to be trained; a training module is used to train the planting environment parameter model of the i-1 control period according to the loss function of the planting environment parameter model of the i-1 control period to obtain the planting environment parameter model of the i-1 control period; a determination module is used to determine the planting environment parameters of the i+1 control period through the planting environment parameter model of the i-th control period.
[0057] It should be understood by those skilled in the art that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.
Claims
1. A method for controlling astragalus cultivation, characterized in that: include: In the ith control cycle, the planting environment of Astragalus is set according to the planting environment parameters determined by the planting environment parameter model of the i-1th control cycle, where i is a positive integer greater than or equal to 1. When i=1, the planting environment of Astragalus is set according to the preset planting environment parameters; sampling is performed among all Astragalus to obtain sample Astragalus, and the Astragalus seedling height, main root length and number of lateral roots of the sample Astragalus at multiple moments in the ith control cycle are obtained; the Astragalus growth status vector is determined according to the Astragalus seedling height, the main root length and the number of lateral roots; the Astragalus growth status vector of the sample Astragalus in the ith control cycle and the Astragalus growth status vector of the sample Astragalus in the i-1th control cycle are judged. Whether the planting environment parameter model of the i-1 control cycle needs to be trained; if the planting environment parameter model of the i-1 control cycle needs to be trained, determine the loss function of the planting environment parameter model of the i-1 control cycle according to the astragalus growth status vector of the sample astragalus of the i-1 control cycle and the astragalus growth status vector of the sample astragalus of the i-1 control cycle; train the planting environment parameter model of the i-1 control cycle according to the loss function of the planting environment parameter model of the i-1 control cycle to obtain the planting environment parameter model of the i control cycle; determine the planting environment parameters of the i+1 control cycle through the planting environment parameter model of the i-1 control cycle; Determining the growth status vector of Astragalus membranaceus according to the Astragalus membranaceus seedling height, the main root length and the number of lateral roots, including: fitting the Astragalus membranaceus seedling height and the moment in the ith control period to obtain the Astragalus membranaceus seedling height function of the Astragalus membranaceus seedling height in the ith control period; obtaining the Astragalus membranaceus seedling height derivative function according to the Astragalus membranaceus seedling height function; determining the seedling height change rate at multiple moments in the ith control period according to the Astragalus membranaceus seedling height derivative function; determining the Astragalus membranaceus seedling height status parameter according to the Astragalus membranaceus seedling height and the seedling height change rate; determining the Astragalus membranaceus root status parameter according to the main root length and the number of lateral roots; determining the Astragalus membranaceus growth status vector according to the Astragalus membranaceus seedling height status parameter and the Astragalus membranaceus root status parameter; According to the astragalus seedling height and the seedling height change rate, determining the astragalus seedling height condition parameter includes: according to the formula , determine the astragalus seedling height condition parameter of the sth sample astragalus in the ith control period ,in, is the seedling height of the sth sample Astragalus at the end of the ith control period, To preset the height of Astragalus membranaceus seedlings, is the seedling height change rate of the s-th sample Astragalus membranaceus at the k-th moment in the ith control period, K is the number of moments in the control period, k≤K, and both k and K are positive integers; According to the main root length and the number of lateral roots, the root condition parameters of Astragalus membranaceus are determined, including: according to the main root length and the number of lateral roots of the sample Astragalus membranaceus at the end of the i-th control period, the root vector of Astragalus membranaceus is determined; according to the preset main root length and the preset number of lateral roots, the standard root vector of Astragalus membranaceus is determined; according to the formula , determine the root condition parameters of the sth sample of Astragalus membranaceus in the ith control period ,in, is the taproot length of the sth sample Astragalus at the end of the ith control period, is the number of lateral roots of the sth sample Astragalus at the end of the ith control period, To preset the taproot length, To preset the number of lateral roots, is the root vector of the sth sample of Astragalus membranaceus in the ith control period, is the standard Astragalus root vector.
2. The Astragalus membranaceus planting control method according to claim 1, characterized in that: According to the astragalus growth status vector of the astragalus sample in the i-th control period and the astragalus growth status vector of the astragalus sample in the i-1th control period, it is judged whether the planting environment parameter model of the i-1th control period needs to be trained, including: according to the formula , we obtained the astragalus seedling height condition A, astragalus root condition B and balanced growth condition C, among which, and is the preset weight, is the seedling height condition parameter of the sth sample of Astragalus membranaceus in the i-th control period, is the seedling height parameter of the sth sample of Astragalus membranaceus in the i-1th control period, is the root condition parameter of the sth sample of Astragalus membranaceus in the i-th control period, is the root condition parameter of the sth sample of Astragalus membranaceus in the i-1th control period, is the maximum value of the astragalus seedling height condition parameters of multiple samples of astragalus in the i-th control period, is the minimum value of the astragalus seedling height condition parameter of multiple samples of astragalus in the i-th control period, is the maximum value of the root condition parameters of multiple samples of Astragalus membranaceus in the i-th control period, The minimum value of the root condition parameters of multiple samples of Astragalus membranaceus in the i-th control period, To preset a balanced growth condition, S is the number of sample astragalus, s≤S, and s and S are both positive integers; when at least one of the astragalus seedling height condition A, the astragalus root condition B and the balanced growth condition C is met, it is determined that the planting environment parameter model at the i-1th moment needs to be trained.
3. The Astragalus membranaceus planting control method according to claim 2, characterized in that: In the case where the planting environment parameter model of the i-1th control period needs to be trained, the loss function of the planting environment parameter model of the i-1th control period is determined according to the astragalus growth status vector of the sample astragalus of the i-1th control period and the astragalus growth status vector of the sample astragalus of the i-1th control period, including: determining the conditions satisfied by the astragalus growth status vector of the sample astragalus of the i-1th control period and the astragalus growth status vector of the sample astragalus of the i-1th control period under the astragalus seedling height condition A, astragalus root condition B and balanced growth condition C; determining the loss function of the planting environment parameter model of the i-1th control period according to the conditions satisfied by the astragalus growth status vector of the sample astragalus of the i-1th control period and the astragalus growth status vector of the sample astragalus of the i-1th control period, as well as the astragalus growth status vector of the sample astragalus of the i-1th control period and the astragalus growth status vector of the sample astragalus of the i-1th control period.
4. The Astragalus membranaceus planting control method according to claim 3, characterized in that: According to the conditions satisfied by the astragalus growth status vector of the sample astragalus in the i-th control period and the astragalus growth status vector of the sample astragalus in the i-1th control period, and the astragalus growth status vector of the sample astragalus in the i-th control period and the astragalus growth status vector of the sample astragalus in the i-1th control period, determine the loss function of the planting environment parameter model of the i-1th control period, including: according to the formula , determine the loss function of the planting environment parameter model for the i-1th control period ,in, and is the preset parameter, and , is the maximum value of the astragalus seedling height condition parameter of the sample astragalus in the i-1th control period, is the minimum value of the astragalus seedling height condition parameter of the sample astragalus in the i-1th control period, is the maximum value of the root condition parameter of the Astragalus membranaceus sample in the i-1th control period, is the minimum value of the root condition parameter of the Astragalus membranaceus sample in the i-1th control period, is the threshold value of the Astragalus membranaceus seedling height condition parameter, is the threshold value of the root condition parameter of Astragalus membranaceus, , and is the preset weight, and if is the conditional function.
5. An astragalus planting control system for implementing the astragalus planting control method according to claim 1, characterized in that: include: The planting environment parameter module is used to set the planting environment of Astragalus membranaceus in the i-th control cycle according to the planting environment parameters determined by the planting environment parameter model of the i-1-th control cycle, where i is a positive integer greater than or equal to 1. When i=1, the planting environment of Astragalus membranaceus is set according to the preset planting environment parameters; The sample astragalus module is used to sample all the astragalus to obtain the sample astragalus, and obtain the astragalus seedling height, main root length and lateral root number of the sample astragalus at multiple moments in the i-th control period; the astragalus growth status vector module is used to determine the astragalus growth status vector according to the astragalus seedling height, the main root length and the lateral root number; the judgment module is used to judge whether the planting environment parameter model of the i-1th control period needs to be trained according to the astragalus growth status vector of the sample astragalus in the i-th control period and the astragalus growth status vector of the sample astragalus in the i-1th control period; the loss function module is used to determine the planting environment parameter model of the i-1th control period according to the astragalus growth status vector of the sample astragalus in the i-1th control period. When the planting environment parameter model needs to be trained, the loss function of the planting environment parameter model of the i-1th control period is determined according to the astragalus growth status vector of the sample astragalus of the i-1th control period and the astragalus growth status vector of the sample astragalus of the i-1th control period; a training module is used to train the planting environment parameter model of the i-1th control period according to the loss function of the planting environment parameter model of the i-1th control period to obtain the planting environment parameter model of the i-th control period; a determination module is used to determine the planting environment parameters of the i+1th control period through the planting environment parameter model of the i-th control period.
