Monitoring system for induction effect of sugarcane flowering phase

Abstract

The invention, which relates to the technical field of sugarcane monitoring, discloses a monitoring system for the induction effect of a sugarcane flowering phase, the system comprises a multi-dimensional sensing layer, an intelligent decision-making center and a precise execution layer. The multi-dimensional sensing layer collects illumination, temperature, phenological period, stress index and ScFT6 gene expression data through an environmental factor sensing module, a plant physiological status sensing module and a key gene expression real-time monitoring module; the intelligent decision center corrects a photoperiod and a temperature threshold through a dynamic threshold regulation and control model, and evaluates an induction effect in combination with gene expression; the precise execution layer executes shading, light supplement and temperature and humidity adjustment to form closed-loop regulation, and the system realizes dynamic precise regulation, improves the induction effect stability, adapts to complex environments and variety differences, and is suitable for sugarcane flowering phase induction monitoring.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sugarcane monitoring, in particular to a monitoring system for the induction effect of sugarcane flowering period. BACKGROUND

[0002] Sugarcane is an important sugar crop and economic crop, and its sexual hybridization breeding is the key way to improve varieties. The premise of successful hybridization is to ensure that the parents meet in the flowering period, so it is crucial to artificially induce and regulate the flowering period of sugarcane. Photoperiod induction is a core technical means to regulate the flowering of sugarcane, and its effect is directly related to the success or failure of hybridization breeding. In the prior art, the flowering of sugarcane is usually induced or delayed by controlling the length of light, but the actual induction effect is influenced by many factors such as variety characteristics, plant physiological state and environmental factors, and it is often difficult to achieve precise induction by relying on the regulation of a single environmental factor.

[0003] In the prior art, there are some invention patents related to the regulation of the flowering period of sugarcane. For example, patent (publication number CN111338280A) discloses a control system for inducing the flowering of sugarcane based on photoperiod. The system controls the sugarcane car carrying the sugarcane parents to enter the light induction room through the computer, data switch and PLC control station, controls the opening or closing of the light source according to the comparison result of the target light duration and the natural light duration, so as to realize the stable and precise control of the light duration, and make the flowering of sugarcane not affected by environmental factors. Another patent (publication number CN103081738A) relates to a method for regulating the induction and natural flowering of innovative sugarcane parents to meet the flowering period. The method controls the induction start time, photoperiod induction and temperature and humidity, aiming to solve the problem of low pollen development rate due to the mismatch of the flowering period of the induced parents and the naturally flowering parents. However, the above prior art solutions mainly focus on the one-way and programmed control of environmental factors, and lack real-time sensing and response to the physiological state changes of the plant body during the induction process. The control logic relies on preset fixed environmental parameter thresholds, and cannot adaptively adjust according to the real-time physiological stress and internal flowering signals of sugarcane, so the stability and precision of the induction effect need to be improved when dealing with complex field environments and variety differences.

[0004] Therefore, there is an urgent need in the art for a monitoring system that can break through the limitations of one-way environmental control. The key is to solve the problem of how to realize the transition from one-way environmental control to dynamic and precise regulation based on the physiological state sensing of the plant body in the induction of the flowering period of sugarcane. The ideal system should be able to integrate environmental data and physiological and even molecular level information of the plant body, form a closed-loop regulation system that can provide real-time feedback and dynamic decision-making, and thus truly realize the precise monitoring and optimal control of the induction effect of the flowering period. SUMMARY

[0005] The present application aims to make up for the shortcomings of the prior art, and provides a sugarcane flowering induction effect monitoring system, which integrates field portable gene expression quantitative analysis, multi-spectral phenology and stress perception, and light and temperature environment regulation into an adaptive intelligent system, so as to realize the fundamental change from blind execution of fixed photoperiod to dynamic optimization of induction strategy according to actual physiological response of plants, and overcome the shortcomings of the prior art.

[0006] The present application aims to make up for the shortcomings of the prior art, and provides a sugarcane flowering induction effect monitoring system, which integrates field portable gene expression quantitative analysis, multi-spectral phenology and stress perception, and light and temperature environment regulation into an adaptive intelligent system, so as to realize the fundamental change from blind execution of fixed photoperiod to dynamic optimization of induction strategy according to actual physiological response of plants, and overcome the shortcomings of the prior art.

[0007] The multi-dimensional perception layer is used for collecting environmental factor data, plant physiological state data and key gene expression data, the environmental factor data includes light duration and temperature, the plant physiological state data includes phenological stage and stress index, and the key gene expression data includes the relative expression amount of ScFT6 gene.

[0008] The intelligent decision-making center is used for receiving the data of the multi-dimensional perception layer and generating regulation instructions through a dynamic threshold regulation model, the dynamic threshold regulation model dynamically corrects the photoperiod threshold and the temperature threshold based on the phenological stage and the stress index, and combines the relative expression amount of the ScFT6 gene to evaluate the induction effect.

[0009] The precise execution layer is used for receiving the regulation instructions and executing light regulation and irrigation operation, and the precise execution layer includes a movable light-shading and light-supplementing device and an intelligent irrigation system.

[0010] The environmental factor perception module, the plant physiological state perception module and the key gene expression real-time monitoring module communicate with the intelligent decision-making center through an Internet of Things module.

[0011] The intelligent decision-making center fuses and analyzes the environmental factor data, the plant physiological state data and the key gene expression data through an algorithm model.

[0012] The movable light-shading and light-supplementing device performs zoning light shading or light supplementing based on the photoperiod threshold.

[0013] The intelligent irrigation system performs irrigation based on the stress index.

[0014] Further, the photosynthetically active radiation sensor in the environmental factor perception module measures photosynthetically active radiation intensity and calculates daily light duration, and the temperature and humidity sensor measures air temperature and relative humidity and generates a temperature curve, the Internet of Things module transmits environmental factor data using LoRa or 4G communication protocol, the environmental factor perception module further includes a data acquisition node that periodically reads sensor data and sends it to the intelligent decision-making hub through the Internet of Things module, the data acquisition node includes a microprocessor and a power management unit, the microprocessor is responsible for data preprocessing and caching, and the power management unit supports solar power supply and battery backup, the grid deployment is used to cover multiple areas of the sugarcane field to capture micro-environmental changes, the data preprocessing includes data filtering and outlier detection, the period of periodic reading can be configured, ranging from 1 minute to 1 hour, the environmental factor data is stored in time series format with time stamp and location identification.

[0015] Further, the multi-spectral unmanned aerial vehicle in the plant physiological state perception module regularly flies to collect multi-spectral image data, the multi-spectral image data includes normalized difference vegetation index NDVI and photochemical reflectance index PRI, the ground mobile robot carries a chlorophyll fluorometer to measure chlorophyll fluorescence parameters, the phenological stage is automatically determined by analyzing the NDVI historical curve and using a machine learning model, the machine learning model uses a time series clustering algorithm, the stress index is calculated by PRI data and soil moisture sensor data, the flight height and path of the multi-spectral unmanned aerial vehicle are programmable, the multi-spectral image data is subjected to geometric correction and radiation correction, the chlorophyll fluorescence parameters include maximum fluorescence yield and actual photosynthetic efficiency, the time series clustering algorithm uses the DBSCAN algorithm, the phenological stage includes the tillering stage, the elongation stage and the induction critical stage, the stress index includes the water stress index and the nutrient stress index, the water stress index is calculated based on the correlation between the PRI value and the soil moisture value, and the nutrient stress index is calculated based on the deviation of the NDVI value from the historical benchmark, the multi-spectral unmanned aerial vehicle and the ground mobile robot communicate with the intelligent decision-making hub through a wireless network.

[0016] Further, the portable field qPCR instrument in the key gene expression real-time monitoring module is used for non-destructive collection of sugarcane leaf tissue samples and real-time quantification of ScFT6 gene expression level, the portable field qPCR instrument comprises a sample processing unit and a detection unit, the sample processing unit integrates a rapid nucleic acid extraction kit, the detection unit is pre-stored with primers and probes of ScFT6 gene and a reference gene, the primers and probes are stored in the form of freeze-dried powder, the non-destructive collection uses a special sampler to collect leaf sheath tissue, the sample processing unit completes RNA extraction and reverse transcription within 10 minutes, the detection unit runs a qPCR program and outputs the relative expression amount of the ScFT6 gene, the relative expression amount is calculated by using a 2^-ΔΔCt method, the portable field qPCR instrument transmits data to the intelligent decision-making center through Bluetooth, the sampling cycle is 3 to 5 days, the sample processing unit supports batch processing of multiple samples, the detection unit has temperature control and fluorescence detection functions, the primers and probes are designed for ScFT6 gene and ScTPS1 gene, the reference gene comprises a UBQ gene, and the relative expression amount data is stored in association with the sampling time point and position.

[0017] Further, the dynamic threshold adjustment model in the intelligent decision-making center dynamically corrects the photoperiod threshold value by using the following mathematical formula:

[0018]

[0019] wherein, represents the corrected photoperiod threshold value, represents the reference photoperiod threshold value, represents a phenophase correction coefficient, represents a current phenophase value, represents a reference phenophase value, represents a stress correction coefficient, represents a current stress index value, represents a reference stress index value;

[0020] The phenophase value is assigned according to the phenophase category, the tillering period is assigned a value of 1, the elongation period is assigned a value of 2, and the induction critical period is assigned a value of 3, the reference phenophase value is set to 2, the stress index value is calculated by normalizing the water stress index, and ranges from 0 to 1, the reference stress index value is set to 0.5, the phenophase correction coefficient ranges from -0.1 to 0.1, and the stress correction coefficient ranges from -0.05 to 0.05;

[0021] The dynamic threshold control model also uses a temperature threshold correction formula, which has the same structure as the photoperiod threshold correction formula but different parameters. The reference photoperiod threshold... Based on historical data and literature settings, the phenological period values ​​and stress index values ​​are derived from the plant physiological state sensing module, and the corrected photoperiod threshold is used to generate control instructions for the precise execution layer.

[0022] Furthermore, the phenological period correction coefficient in the dynamic threshold control model and stress correction coefficient Adaptive adjustment is achieved through the following mathematical formula:

[0023]

[0024]

[0025] in This represents the initial value of the phenological period correction coefficient. Indicates the learning rate during the phenological period. This indicates the current expression level of the ScFT6 gene. This indicates the expression level of the target ScFT6 gene. This represents the initial value of the stress correction coefficient. Indicates the rate of forced learning. This indicates the current stress index value. This represents the optimal stress index value;

[0026] The initial value of the phenological period correction coefficient The phenological learning rate is set to 0.05. The current ScFT6 gene expression level is set to 0.1. The expression level of the target ScFT6 gene, obtained from the real-time monitoring module for key gene expression. The initial value of the stress correction coefficient is set based on variety-specific data. The stress learning rate is set to 0.02. The optimal stress index value is set to 0.05. Set to 0.2;

[0027] The adaptive adjustment is optimized using gradient descent based on historical successful induction data, and the phenological period learning rate is... and coerced learning rate The mathematical formula is dynamically updated during system operation and is used to correct for nonlinear changes in coefficients with gene expression and stress state.

[0028] Furthermore, the induction effect evaluation of the intelligent decision-making center is carried out by comparing environmental factor data with dynamic thresholds and combining them with the expression level of the ScFT6 gene. The induction effect evaluation includes three scenario judgments.

[0029] Scenario A indicates that the environmental data meets the criteria and ScFT6 gene expression is upregulated, which is considered a successful induction. Scenario B indicates that the environmental data meets the criteria but ScFT6 gene expression is not upregulated, which is considered a risk of induction failure. Scenario C indicates that the environmental data does not meet the criteria but ScFT6 gene expression is upregulated, which is considered a response from the plant.

[0030] The environmental data meeting the criteria refers to the light duration and temperature being within the corrected threshold range. The upregulation of ScFT6 gene expression refers to a relative increase in expression level exceeding 20%. The risk of induction failure triggers an early warning signal. The plant's response triggers the model learning mechanism. The early warning signal is displayed through the user interface. The model learning mechanism records environmental data and gene expression data to update the dynamic threshold regulation model. The comparison operation uses a numerical comparison algorithm, which includes threshold comparison and trend analysis. The trend analysis uses a sliding window to calculate the rate of change of ScFT6 gene expression level. The rate of change is calculated using a slope value. The sliding window size is the data from the three most recent samples. The induction effect evaluation period is synchronized with the sampling period.

[0031] Furthermore, the movable shading and supplemental lighting equipment in the precision execution layer moves through the sugarcane field based on a track system. This movable shading and supplemental lighting equipment includes a shading curtain and LED supplemental lights. The shading curtain reduces the duration of illumination, and the LED supplemental lights increase the light intensity. The intelligent irrigation system includes sprinkler heads and solenoid valves. The sprinkler heads control the irrigation volume based on a stress index. The track system covers multiple zones of the sugarcane field. The control commands include shading time, supplemental lighting intensity, and irrigation duration. The movable shading and supplemental lighting equipment receives photoperiod threshold commands and calculates the shading or supplemental lighting time. The intelligent irrigation system receives stress index commands and calculates irrigation volume. The shading time is determined by the difference between daily light duration and photoperiod threshold. The supplemental lighting intensity is determined by the difference between photosynthetically active radiation sensor data and target value. The irrigation volume is determined by the difference between stress index and threshold. The portable shading and supplemental lighting equipment and the intelligent irrigation system receive commands via wireless communication, which uses WiFi or ZigBee protocols. The zone management allows different areas to implement different control strategies. The switching of the shading curtain and LED supplemental lighting is controlled by a programmable logic controller.

[0032] Compared with existing technologies, this monitoring system for sugarcane flowering induction has the following advantages:

[0033] I. This invention collects environmental factor data, plant physiological state data, and key gene expression data by setting up a multi-dimensional sensing layer. Combined with the dynamic threshold regulation model of the intelligent decision-making center, the photoperiod and temperature thresholds are dynamically corrected based on phenological period and stress index. The induction effect is evaluated by combining the expression level of the ScFT6 gene. Then, the precise execution layer performs light regulation and irrigation operations, forming a closed-loop regulation system of perception-decision-execution. This enables the sugarcane flowering induction to shift from unidirectional environmental control to dynamic and precise regulation based on the perception of the plant's own physiological state. It effectively solves the problem that existing technologies lack real-time perception and response to the plant's physiological state and rely on fixed parameter thresholds, resulting in insufficient stability and accuracy of the induction effect under complex field environments and varietal differences.

[0034] Second, this invention improves the comprehensiveness and reliability of sugarcane field microenvironment data collection through the gridded deployment of the environmental factor sensing module and the data preprocessing mechanism; it utilizes multispectral UAVs and ground mobile robots to collaboratively collect plant physiological data, and combines portable field qPCR instruments to acquire gene expression information in real time, achieving multi-dimensional data fusion analysis; at the same time, the precise execution layer supports regional regulation, and the dynamic threshold regulation model has the ability to adaptively adjust the correction coefficient, which can improve the targeting and flexibility of regulation operations, reduce resource waste, and further ensure the stability of the sugarcane flowering induction effect in different sugarcane field areas and different varieties.

[0035] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0037] Figure 1 This is a diagram of the overall system architecture of the present invention;

[0038] Figure 2 This is a flowchart illustrating the data closed loop and decision-making process of the present invention.

[0039] Figure 3 This is a flowchart illustrating the operation of the present invention. Detailed Implementation

[0040] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0041] Example 1

[0042] like Figure 1 and Figure 2 As shown in this embodiment, a specific implementation of a monitoring system for sugarcane flowering induction is disclosed. The system aims to achieve integrated acquisition, dynamic decision-making, and precise regulation of environmental factors, plant physiological states, and key gene expression data during sugarcane flowering induction through the collaborative work of a multi-dimensional perception layer, an intelligent decision-making center, and a precise execution layer. This system uses a dynamic threshold regulation model to correct photoperiod and temperature thresholds in real time, and combines this with the expression level of the ScFT6 gene to assess the induction effect. This solves the problem in existing technologies where unidirectional environmental control cannot adapt to the real-time physiological response of plants, improving the accuracy and stability of sugarcane flowering induction and providing technical support for the matching of parental flowering periods in sugarcane sexual hybridization breeding.

[0043] The sugarcane flowering induction effect monitoring system in this embodiment includes a multi-dimensional sensing layer, an intelligent decision-making center, and a precise execution layer. Each layer interacts with data and transmits commands through an Internet of Things (IoT) module or wireless network, forming a closed-loop control system. The multi-dimensional sensing layer is deployed in the sugarcane field and plant monitoring area, the intelligent decision-making center is deployed in the sugarcane field management center, and the precise execution layer is deployed in a zoned manner within the sugarcane field. The overall architecture follows a closed-loop logic of "collection-analysis-decision-execution," ensuring real-time data and timely control.

[0044] The deployment and data acquisition of the environmental factor sensing module are implemented as follows:

[0045] The environmental factor sensing module includes photosynthetically active radiation sensors, temperature and humidity sensors, and data acquisition nodes, which are deployed in a grid-like manner in the experimental sugarcane field. The grid-like deployment is based on the topography and planting density of the sugarcane field, ensuring that each type of sensor can cover the preset sugarcane field area, thereby capturing microenvironmental changes in different areas.

[0046] The core function of the photosynthetically active radiation sensor is to measure the photosynthetically active radiation intensity in sugarcane fields and calculate the daily sunshine duration based on continuous measurement data. The temperature and humidity sensors measure air temperature and relative humidity and generate continuous temperature curves based on periodic measurement data. The data acquisition node connects to each sensor, and its built-in microprocessor is responsible for preprocessing the raw data collected by the sensors. Data preprocessing includes data filtering and outlier detection. Data filtering uses a moving average filtering method to remove high-frequency noise, and outlier detection uses the 3σ criterion to remove data exceeding a reasonable range. Simultaneously, the preprocessed data is temporarily cached. The power management unit of the data acquisition node supports solar power and battery backup to ensure continuous operation under different weather conditions.

[0047] Data acquisition nodes read sensor data at configurable intervals, ranging from 1 minute to 1 hour. In this embodiment, based on the fluctuating characteristics of the sugarcane field environment, the interval is configured to 10 minutes. The read data is stored in time-series format and associated with timestamps and sensor deployment location identifiers. Environmental factor data is transmitted to the intelligent decision-making center via an IoT module. The IoT module uses the LoRa communication protocol, which is suitable for long-distance, low-power data transmission and adapts to the large-scale deployment needs of sugarcane fields.

[0048] The deployment and data collection of the plant physiological state sensing module are implemented as follows:

[0049] The plant physiological state perception module includes a multispectral drone and a ground mobile robot. The two communicate with the intelligent decision-making center through a wireless network to collaboratively complete the collection of plant physiological state data.

[0050] The flight altitude and path of the multispectral UAV are programmably controlled via a ground control terminal. The flight altitude is set according to the height of the sugarcane plants in the field to ensure that the multispectral camera can clearly capture information about the plant canopy. The flight path covers the entire experimental sugarcane field, and flights are conducted regularly. In this embodiment, the flight cycle is set to 7 days to collect multispectral image data. The collected multispectral image data includes normalized difference vegetation index (NDVI) and photochemical reflectance index (PAR). The image data undergoes geometric and radiometric correction. Geometric correction is used to correct image distortion caused by flight attitude deviations, while radiometric correction is used to eliminate the influence of atmospheric scattering and differences in light intensity on image grayscale values, ensuring data accuracy.

[0051] The ground-based mobile robot, equipped with a chlorophyll fluorometer, moves along a preset path in the sugarcane field to measure chlorophyll fluorescence parameters of the plant leaves. The measured parameters include maximum fluorescence yield and actual photosynthetic efficiency, and the measurement data is transmitted to the intelligent decision-making center in real time.

[0052] Processing plant physiological state data includes phenological stage determination and stress index calculation:

[0053] Phenological stage determination: Based on the historical NDVI curves collected by multispectral UAVs, a machine learning model is used to automatically determine the phenological stage. The DBSCAN algorithm classifies the phenological stage by identifying the characteristic inflection points of the NDVI curve. For example, NDVI rises rapidly during the tillering stage, NDVI tends to stabilize during the elongation stage, and NDVI shows specific fluctuations during the induction critical stage. The phenological stage includes the tillering stage, the elongation stage, and the induction critical stage, and assigns values ​​for subsequent calculations: the tillering stage is assigned a value of 1, the elongation stage is assigned a value of 2, and the induction critical stage is assigned a value of 3.

[0054] Stress index calculation: The stress index includes water stress index and nutrient stress index. The water stress index is calculated based on the correlation between PRI value and soil moisture sensor data. The PRI value decreases as water stress intensifies. The degree of water stress is calculated by fitting the relationship between the two. The nutrient stress index is calculated based on the deviation between NDVI value and historical baseline. The historical baseline is the average NDVI value of this sugarcane variety under normal nutrient conditions. The larger the deviation, the higher the degree of nutrient stress.

[0055] The deployment and data acquisition of the key gene expression real-time monitoring module are implemented as follows:

[0056] The key gene expression real-time monitoring module uses a portable field qPCR instrument to collect sugarcane leaf tissue samples non-destructively and quantify the ScFT6 gene expression level in real time. The sampling cycle is set to 3 to 5 days, and 4 days is used in this embodiment.

[0057] The portable field qPCR instrument includes a sample processing unit and a detection unit. The sample processing unit integrates a rapid nucleic acid extraction kit and uses a dedicated sampler to collect sugarcane leaf sheath tissue, achieving non-destructive sampling and avoiding impact on plant growth. It can process multiple samples in batches, completing RNA extraction and reverse transcription within 10 minutes, converting RNA into cDNA to prepare for subsequent qPCR detection. The detection unit is pre-loaded with primers and probes targeting the ScFT6 and ScTPS1 genes, as well as an internal reference gene. The primers and probes are stored in lyophilized powder form and should be reconstituted according to the instructions before use. The internal reference gene is the UBQ gene, used to correct errors during sample processing. The detection unit has temperature control and fluorescence detection functions. Temperature control ensures accurate temperature at each stage of the qPCR reaction, and fluorescence detection monitors the fluorescence signal intensity in real time during the qPCR reaction.

[0058] After the detection unit runs the qPCR program, it outputs the relative expression level of the ScFT6 gene, which is calculated using the 2^-ΔΔCt method. The calculation first obtains the Ct values ​​of the target gene ScFT6 and the internal reference gene UBQ, calculates the difference between them to obtain ΔCt, then calculates the difference between the ΔCt values ​​of the sample to be tested and the calibration sample to obtain ΔΔCt, and finally calculates the relative expression level using 2^(-ΔΔCt). The calculated relative expression level data is associated with the sampling time point and sampling location and transmitted to the intelligent decision-making center via Bluetooth.

[0059] The intelligent decision-making center is built on an industrial-grade server and has a built-in dynamic threshold control model. Its core function is to receive various types of data transmitted from the multi-dimensional perception layer, perform fusion analysis through algorithm models, and generate control instructions for the precise execution layer.

[0060] The threshold correction in the dynamic threshold control model is implemented as follows:

[0061] The dynamic threshold control model is used to dynamically correct the photoperiod threshold and the temperature threshold. The temperature threshold correction formula has the same structure as the photoperiod threshold correction formula, only the parameter values ​​are different. This embodiment takes the photoperiod threshold correction as an example for detailed explanation.

[0062] The optical period threshold correction is performed using the following formula:

[0063]

[0064] In the formula, The corrected photoperiod threshold, in hours, is used to precisely execute the generation of illumination control commands for the layer. The baseline photoperiod threshold is set based on historical successful induction data and relevant literature for this sugarcane variety. For example, the baseline photoperiod threshold for a certain tropical sugarcane variety is set to 12 hours. : Phenological period correction factor, ranging from -0.1 to 0.1, is used to quantify the impact of the difference between the current phenological period and the reference phenological period on the photoperiod threshold. The initial value is set to 0.05. The current phenological stage value is assigned by the phenological stage judgment result of the plant physiological state sensing module. The reference phenological period value is set to 2, corresponding to the elongation period. The elongation period is a key preparatory stage for inducing sugarcane flowering, and therefore serves as a reference benchmark. Stress correction coefficient, ranging from -0.05 to 0.05, is used to quantify the impact of the difference between the current stress index and the reference stress index on the photoperiod threshold. The initial value is set to 0.02. The current stress index value is calculated using the normalized water stress index, ranging from 0 to 1, where 0 indicates no stress and 1 indicates severe stress. The reference stress index value was set to 0.5, corresponding to a moderate stress level, as a benchmark for balancing the regulatory effect and plant tolerance.

[0065] The temperature threshold correction formula has the same structure as the formula above, only the parameter values ​​are different, such as the reference temperature threshold. Set to 25℃, phenological period correction factor The value ranges from -0.5 to 0.5, and the stress correction coefficient is... The value ranges from -0.2 to 0.2, referencing the stress index value. It remains at 0.5, with the specific value determined based on the temperature sensitivity of the sugarcane variety.

[0066] The adaptive adjustment of the correction coefficient is implemented as follows:

[0067] Phenological period correction coefficient With stress correction factor The following formula is used for adaptive adjustment to ensure that the threshold correction can adapt to the plant's physiological response and induction effect in real time:

[0068]

[0069]

[0070] In the formula, The initial value of the phenological period correction coefficient is set to 0.05. The phenological learning rate is set to 0.1 to control... Adjustment rate, The larger, The more sensitive the gene expression differences, the better. The current ScFT6 gene expression level is transmitted by the key gene expression real-time monitoring module. The target ScFT6 gene expression level was set based on gene expression data at the time of successful induction in this sugarcane variety. The initial value of the stress correction coefficient is set to 0.02. The coercive learning rate, set to 0.05, is used to control... Adjustment rate, The current stress index value, and the value in the photoperiod threshold correction formula. Consistent, The optimal stress index value is set to 0.2, which corresponds to a mild stress level. At this level, the plant will not be damaged by stress and can maintain its sensitivity to induced signals.

[0071] The adaptive adjustment of the correction coefficient is based on historical successful induction data and optimized using the gradient descent method: the system periodically calls historical data, which is set to 15 days in this embodiment, to calculate the current... , The corresponding induction effect deviation is determined by whether the ScFT6 gene expression level reaches [a certain level]. As the evaluation criterion, the gradient descent method is used to adjust... and , thereby optimizing and The value of is determined to ensure the adaptability of the dynamic threshold control model.

[0072] The implementation of induction effect evaluation and control instruction generation is as follows:

[0073] The intelligent decision-making center evaluates the induction effect by comparing environmental factor data with a corrected dynamic threshold and combining it with ScFT6 gene expression levels. The evaluation period is synchronized with the sampling period for key gene expression monitoring; in this example, it is 4 days. The evaluation includes three scenario judgments:

[0074] Scenario A: If the environmental data meets the requirements and the ScFT6 gene expression is upregulated, the induction is considered successful, and the system generates instructions to maintain the current regulatory parameters.

[0075] Scenario B: Environmental data meets the standards, but ScFT6 gene expression is not upregulated, indicating a risk of induction failure. The system triggers an early warning signal, which is displayed through the user interface of the management center, prompting the administrator to check the plant status or adjust the model parameters. At the same time, it generates instructions to adjust the photoperiod / temperature threshold, such as appropriately shortening the photoperiod threshold to enhance the induction signal.

[0076] Scenario C: Environmental data does not meet the standard, but ScFT6 gene expression is upregulated, indicating that the plant has responded. The system triggers the model learning mechanism, records the current environmental data and gene expression data, and uses them to optimize the parameters of the dynamic threshold regulation model. At the same time, it generates instructions to fine-tune the environmental parameters, so that the environmental data moves closer to the corrected threshold and avoids the impact of environmental fluctuations on the subsequent induction effect.

[0077] The above comparison operation uses a numerical comparison algorithm, including threshold comparison and trend analysis. The trend analysis uses the sliding window method to calculate the rate of change of ScFT6 gene expression. The sliding window size is set to the three most recent sampled data. The rate of change is calculated by linear fitting to obtain the slope value. A positive slope indicates an increase in expression.

[0078] The intelligent decision-making center generates control instructions based on the evaluation results of the induction effect. The instructions include the shading time of the portable shading and supplemental lighting equipment, the supplemental lighting intensity, and the irrigation duration of the intelligent irrigation system. The instructions are transmitted to the precision execution layer via wireless communication using WiFi or ZigBee protocols.

[0079] The precision execution layer includes mobile shading and supplemental lighting equipment and an intelligent irrigation system, which are deployed in accordance with the sugarcane field zones and execute operations based on the control commands of the intelligent decision-making center.

[0080] The operation of the portable shading and supplemental lighting equipment is as follows:

[0081] The portable shading and supplemental lighting equipment moves through sugarcane fields using a track system that covers multiple zones, supporting zone management. Different zones can implement different control strategies to adapt to the microenvironmental differences in the sugarcane field. The equipment includes a shading curtain and LED supplemental lighting, controlled by a programmable logic controller (PLC).

[0082] After receiving the light cycle threshold command from the intelligent decision-making center, the device calculates the shading time or supplemental light intensity:

[0083] Shading time calculation: based on daily light duration collected by the environmental factor sensing module and The difference, if the daily light exposure duration exceeds The shading time is the difference between the two, for example, if the daily light exposure is 13 hours. If the duration is 12 hours, then the shading time is 1 hour.

[0084] Complementary light intensity calculation: Based on the difference between the real-time radiation intensity collected by the photosynthetically active radiation sensor and the target value, if the real-time radiation intensity is lower than the target value, the supplementary light intensity is the difference between the two. For example, if the target value is 1000 μmol / (m²·s) and the real-time value is 800 μmol / (m²·s), then the supplementary light intensity is 200 μmol / (m²·s).

[0085] The PLC controls the unfolding / retraction of the shading curtain or the activation / adjustment of the LED supplemental lighting based on the calculation results to ensure that the sugarcane field's lighting conditions meet the requirements. Require.

[0086] The operation and implementation of the intelligent irrigation system are as follows:

[0087] The intelligent irrigation system includes sprinkler heads and solenoid valves. The sprinkler heads are deployed between the rows of sugarcane plants, and the solenoid valves are connected to the sprinkler heads to control the flow of irrigation water.

[0088] After receiving the stress index command from the intelligent decision-making center, the system calculates the irrigation amount based on the difference between the current stress index and a threshold. The threshold is set to 0.3; below the threshold indicates no significant stress, while above the threshold indicates irrigation is required. The larger the difference, the larger the irrigation amount. For example, with a stress index of 0.5, a threshold of 0.3, and a difference of 0.2, the corresponding irrigation amount is 1.2 times the preset baseline irrigation amount. The solenoid valve controls the working time of the sprinkler head according to the calculated irrigation amount and duration command, achieving precise irrigation and alleviating plant stress.

[0089] In summary, this embodiment achieves comprehensive data collection of environmental factors, plant physiological state, and key gene expression data through a multi-dimensional sensing layer. It utilizes a dynamic threshold regulation model within an intelligent decision-making center to complete data fusion analysis and dynamic threshold correction. Combined with ScFT6 gene expression levels, it achieves precise evaluation of the induction effect. Finally, a precise execution layer executes light regulation and irrigation operations, forming a closed-loop system of "sensing-decision-execution." This embodiment overcomes the limitations of unidirectional environmental control in existing technologies, achieving dynamic regulation based on real-time plant physiological responses. It improves the accuracy and stability of sugarcane flowering induction, effectively ensuring the matching of parent flowering periods in sugarcane sexual hybridization breeding, and providing reliable technical support for sugarcane variety improvement.

[0090] Example 2

[0091] like Figure 3 As shown in Example 1, this example details the specific steps of a monitoring system for sugarcane flowering induction during operation. The specific steps are as follows:

[0092] 1. System startup and initialization:

[0093] The system performs a power-on self-test to confirm that all modules are communicating normally.

[0094] The environmental factor sensing module deployed in the sugarcane field activates the sensors and initializes data collection according to the gridded locations.

[0095] The multispectral drone and ground mobile robot in the plant physiological state perception module start the self-test program and load the preset flight path and movement trajectory.

[0096] The portable field qPCR instrument has been calibrated and the pre-prepared primers and probes have been reconstituted and are ready for use.

[0097] The intelligent decision-making central server loads the dynamic threshold control model and initial parameters, and establishes data connections with each module.

[0098] 2. Environmental factor data collection and transmission:

[0099] The environmental factor sensing module reads photosynthetically active radiation intensity, air temperature, and relative humidity at configurable intervals.

[0100] The data acquisition node filters and detects outliers in the raw data, and then adds timestamps and location identifiers after processing.

[0101] The processed environmental factor data is sent to the intelligent decision-making center via LoRa or 4G communication protocols.

[0102] 3. Data collection and processing of plant physiological status:

[0103] Multispectral drones fly according to a preset cycle to collect multispectral image data of sugarcane fields.

[0104] Image data is transmitted to the intelligent decision-making center after geometric and radiometric correction.

[0105] The ground mobile robot moves along a preset path, equipped with a chlorophyll fluorometer to measure chlorophyll fluorescence parameters, and the data is uploaded in real time.

[0106] The intelligent decision-making center automatically determines the phenological period based on the historical NDVI curve using a time series clustering algorithm, and calculates the stress index by combining PRI and soil moisture data.

[0107] 4. Key gene expression data collection and analysis:

[0108] Portable field qPCR instruments were used to collect sugarcane leaf sheath tissue samples non-destructively according to the sampling cycle.

[0109] The sample processing unit completes RNA extraction and reverse transcription within 10 minutes, and the detection unit runs a qPCR program to quantify the expression level of the ScFT6 gene.

[0110] Relative expression level The method calculates the results, which are then transmitted to the intelligent decision-making center via Bluetooth, and are associated with the sampling time and location.

[0111] 5. Data fusion and dynamic threshold correction:

[0112] The intelligent decision-making center receives and integrates data on environmental factors, plant physiological status, and gene expression.

[0113] The dynamic threshold regulation model dynamically corrects the photoperiod threshold and temperature threshold based on the current phenological period and stress index. The phenological period correction coefficient and stress correction coefficient are adaptively adjusted according to the expression level of the ScFT6 gene and the stress state.

[0114] The revised threshold is used to generate the control command baseline.

[0115] 6. Evaluation of the induction effect and decision generation:

[0116] The intelligent decision-making center compares the current environmental data with the corrected threshold and evaluates the induction effect by combining the changing trend of ScFT6 gene expression:

[0117] If the environmental data meet the requirements and the expression of the ScFT6 gene is upregulated, the induction is considered successful, and the current regulation is maintained.

[0118] If environmental data meets the standards but ScFT6 gene expression is not upregulated, it is judged as a risk of induction failure, triggering an early warning signal and generating instructions to adjust the photoperiod or temperature threshold.

[0119] If environmental data does not meet the standards but ScFT6 gene expression is upregulated, it is determined that the plant has responded. The data is recorded for model learning, and instructions for fine-tuning environmental parameters are generated.

[0120] The assessment results are translated into specific control instructions, which are then sent to the precise execution layer via WiFi or ZigBee protocols.

[0121] 7. Precisely execute control measures:

[0122] After receiving instructions, the mobile shading and supplemental lighting equipment moves to the designated sugarcane field section using a track system:

[0123] If shading is required, calculate the difference between the daily illumination duration and the light cycle threshold, and control the shading curtain to open for the corresponding time.

[0124] If supplemental lighting is needed, adjust the intensity of the LED supplemental lights according to the difference between the real-time photosynthetically effective radiation intensity and the target value.

[0125] The intelligent irrigation system calculates the irrigation volume based on the difference between the stress index and the threshold, and controls the solenoid valves and sprinkler heads to perform precise irrigation.

[0126] 8. System monitoring and adaptive optimization:

[0127] The system continues to run the above steps, forming a closed loop of "collection-analysis-decision-execution".

[0128] The intelligent decision-making center regularly optimizes model parameters based on historical successful induction data, updates the phenological period learning rate and stress learning rate, and improves the accuracy of threshold correction.

[0129] The user interface displays the system status, early warning information, and control records in real time for administrators to view and intervene.

[0130] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A monitoring system for the induction effect of sugarcane flowering period, characterized in that, It includes a multi-dimensional perception layer, an intelligent decision-making center, and a precise execution layer; The multidimensional sensing layer is used to collect environmental factor data, plant physiological state data, and key gene expression data. The environmental factor data includes light duration and temperature; the plant physiological state data includes phenological stage and stress index; and the key gene expression data includes the relative expression level of the ScFT6 gene. The multidimensional sensing layer includes an environmental factor sensing module, a plant physiological state sensing module, and a key gene expression real-time monitoring module. The environmental factor sensing module includes a photosynthetically active radiation sensor and a temperature and humidity sensor; the plant physiological state sensing module includes a multispectral UAV and a ground mobile robot; and the key gene expression real-time monitoring module includes a portable field qPCR instrument. The intelligent decision-making center is used to receive data from the multidimensional sensing layer and generate control instructions through a dynamic threshold control model. The dynamic threshold control model dynamically corrects the photoperiod threshold and temperature threshold based on phenological period and stress index, and evaluates the induction effect by combining the relative expression level of the ScFT6 gene. The precision execution layer is used to receive control commands and execute light control and irrigation operations. The precision execution layer includes a portable shading and supplemental lighting device and an intelligent irrigation system. The environmental factor sensing module, plant physiological state sensing module, and key gene expression real-time monitoring module communicate with the intelligent decision-making center through the Internet of Things module. The intelligent decision-making center integrates and analyzes environmental factor data, plant physiological state data, and key gene expression data through an algorithm model. The portable shading and supplemental lighting device performs zoned shading or supplemental lighting based on the photoperiod threshold. The intelligent irrigation system irrigates based on a stress index.

2. The monitoring system for sugarcane flowering induction effect according to claim 1, characterized in that, The photosynthetically active radiation (PAR) sensor and temperature and humidity sensor in the environmental factor sensing module are deployed in a grid-like manner in the sugarcane field. The PAR sensor measures the PAR intensity and calculates the daily sunshine duration, while the temperature and humidity sensor measures the air temperature and relative humidity and generates a temperature curve. The IoT module uses LoRa or 4G communication protocols to transmit environmental factor data. The environmental factor sensing module also includes a data acquisition node, which periodically reads sensor data and sends it to the intelligent decision-making center through the IoT module. The data acquisition node includes a microprocessor and a power management unit. The microprocessor is responsible for data preprocessing and caching, while the power management unit supports solar power and battery backup. The grid-like deployment allows the sensors to cover multiple areas of the sugarcane field to capture microenvironmental changes. The data preprocessing includes data filtering and outlier detection. The periodic reading period is configurable, ranging from 1 minute to 1 hour. The environmental factor data is stored in a time-series format with timestamps and location identifiers.

3. The monitoring system for sugarcane flowering induction effect according to claim 1, characterized in that, The multispectral UAV in the plant physiological state sensing module periodically flies to collect multispectral image data, including the Normalized Difference Vegetation Index (NDVI) and the Photochemical Reflectance Index (PRI). A ground-based mobile robot equipped with a chlorophyll fluorometer measures chlorophyll fluorescence parameters. Phenological stages are automatically determined by analyzing historical NDVI curves and using a machine learning model employing a time-series clustering algorithm. Stress indices are calculated using PRI data and soil moisture sensor data. The multispectral UAV's flight altitude and path are programmable. The multispectral image data undergoes geometric and radiometric correction. Chlorophyll fluorescence parameters include maximum fluorescence yield and actual photosynthetic efficiency. The time-series clustering algorithm uses the DBSCAN algorithm. Phenological stages include tillering, elongation, and induction critical periods. Stress indices include water stress index and nutrient stress index. The water stress index is calculated based on the correlation between PRI values ​​and soil moisture values, while the nutrient stress index is calculated based on the deviation of NDVI values ​​from historical benchmarks. The multispectral UAV and the ground-based mobile robot communicate with the intelligent decision-making center via a wireless network.

4. The monitoring system for sugarcane flowering induction effect according to claim 1, characterized in that, The portable field qPCR instrument in the key gene expression real-time monitoring module is used for non-destructive collection of sugarcane leaf tissue samples and real-time quantification of ScFT6 gene expression levels. The portable field qPCR instrument includes a sample processing unit and a detection unit. The sample processing unit integrates a rapid nucleic acid extraction kit, and the detection unit is pre-loaded with primers and probes for the ScFT6 gene and an internal reference gene. The primers and probes are stored in lyophilized powder form. Non-destructive collection uses a dedicated sampler to collect leaf sheath tissue. The sample processing unit completes RNA extraction and reverse transcription within 10 minutes. The detection unit runs the qPCR program and outputs the relative expression level of the ScFT6 gene, which is calculated using the 2^-ΔΔCt method. The portable field qPCR instrument transmits data to the intelligent decision-making center via Bluetooth. The sampling cycle is 3 to 5 days. The sample processing unit supports batch processing of multiple samples. The detection unit has temperature control and fluorescence detection functions. The primers and probes are designed for the ScFT6 and ScTPS1 genes. The internal reference gene includes the UBQ gene. The relative expression level data is stored in association with the sampling time point and location.

5. The monitoring system for sugarcane flowering induction effect according to claim 1, characterized in that, The dynamic threshold control model in the intelligent decision-making center uses the following mathematical formula to dynamically correct the photoperiod threshold: in, This represents the corrected photoperiod threshold. Indicates the reference optical period threshold. This represents the phenological period correction factor. This indicates the current phenological period value. Indicates the reference phenological period value. This represents the stress correction factor. This indicates the current stress index value. This indicates the reference stress index value; The phenological period values The tillering stage is assigned a value of 1, the elongation stage a value of 2, and the induction critical stage a value of 3, based on the phenological stage categories. The reference phenological stage values ​​are... The stress index value is set to 2. The reference stress index value is calculated using a normalized water stress index, ranging from 0 to 1. The phenological period correction coefficient is set to 0.

5. The value range is from -0.1 to 0.1, and the stress correction coefficient is... The value range is from -0.05 to 0.05; The dynamic threshold control model also uses a temperature threshold correction formula, which has the same structure as the photoperiod threshold correction formula but different parameters. The reference photoperiod threshold... Based on historical data and literature settings, the phenological period values ​​and stress index values ​​are derived from the plant physiological state sensing module, and the corrected photoperiod threshold is used to generate control instructions for the precise execution layer.

6. The monitoring system for sugarcane flowering induction effect according to claim 5, characterized in that, The phenological period correction coefficient in the dynamic threshold control model and stress correction coefficient Adaptive adjustment is achieved through the following mathematical formula: in This represents the initial value of the phenological period correction coefficient. Indicates the learning rate during the phenological period. This indicates the current expression level of the ScFT6 gene. Indicates the expression level of the target ScFT6 gene. This represents the initial value of the stress correction coefficient. Indicates the forced learning rate. This indicates the current stress index value. This represents the optimal stress index value; The initial value of the phenological period correction coefficient The phenological learning rate is set to 0.

05. The current ScFT6 gene expression level is set to 0.

1. The expression level of the target ScFT6 gene, obtained from the real-time monitoring module for key gene expression. The initial value of the stress correction coefficient is set based on variety-specific data. The stress learning rate is set to 0.

02. The optimal stress index value is set to 0.

05. Set to 0.2; The adaptive adjustment is optimized using gradient descent based on historical successful induction data, and the phenological period learning rate is... and coerced learning rate The mathematical formula is dynamically updated during system operation and is used to correct for nonlinear changes in coefficients with gene expression and stress state.

7. The monitoring system for sugarcane flowering induction effect according to claim 1, characterized in that, The induction effect evaluation of the intelligent decision-making center is carried out by comparing environmental factor data with dynamic thresholds and combining ScFT6 gene expression levels. The induction effect evaluation includes three scenario judgments. Scenario A indicates that the environmental data meets the criteria and ScFT6 gene expression is upregulated, which is considered a successful induction. Scenario B indicates that the environmental data meets the criteria but ScFT6 gene expression is not upregulated, which is considered a risk of induction failure. Scenario C indicates that the environmental data does not meet the criteria but ScFT6 gene expression is upregulated, which is considered a response from the plant. The environmental data meeting the criteria refers to the light duration and temperature being within the corrected threshold range. The upregulation of ScFT6 gene expression refers to a relative increase in expression level exceeding 20%. The risk of induction failure triggers an early warning signal. The plant's response triggers the model learning mechanism. The early warning signal is displayed through the user interface. The model learning mechanism records environmental data and gene expression data to update the dynamic threshold regulation model. The comparison operation uses a numerical comparison algorithm, which includes threshold comparison and trend analysis. The trend analysis uses a sliding window to calculate the rate of change of ScFT6 gene expression level. The rate of change is calculated using a slope value. The sliding window size is the data from the three most recent samples. The induction effect evaluation period is synchronized with the sampling period.

8. The monitoring system for sugarcane flowering induction effect according to claim 1, characterized in that, The movable shading and supplemental lighting equipment in the precision execution layer moves through the sugarcane field based on a track system. This equipment includes a shading curtain and LED supplemental lights. The shading curtain reduces the duration of sunlight exposure, while the LED supplemental lights increase light intensity. The intelligent irrigation system includes sprinkler heads and solenoid valves. The sprinkler heads control the irrigation volume based on a stress index. The track system covers multiple zones within the sugarcane field. The control commands include shading time, supplemental lighting intensity, and irrigation duration. The movable shading and supplemental lighting equipment receives a photoperiod threshold command and calculates the shading or supplemental lighting time. The intelligent irrigation system receives stress index commands and calculates irrigation volume. The shading time is determined by the difference between the daily light duration and the photoperiod threshold. The supplemental lighting intensity is determined by the difference between the photosynthetically active radiation sensor data and the target value. The irrigation volume is determined by the difference between the stress index and the threshold. The portable shading and supplemental lighting equipment and the intelligent irrigation system receive commands via wireless communication, which uses WiFi or ZigBee protocols. The zone management allows different areas to implement different control strategies. The switching of the shading curtain and LED supplemental lighting is controlled by a programmable logic controller.

Images