AIOT-based intelligent garden virtual-real collaborative teaching practical training method and system

Abstract

The invention discloses an AIOT-based smart garden virtual-real collaborative teaching training method and system, and belongs to the technical field of AIOT. The method comprises the following steps: performing grid division on a target smart garden site area, constructing teaching practical training sub-areas, deploying multi-source environment sensing equipment in each sub-area, collecting soil humidity, environment temperature, air humidity and illumination intensity, synchronously mapping the soil humidity, the environment temperature, the air humidity and the illumination intensity to a virtual teaching environment, and constructing a virtual practical training unit; comprehensive evaluation and fluctuation factor calculation are carried out on the multi-dimensional environment data in the virtual practical training unit, a dynamic allowable fluctuation interval is constructed, comparison analysis is carried out on the environment data of adjacent sampling time points, and when abnormal fluctuation is detected, early warning is triggered and dynamic teaching practical training is carried out in a linkage mode. The perceptible, evaluable and respondable garden teaching environment state is realized, and the authenticity, interactivity and practical teaching effect of teaching practical training are improved.

Description

Technical Field

[0001] This invention relates to the field of AIoT technology, specifically to a smart garden virtual-real collaborative teaching and training method and system based on AIoT. Background Technology

[0002] With the rapid development of new-generation information technology, the integration of artificial intelligence, the Internet of Things, cloud computing, big data, and other technologies in education and industry applications is deepening, gradually forming a smart application system centered on AIoT. In the field of teaching related to landscape architecture and landscape engineering, traditional practical teaching is evolving from a model primarily based on experience transmission to a teaching form characterized by data-driven approaches, intelligent sensing, and virtual-real integration. On the one hand, smart landscape technology, by deploying various types of environmental sensing devices in landscape scenes, continuously collects key ecological parameters such as soil, water, climate, and light, providing objective data support for landscape maintenance, ecological regulation, and landscape management. On the other hand, the introduction of virtual simulation teaching, digital twins, and immersive teaching environments allows complex landscape ecological processes and environmental change patterns to be presented in a visualized and interactive way, effectively compensating for the limitations imposed by site, season, and safety conditions in traditional practical training. In recent years, some research and applications have begun to attempt to map real landscape environment data onto virtual teaching platforms to achieve linkage between "real scenes and virtual environments," thereby improving the authenticity and repeatability of teaching. However, overall, it is still in a development stage mainly based on static display or one-way data retrieval.

[0003] However, existing technologies related to smart garden teaching and virtual training still have significant shortcomings. First, most teaching systems focus on the collection and display of environmental data, lacking a systematic and comprehensive evaluation mechanism for multi-dimensional environmental data. This makes it difficult to reflect the overall changes in the garden's ecological state over time, resulting in fragmented teaching analysis results. Second, existing virtual simulation teaching platforms typically use preset scenarios or idealized parameters for demonstrations, lacking a high-frequency, dynamic data synchronization mechanism with the real garden environment. This makes it difficult for virtual training content to reflect environmental fluctuations and anomalies in the actual site in a timely manner, limiting the authenticity and relevance of teaching. Furthermore, in existing smart teaching applications, environmental anomaly identification largely relies on fixed thresholds or human experience judgment, failing to dynamically adjust based on different teaching areas, teaching types, and multi-parameter coupling relationships. This easily leads to misjudgments or delayed warnings, making it difficult to support high-level practical training aimed at process cognition and scenario response. Existing technologies lack regionalized modeling and virtual-real collaborative control mechanisms for teaching scenarios, making it difficult to meet the needs of refined teaching and practical training evaluation. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for collaborative teaching and training in smart gardens based on AIoT, so as to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] A method for collaborative teaching and training in smart gardens based on AIoT, comprising the following steps: Step S1: Divide the target smart garden site into several teaching and training sub-regions, and construct a set of teaching and training sub-regions; construct a sub-region status dataset of the teaching and training sub-regions at the data sampling time point; Step S2: Synchronously transmit the sub-region status dataset to the virtual teaching environment through the data interaction module in the teaching platform, and construct virtual training units corresponding to the teaching and training sub-regions; Step S3: Based on the virtual training units, obtain the environmental data of the corresponding teaching and training sub-regions at the data sampling time point, and calculate the comprehensive evaluation value and data fluctuation factor; Step S4: Based on the data fluctuation factor, calculate the allowable data fluctuation range of the virtual training units at the data sampling time point; obtain the environmental data of the virtual training units at the next data sampling time point, and compare it with the allowable data fluctuation range. If there is abnormal environmental fluctuation, issue an early warning to relevant personnel and conduct dynamic teaching and training.

[0007] As a preferred embodiment of the AIOT-based smart garden virtual-real collaborative teaching and training method of this invention, the target smart garden site area for teaching and training within the current teaching cycle is obtained, and a grid-based partitioning technique is used to divide the target smart garden site area into several teaching and training sub-regions, wherein one teaching and training sub-region corresponds to one teaching type within one teaching cycle; a set of teaching and training sub-regions is constructed. ,in, Let i represent the i-th teaching and training sub-region, and I represent the total number of teaching and training sub-regions.

[0008] Sensing data collection points are deployed within the teaching and training sub-area. These points include soil moisture sensors, ambient temperature sensors, air humidity sensors, and light intensity sensors. Based on these sensing data collection points, data about the teaching and training sub-area is obtained. Soil moisture data, ambient temperature data, air humidity data, and light intensity data.

[0009] As a preferred embodiment of the AIOT-based smart garden virtual-real collaborative teaching and training method described in this invention, a data sampling time period is constructed, denoted as... ,in, Let A represent the a-th data sampling time point, and A represent the total number of data sampling time points; the data sampling time points are respectively... Teaching and training sub-regions collected below The soil moisture data, ambient temperature data, air humidity data, and light intensity data are denoted as... , , and ;

[0010] Construct teaching and training sub-regions At the data sampling time point The sub-region state dataset is denoted as follows: .

[0011] As a preferred embodiment of the AIOT-based smart garden virtual-real collaborative teaching and training method described in this invention, the teaching and training sub-regions are... At the data sampling time point Sub-region state dataset The data is synchronously transmitted to the virtual teaching environment through the data interaction module in the teaching platform, and integrated with the teaching and training sub-areas. Corresponding virtual training unit The details are as follows:

[0012] Teaching and training sub-regions based on transmission to the virtual teaching environment At the data sampling time point Sub-region state dataset In the virtual teaching environment, the environmental parameters of the virtual training unit are initialized, including soil moisture data, ambient temperature data, air humidity data, and light intensity data.

[0013] Construct teaching and training sub-regions With virtual training units A bidirectional mapping table between them is denoted as .

[0014] As a preferred embodiment of the AIOT-based smart garden virtual-real collaborative teaching and training method described in this invention, based on virtual training units... From the bidirectional mapping table Obtain the corresponding teaching and training sub-region from the middle At the data sampling time point The soil moisture data, ambient temperature data, air humidity data, and light intensity data are recorded separately as follows: , , and And calculate the virtual training unit At the data sampling time point The comprehensive evaluation value is calculated using the following formula:

[0015] ;

[0016] in, Represents virtual training unit At the data sampling time point The overall evaluation value is as follows: This represents the weighting coefficient of the preset soil moisture data. This represents the weighting coefficient of the preset ambient temperature data. This represents the weighting coefficient of the preset air humidity data. This represents the weighting coefficient of the preset light intensity data. , , and This represents the preset ideal values ​​for soil moisture, ambient temperature, air humidity, and light intensity. , , and This indicates the maximum physiological fluctuation range of the preset soil moisture, ambient temperature, air humidity, and light intensity;

[0017] Based on virtual training units At the data sampling time point The comprehensive evaluation value is used to calculate the fluctuation factor of soil moisture data, ambient temperature data, air humidity data, and light intensity data. The calculation formula is as follows:

[0018] ;

[0019] in, , , and These represent the fluctuation factors for soil moisture data, ambient temperature data, air humidity data, and light intensity data, respectively.

[0020] As a preferred embodiment of the AIOT-based smart garden virtual-real collaborative teaching and training method described in this invention, based on soil moisture data fluctuation factors... Ambient temperature data fluctuation factor Air humidity data fluctuation factor and light intensity data fluctuation factor Computational Virtual Training Unit At the data sampling time point The permissible fluctuation ranges for soil moisture data, ambient temperature data, air humidity data, and light intensity data are as follows:

[0021] ;

[0022] in, , , and These represent the soil moisture data, ambient temperature data, air humidity data, and light intensity data at the data sampling time points, respectively. The allowable fluctuation range below and These represent the preset upper and lower physical boundaries of soil moisture data, respectively. and These represent the upper and lower physical boundaries of the preset ambient temperature data, respectively. and These represent the upper and lower bounds of the preset air humidity data, respectively. and These represent the upper and lower bounds of the preset light intensity data (preset based on the effective range of the sensor collecting the corresponding environmental parameters, and in conjunction with the physiological safety range of the teaching and training subjects).

[0023] Obtaining virtual training units At the data sampling time point The data includes soil moisture, ambient temperature, air humidity, and light intensity, and is compared with those of the virtual training unit. At the data sampling time point The allowable fluctuation ranges of soil moisture data, ambient temperature data, air humidity data, and light intensity data are compared. If any data does not fall within the allowable fluctuation range, the data sampling time point is determined. If there are abnormal fluctuations in the environment, an early warning will be issued to relevant personnel and dynamic teaching and training will be conducted.

[0024] This AIOT-based smart garden virtual-real collaborative teaching and training system includes: a set construction module, a data transmission module, an evaluation value calculation and factor calculation module, and a fluctuation range calculation and anomaly judgment module.

[0025] The set construction module: divides the target smart garden site area into several teaching and training sub-areas, constructs a set of teaching and training sub-areas; and constructs a sub-area status dataset of the teaching and training sub-areas at the data sampling time point.

[0026] The data transmission module synchronously transmits the sub-region status dataset to the virtual teaching environment through the data interaction module in the teaching platform, and constructs virtual training units corresponding to the teaching and training sub-regions.

[0027] The evaluation value calculation and factor calculation module: Based on the virtual training unit, it obtains the environmental data of the corresponding teaching and training sub-region at the data sampling time point, and calculates the comprehensive evaluation value and data fluctuation factor;

[0028] The fluctuation range calculation and anomaly judgment module calculates the allowable fluctuation range of the virtual training unit at the data sampling time point based on the data fluctuation factor; obtains the environmental data of the virtual training unit at the next data sampling time point and compares it with the allowable fluctuation range; if there is an abnormal environmental fluctuation, it issues an early warning to relevant personnel and conducts dynamic teaching and training.

[0029] Furthermore, the data transmission module includes a data transmission unit;

[0030] The data transmission unit synchronously transmits the sub-region state dataset of the teaching and training sub-region at the data sampling time point to the virtual teaching environment through the data interaction module in the teaching platform, and constructs a virtual training unit corresponding to the teaching and training sub-region. Specifically, based on the sub-region state dataset of the teaching and training sub-region at the data sampling time point transmitted to the virtual teaching environment, the environmental parameters of the virtual training unit are initialized in the virtual teaching environment, including soil moisture data, ambient temperature data, air humidity data, and light intensity data; a bidirectional mapping relationship table between the teaching and training sub-region and the virtual training unit is constructed.

[0031] Furthermore, the evaluation value calculation and factor calculation module includes an evaluation value calculation unit and a factor calculation unit;

[0032] The evaluation value calculation unit: Based on the virtual training unit, it obtains the soil moisture data, ambient temperature data, air humidity data and light intensity data of the corresponding teaching and training sub-region at the data sampling time point from the bidirectional mapping relationship table, and calculates the comprehensive evaluation value of the virtual training unit at the data sampling time point;

[0033] The factor calculation unit calculates the fluctuation factors of soil moisture data, ambient temperature data, air humidity data, and light intensity data based on the comprehensive evaluation value of the virtual training unit at the data sampling time point.

[0034] Furthermore, the fluctuation range calculation and anomaly determination module includes a fluctuation range calculation unit and an anomaly determination unit;

[0035] The fluctuation range calculation unit calculates the allowable fluctuation range of soil moisture data, ambient temperature data, air humidity data, and light intensity data of the virtual training unit at the data sampling time point, based on the soil moisture data fluctuation factor, ambient temperature data fluctuation factor, air humidity data fluctuation factor, and light intensity data fluctuation factor.

[0036] The anomaly determination unit acquires soil moisture data, ambient temperature data, air humidity data, and light intensity data of the virtual training unit at the next data sampling time point, and compares them with the allowable fluctuation range of soil moisture data, ambient temperature data, air humidity data, and light intensity data of the virtual training unit at the data sampling time point. If any data does not fall within the allowable fluctuation range, it determines that there is an abnormal environmental fluctuation at the next data sampling time point, issues an early warning to relevant personnel, and conducts dynamic teaching and training.

[0037] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: The AIOT-based smart garden virtual-real collaborative teaching and training method and system provided by this invention divides the real smart garden site into grids, constructing teaching and training sub-regions that correspond one-to-one with specific teaching types. Multi-source environmental perception data is collected within each sub-region to form a sub-region state dataset with time-series characteristics, thereby achieving a refined description of the teaching object and environmental state. Based on this, the aforementioned state data is synchronously mapped to the virtual teaching environment through a data interaction module, constructing virtual training units that strictly correspond to the real sub-regions and establishing a two-way mapping relationship. This allows the virtual training environment to realistically reflect the dynamic changes of the garden site, thus overcoming the limitations of traditional virtual... To address the disconnect between simulated teaching and real-world scenarios, this paper further explores a comprehensive evaluation and fluctuation factor calculation of multi-dimensional environmental parameters based on virtual training units. This allows for the analysis of changes in individual environmental parameters within the constraints of the overall environmental state, thereby improving the scientific rigor and stability of teaching evaluation and state assessment. Finally, by introducing a dynamic permissible fluctuation range based on fluctuation factors and comparing data from subsequent sampling time points with physical upper and lower bounds, the paper achieves accurate identification and timely warning of abnormal environmental fluctuations. Abnormal states are directly transformed into dynamic teaching and training content, thus enabling the teaching process to be driven by a real environment, with reproducible scenarios and teachable anomalies. Ultimately, this effectively enhances the authenticity, interactivity, and practical guidance value of smart garden teaching and training. Attached Figure Description

[0038] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0039] Figure 1 This is a schematic diagram illustrating the steps of the AIOT-based smart garden virtual-real collaborative teaching and training method of the present invention;

[0040] Figure 2 This is a schematic diagram of the structure of the AIOT-based smart garden virtual-real collaborative teaching and training system of the present invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] Please see Figure 1 In this first embodiment: a smart garden virtual-real collaborative teaching and training method based on AIoT is provided, which includes the following steps:

[0043] Step S1: Divide the target smart garden site into several teaching and training sub-regions and construct a set of teaching and training sub-regions; construct a sub-region status dataset of the teaching and training sub-regions at the data sampling time point.

[0044] Specifically, the target smart garden site area for teaching and training within the current teaching cycle is obtained, and grid-based partitioning technology is used to divide the target smart garden site area into several teaching and training sub-areas. Each teaching and training sub-area corresponds to one teaching type within one teaching cycle; a set of teaching and training sub-areas is then constructed. ,in, Let i represent the i-th teaching and training sub-region, and I represent the total number of teaching and training sub-regions.

[0045] Sensing data collection points are deployed within the teaching and training sub-area. These points include soil moisture sensors, ambient temperature sensors, air humidity sensors, and light intensity sensors. Based on these sensing data collection points, data about the teaching and training sub-area is obtained. Soil moisture data, ambient temperature data, air humidity data, and light intensity data.

[0046] Furthermore, the data sampling time period is constructed, denoted as . ,in, Let A represent the a-th data sampling time point, and A represent the total number of data sampling time points; the data sampling time points are respectively... Teaching and training sub-regions collected below The soil moisture data, ambient temperature data, air humidity data, and light intensity data are denoted as... , , and ;

[0047] Construct teaching and training sub-regions At the data sampling time point The sub-region state dataset is denoted as follows: .

[0048] In this invention, the target smart garden site is divided into grids, and each sub-region is associated with a unique teaching type within each teaching cycle. Multiple environmental sensing data collection points are deployed within each sub-region, constructing a temporal state data system with the sub-region as the smallest teaching unit. This sub-region division process achieves a precise binding between the teaching object and the spatial environment, allowing teaching and training to be conducted independently by region, rather than relying on the entire site. This improves the flexibility and scalability of teaching organization. Unified collection and identification of multi-source sensing data enables the quantification and traceability of environmental states, providing a reliable data foundation for subsequent evaluation and early warning. The introduction of the time dimension provides a continuously evolving state description for teaching and training, elevating teaching analysis from static display to dynamic process analysis, which helps cultivate students' understanding of the changing patterns of the garden environment.

[0049] Step S2: Transmit the sub-region status dataset synchronously to the virtual teaching environment through the data interaction module in the teaching platform, and construct virtual training units corresponding to the teaching and training sub-regions.

[0050] Specifically, the teaching and training sub-areas will be divided into... At the data sampling time point Sub-region state dataset The data is synchronously transmitted to the virtual teaching environment through the data interaction module in the teaching platform, and integrated with the teaching and training sub-areas. Corresponding virtual training unit The details are as follows:

[0051] Teaching and training sub-regions based on transmission to the virtual teaching environment At the data sampling time point Sub-region state dataset In the virtual teaching environment, the environmental parameters of the virtual training unit are initialized, including soil moisture data, ambient temperature data, air humidity data, and light intensity data.

[0052] Construct teaching and training sub-regions With virtual training units A bidirectional mapping table between them is denoted as .

[0053] In this invention, the state data of real sub-regions are synchronously mapped to the virtual teaching environment to construct corresponding virtual training units. A two-way mapping relationship is established between the real sub-regions and the virtual training units. Through the synchronous construction steps of virtual and real, the virtual teaching environment is no longer a preset model, but a "digital twin teaching unit" that reflects the real state of the garden in real time, significantly improving the authenticity of virtual training. The two-way mapping mechanism provides a structural foundation for subsequent virtual analysis to guide real teaching and management, avoiding the virtual system from becoming a one-way display tool. Through unified data sources and mapping rules, the data separation problem between virtual teaching and on-site teaching is eliminated, enabling virtual and real teaching to be compared and coordinated under the same evaluation framework.

[0054] Step S3: Based on the virtual training unit, obtain the environmental data of the corresponding teaching and training sub-region at the data sampling time point, and calculate the comprehensive evaluation value and data fluctuation factor.

[0055] Specifically, based on virtual training units From the bidirectional mapping table Obtain the corresponding teaching and training sub-region from the middle At the data sampling time point The soil moisture data, ambient temperature data, air humidity data, and light intensity data are recorded separately as follows: , , and And calculate the virtual training unit At the data sampling time point The comprehensive evaluation value is calculated using the following formula:

[0056] ;

[0057] in, Represents virtual training unit At the data sampling time point The overall evaluation value is as follows: This represents the weighting coefficient of the preset soil moisture data. This represents the weighting coefficient of the preset ambient temperature data. This represents the weighting coefficient of the preset air humidity data. This represents the weighting coefficient of the preset light intensity data. , , and This represents the preset ideal values ​​for soil moisture, ambient temperature, air humidity, and light intensity. , , and This indicates the maximum physiological fluctuation range of the preset soil moisture, ambient temperature, air humidity, and light intensity;

[0058] It should be noted that the core function of each parameter is to eliminate the difference in units, that is, the units of soil moisture (percentage), temperature (°C), and light intensity (lux) are different and cannot be directly added. Through this calculation, all parameters are converted into a deviation coefficient relative to the ideal value. The closer the GSI is to 0, the closer the environment is to the ideal state; the further it deviates from 0 (positive / negative), the more unsuitable the environment is (e.g., a GSI of -0.8 may indicate that the soil moisture is too low; a GSI of +0.6 may indicate that the temperature is too high). This avoids the one-sidedness of traditional evaluations that "only use absolute values ​​to judge (e.g., a temperature above 30℃ is abnormal)"—for example, although a temperature of 28℃ is higher than the ideal value of 25℃, it is still within the physiological fluctuation range (15-35℃). After normalization, the contribution of this item is only 0.1, and the GSI may still be close to 0, which will not be misjudged as an environmental anomaly, making the evaluation more in line with the actual laws of plant growth.

[0059] Based on virtual training units At the data sampling time point The comprehensive evaluation value is used to calculate the fluctuation factor of soil moisture data, ambient temperature data, air humidity data, and light intensity data. The calculation formula is as follows:

[0060] ;

[0061] in, , , and These represent the fluctuation factors for soil moisture data, ambient temperature data, air humidity data, and light intensity data, respectively.

[0062] It should be noted that the denominator is introduced The core principle is that the better the overall environment, the lower the tolerance for fluctuations in a single parameter; the worse the overall environment, the higher the tolerance, thus avoiding the neglect of small fluctuations when the overall environment is suitable or the oversensitivity to trigger large fluctuations when the overall environment is unsuitable. The denominator is The reason is that light intensity has a more significant instantaneous impact on most garden plants (such as strong sunlight exposure may quickly cause leaves to wither), and they need to be more sensitive to its fluctuations. Even if the overall environment is good, small deviations in light intensity will be amplified into larger fluctuation factors, ensuring the capture of key emergency scenarios in teaching.

[0063] In teaching, if the light intensity suddenly exceeds the ideal value, its fluctuation factor will be amplified (the denominator will be smaller), and the subsequent allowable fluctuation range will be narrower, making it easier to trigger an early warning. This can quickly capture emergency scenarios such as sudden changes in light intensity, guide students to learn practical skills such as setting up shade nets and adjusting supplementary lighting equipment, and improve the relevance of practical training.

[0064] In this invention, the comprehensive evaluation value calculation process enables a holistic characterization of the suitability of the garden environment for teaching, avoiding the one-sidedness of judging the quality of the environment based on a single indicator. By constructing a fluctuation factor, changes in environmental parameters are analyzed within the context of the overall teaching environment, effectively improving the rationality and stability of anomaly judgment. By introducing a mechanism for regulating individual fluctuations through overall evaluation, the system's adaptability to complex environmental coupling changes is enhanced, avoiding the misjudgment of local normal fluctuations as teaching anomalies.

[0065] Step S4: Based on the data fluctuation factor, calculate the allowable fluctuation range of the virtual training unit at the data sampling time point; obtain the environmental data of the virtual training unit at the next data sampling time point and compare it with the allowable fluctuation range. If there is abnormal environmental fluctuation, issue an early warning to relevant personnel and conduct dynamic teaching and training.

[0066] Specifically, based on soil moisture data fluctuation factors Ambient temperature data fluctuation factor Air humidity data fluctuation factor and light intensity data fluctuation factor Computational Virtual Training Unit At the data sampling time point The permissible fluctuation ranges for soil moisture data, ambient temperature data, air humidity data, and light intensity data are as follows:

[0067] ;

[0068] in, , , and These represent the soil moisture data, ambient temperature data, air humidity data, and light intensity data at the data sampling time points, respectively. The allowable fluctuation range below and These represent the preset upper and lower physical boundaries of soil moisture data, respectively. and These represent the upper and lower physical boundaries of the preset ambient temperature data, respectively. and These represent the upper and lower bounds of the preset air humidity data, respectively. and These represent the upper and lower bounds of the preset light intensity data (preset based on the effective range of the sensor collecting the corresponding environmental parameters, and in conjunction with the physiological safety range of the teaching and training subjects).

[0069] It should be noted that the allowable fluctuation range is not a fixed value, but is dynamically calculated based on the measured data of the previous sampling point and the fluctuation factor. That is, the range is different for each sampling time point, which is completely synchronized with the dynamic changes of the real garden environment. The abnormal threshold of traditional virtual training is fixed (such as alarming when the soil moisture is below 40%), which cannot adapt to environmental changes such as "sunny day to cloudy day" and "dry season to rainy season". However, the dynamic range of this formula will be adjusted according to the environmental conditions. For example, during the rainy season, the soil moisture base is high, and the upper limit of the fluctuation range will be appropriately widened to avoid the humidity increase caused by normal rainfall being misjudged as abnormal, so as to make the virtual training highly synchronized with the real environment.

[0070] Taking soil moisture as an example, the lower limit of the allowable fluctuation range is the current measured value minus the maximum value of the fluctuation factor and the physical lower limit (to avoid the range being lower than the plant's survival baseline); the upper limit is the current measured value plus the minimum value of the fluctuation factor and the physical upper limit (to avoid the range being higher than the survival upper limit) - taking into account both dynamic fluctuation factors and ensuring the physiological safety of plants.

[0071] The interval employs a dual constraint, considering both dynamic fluctuation factors (flexibility) and limiting it to the physiologically safe range of plants and the sensor's measurement range (safety). For example, the physiologically safe range of soil moisture is 20%-90%. If the previous sampling point measured 60%, the fluctuation factor calculates a "lower limit of 10%", and the final lower limit of the interval remains 20% (the minimum physiologically safe value). This prevents the interval from exceeding the plant's survival range due to excessive fluctuation factors. This design ensures that genuine anomalies are not missed, nor does dynamic adjustment lead to dangerous thresholds, guaranteeing the safety and scientific rigor of the training.

[0072] Obtaining virtual training units At the data sampling time point The data includes soil moisture, ambient temperature, air humidity, and light intensity, and is compared with those of the virtual training unit. At the data sampling time point The allowable fluctuation ranges of soil moisture data, ambient temperature data, air humidity data, and light intensity data are compared. If any data does not fall within the allowable fluctuation range, the data sampling time point is determined. If abnormal environmental fluctuations occur, an early warning will be issued to relevant personnel and dynamic teaching and practical training will be conducted, as detailed below:

[0073] Virtual training units At the data sampling time point The soil moisture data, ambient temperature data, air humidity data, and light intensity data are mapped to the virtual garden model; corresponding abnormal scenario states are generated in the virtual environment; and abnormal environmental elements are associated with plant growth status, disease risk, or teaching indicators for display.

[0074] Analyze the causes of anomalies (such as insufficient irrigation, light blockage, etc.), select or adjust coping strategies in a virtual environment, and observe the impact of different strategies on the evolution of the virtual garden state.

[0075] Once an effective adjustment strategy is determined in the virtual training, the corresponding strategy parameters are sent to the execution equipment in the physical garden through the AIoT system to make targeted adjustments to abnormal environmental elements and continuously monitor the adjustment effect.

[0076] In this invention, by constructing a dynamic allowable fluctuation range, the anomaly judgment criteria are made to change with the environmental state, avoiding false alarms or missed alarms caused by static thresholds; through the physical boundary constraint mechanism, the anomaly analysis in virtual teaching is ensured to have realistic rationality, preventing it from deviating from the actual garden operation conditions; through the linkage between anomaly triggering and teaching, environmental changes are directly transformed into teaching scenarios and training content, enabling students to conduct analysis and decision-making training in real anomaly backgrounds, significantly improving the practicality of teaching and emergency cognition ability.

[0077] Please see Figure 2 In this second embodiment: a smart garden virtual-real collaborative teaching and training system based on AIoT is provided. The system includes: a set construction module, a data transmission module, an evaluation value calculation and factor calculation module, and a fluctuation range calculation and anomaly judgment module.

[0078] The set construction module: divides the target smart garden site area into several teaching and training sub-areas, constructs a set of teaching and training sub-areas; and constructs a sub-area status dataset of the teaching and training sub-areas at the data sampling time point.

[0079] The data transmission module synchronously transmits the sub-region status dataset to the virtual teaching environment through the data interaction module in the teaching platform, and constructs virtual training units corresponding to the teaching and training sub-regions.

[0080] The evaluation value calculation and factor calculation module: Based on the virtual training unit, it obtains the environmental data of the corresponding teaching and training sub-region at the data sampling time point, and calculates the comprehensive evaluation value and data fluctuation factor;

[0081] The fluctuation range calculation and anomaly judgment module calculates the allowable fluctuation range of the virtual training unit at the data sampling time point based on the data fluctuation factor; obtains the environmental data of the virtual training unit at the next data sampling time point and compares it with the allowable fluctuation range; if there is an abnormal environmental fluctuation, it issues an early warning to relevant personnel and conducts dynamic teaching and training.

[0082] Furthermore, the data transmission module includes a data transmission unit;

[0083] The data transmission unit synchronously transmits the sub-region state dataset of the teaching and training sub-region at the data sampling time point to the virtual teaching environment through the data interaction module in the teaching platform, and constructs a virtual training unit corresponding to the teaching and training sub-region. Specifically, based on the sub-region state dataset of the teaching and training sub-region at the data sampling time point transmitted to the virtual teaching environment, the environmental parameters of the virtual training unit are initialized in the virtual teaching environment, including soil moisture data, ambient temperature data, air humidity data, and light intensity data; a bidirectional mapping relationship table between the teaching and training sub-region and the virtual training unit is constructed.

[0084] Furthermore, the evaluation value calculation and factor calculation module includes an evaluation value calculation unit and a factor calculation unit;

[0085] The evaluation value calculation unit: Based on the virtual training unit, it obtains the soil moisture data, ambient temperature data, air humidity data and light intensity data of the corresponding teaching and training sub-region at the data sampling time point from the bidirectional mapping relationship table, and calculates the comprehensive evaluation value of the virtual training unit at the data sampling time point;

[0086] The factor calculation unit calculates the fluctuation factors of soil moisture data, ambient temperature data, air humidity data, and light intensity data based on the comprehensive evaluation value of the virtual training unit at the data sampling time point.

[0087] Furthermore, the fluctuation range calculation and anomaly determination module includes a fluctuation range calculation unit and an anomaly determination unit;

[0088] The fluctuation range calculation unit calculates the allowable fluctuation range of soil moisture data, ambient temperature data, air humidity data, and light intensity data of the virtual training unit at the data sampling time point, based on the soil moisture data fluctuation factor, ambient temperature data fluctuation factor, air humidity data fluctuation factor, and light intensity data fluctuation factor.

[0089] The anomaly determination unit acquires soil moisture data, ambient temperature data, air humidity data, and light intensity data of the virtual training unit at the next data sampling time point, and compares them with the allowable fluctuation range of soil moisture data, ambient temperature data, air humidity data, and light intensity data of the virtual training unit at the data sampling time point. If any data does not fall within the allowable fluctuation range, it determines that there is an abnormal environmental fluctuation at the next data sampling time point, issues an early warning to relevant personnel, and conducts dynamic teaching and training.

[0090] In this third embodiment, a smart garden virtual-real collaborative teaching and training method based on AIoT is provided, and the application of this method on a physical teaching and training device is specifically described to verify the beneficial effects of the present invention.

[0091] The physical teaching and training device includes an environmental perception module, a central control and execution module, and a human-computer interaction and communication module. The environmental perception module includes a meteorological perception unit, a soil perception unit, and an image perception unit. The central control and execution module includes a central controller, a PLC control unit, and an execution device. The human-computer interaction and communication module includes a 10-inch touch screen, a wireless router, and a training cloud platform.

[0092] The meteorological sensing unit includes a wind direction sensor, a wind speed sensor, a temperature and humidity sensor, a light sensor, a carbon dioxide sensor, and a PM2.5 / PM100 sensor, which are used to collect meteorological parameters such as air temperature, humidity, light intensity, and air quality.

[0093] The soil sensing unit includes a soil temperature and humidity sensor, a soil nitrogen and potassium sensor, a soil phosphorus sensor, a soil EC value sensor, and a soil pH value sensor, which are used to collect soil parameters such as soil moisture, nutrient content, electrical conductivity, and pH.

[0094] The image sensing unit includes a remote observation instrument (camera) for collecting visual data on plant growth status and pest and disease conditions.

[0095] The central controller includes 8-channel and 12-channel central controllers, used to receive sensor data and issue control commands;

[0096] The PLC control unit includes a basic PLC unit and a four-channel analog input expansion module for logic control and analog signal processing.

[0097] The execution equipment includes a water pump (for irrigation), a fertilization module (for precision fertilization), a full-spectrum supplemental lighting (for light regulation), and a soil-grown planting trough, supporting active control of the garden environment.

[0098] The application of this method in physical teaching and training devices can refer to the following approaches:

[0099] The smart garden teaching area represented by the training platform (such as a smart greenhouse on campus or an outdoor training base) is divided into several sub-areas, each sub-area corresponding to an independent teaching type (such as a soil nutrient management training area or a microclimate control training area).

[0100] Within each sub-region, the corresponding sensors from the aforementioned environmental sensing module are deployed. For example, soil nitrogen, phosphorus, and potassium sensors, EC value sensors, and pH value sensors are deployed in the soil nutrient management training area; and temperature, humidity, light, and carbon dioxide sensors are deployed in the microclimate control training area.

[0101] The central controller and PLC unit collect data on soil moisture, ambient temperature, air humidity, and light intensity (corresponding to the four core parameters in this invention) of each sub-region according to a preset sampling period (e.g., every 10 minutes). At the same time, extended parameters (such as soil nutrients, CO2 concentration, etc.) can also be collected to form a sub-region status dataset.

[0102] The collected sub-region status dataset is uploaded to the training cloud platform (i.e., the virtual teaching environment in this invention) in real time via a wireless router.

[0103] Based on the uploaded data, the cloud platform creates virtual training units in the virtual environment that correspond one-to-one with the real sub-regions, and visualizes parameters such as soil moisture, temperature distribution, and light intensity in the 3D model.

[0104] Establish a two-way mapping table to ensure that each real sensor data point is synchronized with the corresponding data node in the virtual model in real time.

[0105] In the cloud platform, for each virtual training unit, the comprehensive evaluation value calculation formula in this invention is called, and combined with the real collected soil moisture, temperature, humidity and light data, the comprehensive evaluation value of the unit is calculated.

[0106] Furthermore, based on the comprehensive evaluation value, the fluctuation factor of each parameter is calculated. For example, if the light intensity of a certain sub-region changes abruptly (such as due to weather changes), the system will dynamically adjust the fluctuation tolerance of the parameter according to the current comprehensive environmental status.

[0107] In addition to the four core parameters, the system can also incorporate parameters such as soil nutrients and carbon dioxide concentration into the evaluation system, and achieve a more comprehensive environmental status assessment by configuring corresponding weights.

[0108] The system continuously monitors the data at the next sampling point based on the calculated allowable fluctuation range. For example, if the soil moisture in a certain sub-region suddenly falls below the lower limit of the allowable range, the system determines that the irrigation is abnormal.

[0109] Once an anomaly is identified, the cloud platform immediately issues an alert to both teachers and students, highlights the abnormal area in the virtual environment, and simultaneously pushes relevant teaching materials, such as measures to address soil drought.

[0110] Students can conduct dynamic practical training in a virtual environment, such as simulating the adjustment of irrigation strategies and setting up fertilization plans. They can also send instructions to the actual execution equipment on the training platform (such as starting a water pump to replenish water) and observe the feedback from the actual environment, forming a closed-loop teaching process.

[0111] For example, if the system detects a continuous decrease in nitrogen content in the soil nutrient management training area and the nitrogen content exceeds the allowable fluctuation range through the soil nitrogen, phosphorus and potassium sensors, the cloud platform will mark a low nitrogen warning in the corresponding virtual unit and display the types of plants that may be affected (such as leafy vegetables being hindered from growing).

[0112] After receiving the warning, the teacher can immediately issue a practical training task on soil nitrogen supplementation regulation, requiring students to analyze the cause and design a regulation plan;

[0113] Students simulate adding nitrogen fertilizer in a virtual environment, observe the growth response curve of virtual plants, and learn about the coupled effects of different fertilizer application rates on soil EC and pH values.

[0114] After determining the optimization plan, instructions are sent to the fertilization module of the training platform through the platform to execute precise nitrogen supplementation and monitor changes in soil parameters in real time.

[0115] The system automatically records students' operation process, environmental response data, and control effects, generates training reports, and supports teaching evaluation and feedback.

[0116] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0117] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A smart garden virtual-real collaborative teaching and training method based on AIoT, characterized in that, The method includes the following steps: Step S1: Divide the target smart garden site into several teaching and training sub-regions and construct a set of teaching and training sub-regions; construct a sub-region status dataset of the teaching and training sub-regions at the data sampling time point; Step S2: Transmit the sub-region status dataset synchronously to the virtual teaching environment through the data interaction module in the teaching platform, and construct virtual training units corresponding to the teaching and training sub-regions; Step S3: Based on the virtual training unit, obtain the environmental data of the corresponding teaching and training sub-area at the data sampling time point, and calculate the comprehensive evaluation value and data fluctuation factor; Step S4: Based on the data fluctuation factor, calculate the allowable fluctuation range of the virtual training unit at the data sampling time point; obtain the environmental data of the virtual training unit at the next data sampling time point and compare it with the allowable fluctuation range. If there is abnormal environmental fluctuation, issue an early warning to relevant personnel and conduct dynamic teaching and training.

2. The AIOT-based smart garden virtual-real collaborative teaching and training method according to claim 1, characterized in that, The specific implementation process of step S1 includes: The target smart garden site area for teaching and training within the current teaching cycle is obtained, and using grid partitioning technology, it is divided into several teaching and training sub-regions. Each teaching and training sub-region corresponds to one teaching type within one teaching cycle; a set of teaching and training sub-regions is then constructed. ,in, Let i represent the i-th teaching and training sub-region, and I represent the total number of teaching and training sub-regions. Sensing data collection points are deployed within the teaching and training sub-area. These points include soil moisture sensors, ambient temperature sensors, air humidity sensors, and light intensity sensors. Based on these sensing data collection points, data about the teaching and training sub-area is obtained. Soil moisture data, ambient temperature data, air humidity data, and light intensity data.

3. The AIOT-based smart garden virtual-real collaborative teaching and training method according to claim 2, characterized in that, The specific implementation process of step S1 also includes: The data sampling time period is defined as follows: ,in, Let A represent the a-th data sampling time point, and A represent the total number of data sampling time points; the data sampling time points are respectively... Teaching and training sub-regions collected below The soil moisture data, ambient temperature data, air humidity data, and light intensity data are denoted as... , , and ; Construct teaching and training sub-regions At the data sampling time point The sub-region state dataset is denoted as follows: .

4. The AIOT-based smart garden virtual-real collaborative teaching and training method according to claim 3, characterized in that, The specific implementation process of step S2 includes: Teaching and training sub-areas At the data sampling time point Sub-region state dataset The data is synchronously transmitted to the virtual teaching environment through the data interaction module in the teaching platform, and integrated with the teaching and training sub-areas. Corresponding virtual training unit The details are as follows: Teaching and training sub-regions based on transmission to the virtual teaching environment At the data sampling time point Sub-region state dataset In the virtual teaching environment, the environmental parameters of the virtual training unit are initialized, including soil moisture data, ambient temperature data, air humidity data, and light intensity data. Construct teaching and training sub-regions With virtual training units A bidirectional mapping table between them is denoted as .

5. The AIOT-based smart garden virtual-real collaborative teaching and training method according to claim 4, characterized in that, The specific implementation process of step S3 includes: Based on virtual training units From the bidirectional mapping table Obtain the corresponding teaching and training sub-regions from the middle. At the data sampling time point The soil moisture data, ambient temperature data, air humidity data, and light intensity data are recorded separately as follows: , , and And calculate the virtual training unit At the data sampling time point The comprehensive evaluation value is calculated using the following formula: ； in, Virtual training unit At the data sampling time point The overall evaluation value is as follows: This represents the weighting coefficient of the preset soil moisture data. This represents the weighting coefficient of the preset ambient temperature data. This represents the weighting coefficient of the preset air humidity data. This represents the weighting coefficient of the preset light intensity data. , , and This represents the preset ideal values ​​for soil moisture, ambient temperature, air humidity, and light intensity. , , and This indicates the maximum physiological fluctuation range of the preset soil moisture, ambient temperature, air humidity, and light intensity; Based on virtual training units At the data sampling time point The comprehensive evaluation value is used to calculate the fluctuation factor of soil moisture data, ambient temperature data, air humidity data, and light intensity data. The calculation formula is as follows: ； in, , , and These represent the fluctuation factors for soil moisture data, ambient temperature data, air humidity data, and light intensity data, respectively.

6. The AIOT-based smart garden virtual-real collaborative teaching and training method according to claim 5, characterized in that, The specific implementation process of step S4 includes: Based on soil moisture data fluctuation factor Ambient temperature data fluctuation factor Air humidity data fluctuation factor and light intensity data fluctuation factor Computational Virtual Training Unit At the data sampling time point The permissible fluctuation ranges for soil moisture data, ambient temperature data, air humidity data, and light intensity data are as follows: ； in, , , and These represent the soil moisture data, ambient temperature data, air humidity data, and light intensity data at the data sampling time points, respectively. The allowable fluctuation range below and These represent the preset upper and lower physical boundaries of soil moisture data, respectively. and These represent the upper and lower physical boundaries of the preset ambient temperature data, respectively. and These represent the upper and lower bounds of the preset air humidity data, respectively. and These represent the upper and lower physical bounds of the preset illumination intensity data, respectively. Obtaining virtual training units At the data sampling time point The data includes soil moisture, ambient temperature, air humidity, and light intensity, and is compared with those of the virtual training unit. At the data sampling time point The allowable fluctuation ranges of soil moisture data, ambient temperature data, air humidity data, and light intensity data are compared. If any data does not fall within the allowable fluctuation range, the data sampling time point is determined. If there are abnormal fluctuations in the environment, an early warning will be issued to relevant personnel and dynamic teaching and training will be conducted.

7. An AIOT-based smart garden virtual-real collaborative teaching and training system, executing the AIOT-based smart garden virtual-real collaborative teaching and training method as described in any one of claims 1-6, characterized in that, The system includes: a set construction module, a data transmission module, an evaluation value calculation and factor calculation module, and a fluctuation range calculation and anomaly determination module; The set construction module: divides the target smart garden site area into several teaching and training sub-areas, constructs a set of teaching and training sub-areas; and constructs a sub-area status dataset of the teaching and training sub-areas at the data sampling time point. The data transmission module synchronously transmits the sub-region status dataset to the virtual teaching environment through the data interaction module in the teaching platform, and constructs virtual training units corresponding to the teaching and training sub-regions. The evaluation value calculation and factor calculation module: Based on the virtual training unit, it obtains the environmental data of the corresponding teaching and training sub-region at the data sampling time point, and calculates the comprehensive evaluation value and data fluctuation factor; The fluctuation range calculation and anomaly judgment module calculates the allowable fluctuation range of the virtual training unit at the data sampling time point based on the data fluctuation factor; obtains the environmental data of the virtual training unit at the next data sampling time point and compares it with the allowable fluctuation range; if there is an abnormal environmental fluctuation, it issues an early warning to relevant personnel and conducts dynamic teaching and training.

8. The AIOT-based smart garden virtual-real collaborative teaching and training system according to claim 7, characterized in that: The data transmission module includes a data transmission unit; The data transmission unit synchronously transmits the sub-region state dataset of the teaching and training sub-region at the data sampling time point to the virtual teaching environment through the data interaction module in the teaching platform, and constructs a virtual training unit corresponding to the teaching and training sub-region. Specifically, based on the sub-region state dataset of the teaching and training sub-region at the data sampling time point transmitted to the virtual teaching environment, the environmental parameters of the virtual training unit are initialized in the virtual teaching environment, including soil moisture data, ambient temperature data, air humidity data, and light intensity data; a bidirectional mapping relationship table between the teaching and training sub-region and the virtual training unit is constructed.

9. The AIOT-based smart garden virtual-real collaborative teaching and training system according to claim 8, characterized in that: The evaluation value calculation and factor calculation module includes an evaluation value calculation unit and a factor calculation unit; The evaluation value calculation unit: Based on the virtual training unit, it obtains the soil moisture data, ambient temperature data, air humidity data and light intensity data of the corresponding teaching and training sub-region at the data sampling time point from the bidirectional mapping relationship table, and calculates the comprehensive evaluation value of the virtual training unit at the data sampling time point; The factor calculation unit calculates the fluctuation factors of soil moisture data, ambient temperature data, air humidity data, and light intensity data based on the comprehensive evaluation value of the virtual training unit at the data sampling time point.

10. The AIOT-based smart garden virtual-real collaborative teaching and training system according to claim 9, characterized in that: The fluctuation range calculation and anomaly determination module includes a fluctuation range calculation unit and an anomaly determination unit; The fluctuation range calculation unit calculates the allowable fluctuation range of soil moisture data, ambient temperature data, air humidity data, and light intensity data of the virtual training unit at the data sampling time point, based on the soil moisture data fluctuation factor, ambient temperature data fluctuation factor, air humidity data fluctuation factor, and light intensity data fluctuation factor. The anomaly determination unit acquires soil moisture data, ambient temperature data, air humidity data, and light intensity data of the virtual training unit at the next data sampling time point, and compares them with the allowable fluctuation range of soil moisture data, ambient temperature data, air humidity data, and light intensity data of the virtual training unit at the data sampling time point. If any data does not fall within the allowable fluctuation range, it determines that there is an abnormal environmental fluctuation at the next data sampling time point, issues an early warning to relevant personnel, and conducts dynamic teaching and training.

Images