An online non-destructive monitoring method for nutritional stress of strawberry virus-free seedlings

Abstract

The present application relates to the field of plant tissue physiological state time-resolved photoelectric detection technology, and discloses an online nondestructive monitoring method for strawberry virus-free seedling nutrient stress, comprising: collecting diffuse reflection signals and calculating distortion variance; generating a driving signal according to the variance adjustment pulse width; injecting a reverse charge pulse at the signal cutoff moment to neutralize the residual carriers of the junction capacitor, disconnecting the bias path and turning on the acquisition loop; collecting the fluorescence decay level at discrete time to calculate the first-order difference slope and the second-order difference rate of the relative decay rate characteristic scalar, and determining the nutrient stress state accordingly; the present application eliminates the overshoot noise of the junction capacitor by injecting a reverse neutralization charge pulse, constructs a fast fluorescence front acquisition channel, effectively overcomes the greenhouse inspection vibration and greenhouse diffuse scattering interference, and improves the identification accuracy of the hidden nutrient stress in dynamic inspection.

Description

Technical Field

[0001] This invention belongs to the field of time-resolved photoelectric detection technology of plant tissue physiological state, and particularly relates to an online non-destructive monitoring method for nutrient stress in virus-free strawberry seedlings. Background Technology

[0002] In the current technical system that utilizes pulsed radiation to demodulate the excited energy level transition characteristics of living plant tissues, high-frequency discontinuous excitation pulses are projected onto the surface of the target sample, and transient radiation intensity that decays over time is discretely collected by time-resolved detection components. Then, the relative decay slope of the photochemical conversion rate inside the cell is calculated based on the waveform envelope. This process is a normalized technical path for identifying the light energy capture state of the chlorophyll system in living plants. With the application and development of the closed-loop seedling automated track inspection mode in high-speed online testing scenarios, the mechanical micro-vibrations of the mobile inspection platform and the random multidimensional deflection of the leaf's spatial unfolding posture cause nondeterministic spatial diffuse scattering of the incident light path. Conventional methods increase the radiation power of the excitation pulse to maintain the detection signal-to-noise ratio. However, this power adjustment causes a cumulative deep energy level charge trapping effect in the depletion layer inside the sensing diode, and induces a long tail overshoot waveform of junction capacitance charge during high-frequency turn-off transients, which extensively covers and contaminates the acquisition window of the weak fluorescence decay kinetics leading edge signal.

[0003] To address the early time-domain detection blind zone caused by photogenerated parasitic charge tailing, conventional strategies attempt to simply delay the sampling timing window or incorporate a matching circuit to attenuate the charge. However, delaying the sampling timing inevitably results in the loss of early attenuation information rich in weak physiological mutation characteristics. Incorporating filtering devices, on the other hand, is limited by fixed frequency band constraints, leading to a narrowing of the sampling transient measurement bandwidth and making it difficult to adaptively balance carrier dissipation changes under diffuse scattering mutation conditions. For example, Chinese invention patent application CN105548122A discloses a chlorophyll fluorescence time-sharing monitoring system that uses an overall time period... The time-division control with a fixed preset integral time relies on the ideal state of stable and slow change in ambient light intensity. However, when faced with instantaneous high-frequency mechanical shaking in the greenhouse inspection line and sudden fog diffusion conditions in the greenhouse, the static time period division cannot track sudden distortion of the ambient light path in real time. It fails to delve into the carrier dynamics level inside the depletion layer of the sensing component and cannot offset and clear the residual charge of the junction capacitance caused by the high-frequency excitation pulse. The core preset premise is mismatched with the boundary conditions of the actual working conditions. During high-speed automated inspection, it is difficult to avoid the contamination of the leading window of the weak fluorescence waveform by geometric attitude deflection and long charge tail.

[0004] Therefore, how to control the carrier motion state inside the sensing depletion layer to lock the trap energy level and remove transient residual charge, and stably demodulate the monotonic corresponding relative decay rate index under complex diffuse scattering interference environment, constitutes a technical problem to be solved in this field. Summary of the Invention

[0005] This invention aims to solve the problem of controlling the carrier motion state inside the sensing depletion layer to lock the trap energy level and remove transient residual charge, and stably demodulating the monotonic corresponding relative decay rate index under complex diffuse scattering interference environment.

[0006] In this technical solution, an online non-destructive monitoring method for nutrient stress in virus-free strawberry seedlings includes the following steps: In step S101, under the condition of no external excitation light, the time-resolved photoelectric detection module collects the diffuse reflection light signal of the greenhouse environment. The central processing chip calculates the statistical discrete variance based on multiple sets of diffuse reflection light signals of the greenhouse environment and confirms the statistical discrete variance as the environmental diffuse scattering distortion variance. In step S102, the central processing chip increases the driving pulse width of the pulse period according to the variance of the environmental diffuse scattering distortion to generate a high-frequency excitation driving signal, which drives the excitation light source to radiate a multi-frequency excitation beam to the leaves of the strawberry virus-free seedling. Step S103: At the instant the pulse of the high-frequency excitation driving signal is cut off, the central processing chip controls the reverse pulse source to inject a reverse neutralizing charge pulse into the junction capacitance of the time-resolved photodetector module to neutralize the residual photogenerated charge in the junction capacitance. At the same time, the bias charge injection path is disconnected and the signal acquisition and holding circuit is turned on. In step S104, the time-resolved photoelectric detection module acquires chlorophyll fluorescence decay level waveforms at the first and second discrete moments after the high-frequency excitation drive signal is turned off. The central processing chip uses the logarithmic difference operator to calculate the relative decay rate characteristic scalar, calculates the first-order difference slope and the second-order difference rate of change of the relative decay rate characteristic scalar, and determines the nutrient stress state of the virus-free strawberry seedlings based on the first-order difference slope and the second-order difference rate of change.

[0007] Preferably, step S101 further includes: adjusting the closed state of the internal optical path in the absence of external excitation light, continuously collecting 30 to 50 sets of diffuse reflection light signals of the greenhouse environment through the time-resolved photoelectric detection module; the central processing chip performs mean filtering and normalization calculation on the 30 to 50 sets of diffuse reflection light signals of the greenhouse environment and solves the statistical discrete variance, and confirms the statistical discrete variance as the environmental diffuse scattering distortion variance.

[0008] Preferably, the calculation of the first-order difference slope and second-order difference rate of change of the relative decay rate feature scalar in step S104, and the determination of the nutrient stress state of the strawberry virus-free seedlings based on the first-order difference slope and second-order difference rate of change, is further refined into the following sub-steps: Step S1041, the central processing chip stores the relative decay rate feature scalar of five consecutive cycles into the sliding window shift register in the order of time generation to construct a time-domain scalar time series; Step S1042, the central processing chip performs first-order difference calculation on the time-domain scalar time series to calculate the first-order difference slope, and obtains the second-order difference rate of change through second-order difference calculation based on the first-order difference slope; when the second-order difference rate of change exceeds the preset safety threshold range, the central processing chip outputs the nutrient stress monitoring results of the strawberry virus-free seedlings.

[0009] Preferably, step S102, which increases the driving pulse width of subsequent pulse cycles based on the environmental diffuse scattering distortion variance, is refined into the following sub-steps: Step S1021, the central processing chip compares the calculated environmental diffuse scattering distortion variance with the internally stored reference variance interval; Step S1022, when the environmental diffuse scattering distortion variance is greater than the upper limit of the reference variance interval, the central processing chip increases the pulse width of the high-frequency excitation driving signal based on the deviation of the environmental diffuse scattering distortion variance from the upper limit, thereby increasing the radiation energy of the excitation source in a single pulse cycle.

[0010] Preferably, step S103, which controls the reverse pulse source to inject a reverse neutralizing charge pulse into the junction capacitance of the time-resolved photodetector module, is refined into the following sub-steps: Step S1031, when the falling edge of the pulse cutoff of the high-frequency excitation drive signal is captured, the reverse pulse source is triggered to turn on; Step S1032, the reverse pulse source injects a reverse bias current into the time-resolved photodetector module, and the pulse amplitude and duration of the reverse neutralizing charge pulse are reverse-modulated by the current duty cycle parameter of the high-frequency excitation drive signal.

[0011] Preferably, step S103, which involves simultaneously disconnecting the bias charge injection path and turning on the signal acquisition and holding circuit, is further refined into the following sub-steps: Step S1033, disconnecting the first charge channel at the instant the high-frequency excitation drive signal is cut off to cut off the bias charge injection path; Step S1034, simultaneously turning on the second charge channel to close the signal acquisition and holding circuit, and resolving the electrical signal transmission path between the output terminal of the photoelectric detection module and the capacitor input terminal of the sample-and-hold circuit by the conduction time of the second charge channel.

[0012] Preferably, in step S104, the time-resolved photodetector module collects the chlorophyll fluorescence decay level waveform at the first discrete moment and the second discrete moment after the high-frequency excitation drive signal is turned off. Specifically, after the residual photogenerated charge is cleared by the reverse neutralization charge pulse, the time-resolved photodetector module collects the chlorophyll fluorescence decay level waveform at the first discrete moment and the second discrete moment within a time window of 10μs to 0.5ms after the high-frequency excitation drive signal is turned off.

[0013] Preferably, the modulation frequency of the multi-frequency excitation beam is 10kHz to 500kHz; the sampling rate of the time-resolved photoelectric detection module is 10MSPS to 100MSPS.

[0014] Preferably, the tracking and monitoring steps include the following: Step S105, the central processing chip continuously records the first-order difference slope and second-order difference rate of change of the same strawberry virus-free seedling at the same growth node during a continuous inspection cycle of 7 to 30 days, and constructs a historical dataset; Step S106, the central processing chip calculates the sliding variance of the historical dataset, and when the sliding variance continues to increase within 3 consecutive inspection cycles and exceeds the preset nutrient stress alarm threshold, it outputs a nutrient stress alarm signal.

[0015] Compared with existing technologies, the online non-destructive monitoring method for nutrient stress in virus-free strawberry seedlings of the present invention has the following advantages: 1. In the online non-destructive monitoring of nutrient stress in virus-free strawberry seedlings, a fixed amplitude subthreshold pre-saturation bias current is injected into the photodetector diode in reverse within the time window before the application of a high-frequency excitation pulse. This forces the filling of deep energy level charge traps inside the depletion layer and solidifies the zero-point dark current noise substrate. In conjunction with the reconstructing of the bias charge reverse injection path based on the variance change of the ground state steady-state diffuse reflection electrical signal within the yield delay window triggered by the falling edge of the pulse signal, a heterogeneous reverse pump charge pulse with a positive correlation between amplitude and variance is introduced into the input terminal of the photodetector. This actively absorbs and neutralizes the residual photogenerated parasitic charge inside the junction capacitance in situ, blocking the nonlinear distortion caused by high-frequency mechanical vibration of the inspection track.

[0016] 2. The time-resolved photoelectric detection module continuously reads multiple sets of ground-state steady-state diffuse reflection electrical signals without external excitation. The central processing chip calculates the working condition fluctuation variance characterizing the degree of environmental diffuse scattering distortion. Based on this variance, the high-frequency drive pulse width parameter of the subsequent stage is adaptively broadened across levels. In conjunction with disconnecting the bias charge injection path and turning on the signal acquisition and holding circuit at the moment the pulse signal is cut off, the time-resolved photoelectric detection module accurately captures the nonlinear fluorescence decay waveform level after purification at a limited discrete time point, directly eliminating the nonlinear distortion of light intensity electrical signals caused by sudden fog diffuse scattering in the greenhouse.

[0017] 3. The central processing chip receives the characteristic pulse radiation light intensity level read at discrete moments and calculates the relative attenuation rate characteristic scalar through the internal logarithmic difference operator. By using the relative attenuation transformation slope in the time dimension, the physical dependence of the detection system on the absolute light intensity value is stripped away. Under the condition of avoiding the mechanical adjustment and alignment components of the external optical network geometry, the random spatial tilt caused by the natural spread angle of the leaf and the influence of measurement distance fluctuation are directly offset by pure digital timing algorithm logic. The uncontrollable attenuation of the external optical path is transformed into a dimensionless relative ratio that purely maps the specific index of the chlorophyll excited energy level recombination rate, thereby enhancing the repeatability of optical testing. Attached Figure Description

[0018] Figure 1 This is a flowchart of the time-resolved photoelectric detection method for monitoring nutritional stress in virus-free strawberry seedlings according to the present invention. Figure 2 This is a status diagram of the monitoring of nutritional stress in virus-free strawberry seedlings according to the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0020] A method for online non-destructive monitoring of nutrient stress in virus-free strawberry seedlings includes the following steps: In step S101, under the condition of no external excitation light, the time-resolved photoelectric detection module collects the diffuse reflection light signal of the greenhouse environment. The central processing chip calculates the statistical discrete variance based on multiple sets of diffuse reflection light signals of the greenhouse environment and confirms the statistical discrete variance as the environmental diffuse scattering distortion variance. In step S102, the central processing chip increases the driving pulse width of the pulse period according to the variance of the environmental diffuse scattering distortion to generate a high-frequency excitation driving signal, which drives the excitation light source to radiate a multi-frequency excitation beam to the leaves of the strawberry virus-free seedling. Step S103: At the instant the pulse of the high-frequency excitation driving signal is cut off, the central processing chip controls the reverse pulse source to inject a reverse neutralizing charge pulse into the junction capacitance of the time-resolved photodetector module to neutralize the residual photogenerated charge in the junction capacitance. At the same time, the bias charge injection path is disconnected and the signal acquisition and holding circuit is turned on. In step S104, the time-resolved photoelectric detection module acquires chlorophyll fluorescence decay level waveforms at the first and second discrete moments after the high-frequency excitation drive signal is turned off. The central processing chip uses the logarithmic difference operator to calculate the relative decay rate characteristic scalar, calculates the first-order difference slope and the second-order difference rate of change of the relative decay rate characteristic scalar, and determines the nutrient stress state of the virus-free strawberry seedlings based on the first-order difference slope and the second-order difference rate of change.

[0021] Preferably, step S101 further includes: adjusting the closed state of the internal optical path in the absence of external excitation light, continuously collecting 30 to 50 sets of diffuse reflection light signals of the greenhouse environment through the time-resolved photoelectric detection module; the central processing chip performs mean filtering and normalization calculation on the 30 to 50 sets of diffuse reflection light signals of the greenhouse environment and solves the statistical discrete variance, and confirms the statistical discrete variance as the environmental diffuse scattering distortion variance.

[0022] Preferably, the calculation of the first-order difference slope and second-order difference rate of change of the relative decay rate feature scalar in step S104, and the determination of the nutrient stress state of the strawberry virus-free seedlings based on the first-order difference slope and second-order difference rate of change, is further refined into the following sub-steps: Step S1041, the central processing chip stores the relative decay rate feature scalar of five consecutive cycles into the sliding window shift register in the order of time generation to construct a time-domain scalar time series; Step S1042, the central processing chip performs first-order difference calculation on the time-domain scalar time series to calculate the first-order difference slope, and obtains the second-order difference rate of change through second-order difference calculation based on the first-order difference slope; when the second-order difference rate of change exceeds the preset safety threshold range, the central processing chip outputs the nutrient stress monitoring results of the strawberry virus-free seedlings.

[0023] Preferably, step S102, which increases the driving pulse width of subsequent pulse cycles based on the environmental diffuse scattering distortion variance, is refined into the following sub-steps: Step S1021, the central processing chip compares the calculated environmental diffuse scattering distortion variance with the internally stored reference variance interval; Step S1022, when the environmental diffuse scattering distortion variance is greater than the upper limit of the reference variance interval, the central processing chip increases the pulse width of the high-frequency excitation driving signal based on the deviation of the environmental diffuse scattering distortion variance from the upper limit, thereby increasing the radiation energy of the excitation source in a single pulse cycle.

[0024] Preferably, step S103, which controls the reverse pulse source to inject a reverse neutralizing charge pulse into the junction capacitance of the time-resolved photodetector module, is refined into the following sub-steps: Step S1031, when the falling edge of the pulse cutoff of the high-frequency excitation drive signal is captured, the reverse pulse source is triggered to turn on; Step S1032, the reverse pulse source injects a reverse bias current into the time-resolved photodetector module, and the pulse amplitude and duration of the reverse neutralizing charge pulse are reverse-modulated by the current duty cycle parameter of the high-frequency excitation drive signal.

[0025] Preferably, step S103, which involves simultaneously disconnecting the bias charge injection path and turning on the signal acquisition and holding circuit, is further refined into the following sub-steps: Step S1033, disconnecting the first charge channel at the instant the high-frequency excitation drive signal is cut off to cut off the bias charge injection path; Step S1034, simultaneously turning on the second charge channel to close the signal acquisition and holding circuit, and resolving the electrical signal transmission path between the output terminal of the photoelectric detection module and the capacitor input terminal of the sample-and-hold circuit by the conduction time of the second charge channel.

[0026] Preferably, in step S104, the time-resolved photodetector module collects the chlorophyll fluorescence decay level waveform at the first discrete moment and the second discrete moment after the high-frequency excitation drive signal is turned off. Specifically, after the residual photogenerated charge is cleared by the reverse neutralization charge pulse, the time-resolved photodetector module collects the chlorophyll fluorescence decay level waveform at the first discrete moment and the second discrete moment within a time window of 10μs to 0.5ms after the high-frequency excitation drive signal is turned off.

[0027] Preferably, the modulation frequency of the multi-frequency excitation beam is 10kHz to 500kHz; the sampling rate of the time-resolved photoelectric detection module is 10MSPS to 100MSPS.

[0028] Preferably, the tracking and monitoring steps include the following: Step S105, the central processing chip continuously records the first-order difference slope and second-order difference rate of change of the same strawberry virus-free seedling at the same growth node during a continuous inspection cycle of 7 to 30 days, and constructs a historical dataset; Step S106, the central processing chip calculates the sliding variance of the historical dataset, and when the sliding variance continues to increase within 3 consecutive inspection cycles and exceeds the preset nutrient stress alarm threshold, it outputs a nutrient stress alarm signal.

[0029] Example 1: In a fully automated, track-based optical inspection line within a factory-style enclosed seedling greenhouse operating at a speed of 0.5 m / s, the system conducts online optical monitoring of the leaves of virus-free strawberry seedlings below the inspection track. Due to mechanical vibrations during track operation, spatial geometric tilt caused by the random growth and unfolding angle of the strawberry seedling leaves, and sudden high-fog diffuse scattering losses caused by different spraying conditions within the greenhouse, the absolute radiation intensity received by the time-resolved photoelectric detection module experiences nonlinear waveform distortion. At the moment of high-frequency strong pulse excitation cutoff, the time-resolved photoelectric detection module... The photogenerated parasitic charge accumulated inside the photodiode junction capacitor generates dissipation hysteresis, thus forming a trailing overshoot waveform at the output end. The optical path fluctuations caused by mechanical vibration and leaf spatial tilt cause nonlinear coupling between the dissipation rate of the photogenerated parasitic charge and the initial threshold, as well as dynamic baseline drift that changes with operating conditions. This dynamic baseline drift causes the overshoot charge at the tail edge of the trailing overshoot waveform to cover and overlap with the microsecond-level fast fluorescence decay kinetic leading edge signal acquisition window that follows. This leads to the loss of the monotonic correspondence between the fluorescence radiation intensity level at subsequent sampling points and the early latent nutrient stress state of strawberry virus-free seedlings.

[0030] To eliminate the overlap of overshoot charges caused by mechanical vibration and abrupt changes in background diffuse scattering, the time-resolved photoelectric detection module continuously reads 20 greenhouse environment diffuse reflectance light signals with a sampling period of 100 μs under conditions without external excitation light. The central processing chip then calculates the operating condition fluctuation variance of the current detection window based on the greenhouse environment diffuse reflectance light signals. When the operating conditions fluctuate and the variance When the fluctuation exceeds the set fluctuation threshold of 0.05, the central processing chip issues an adjustment command to control the timing logic control unit to adjust the drive pulse width parameter for the next stage. The initial value of 50 μs was adjusted to 80 μs to increase the excitation photon capture density per sample. This was done if the operating condition showed fluctuations in variance. If the value is less than or equal to 0.05, the initial value of 50 μs is maintained. In actual operation, the engineering calibration basis and rationale for setting the fluctuation threshold to 0.05 are as follows: This value is determined by continuously sampling 50 sets of ambient diffuse reflection light signals using a time-resolved photoelectric detection module in a normal dark room environment without external excitation light in a closed seedling greenhouse, and by calculating the statistical discrete variance benchmark by the central processing chip. This value is three times the standard deviation boundary of the maximum background noise fluctuation variance of the system. Setting the threshold to 0.05 ensures that when the operating condition fluctuation variance is less than or equal to this value, the system determines that the current track mechanical vibration and minor airflow fluctuations are within the safe tolerance range and no energy reconstruction is required. If the variance exceeds 0.05, it indicates that the external sudden diffuse scattering distortion has reached a certain level, triggering the pulse width adaptive widening mechanism to ensure test accuracy. During the 5μs pre-window period before applying the multi-frequency excitation drive signal, the timing logic control unit sends the first set command to the precision constant current source connected to the time-resolved photodetector module, driving it to inject a subthreshold pre-saturation bias current with an amplitude of 15μA into the anode of the photodiode. The constant ground state photon flow and quota carriers provided by the subthreshold pre-saturation bias current pre-fill the energy level charge traps inside the depletion layer of the photodiode in the light-free dissociation state, thereby solidifying the zero-point dark current noise substrate. The timing logic control unit then adjusts the timing pulse width parameters accordingly. The modulated energy radiation source emits a multi-frequency excitation beam with a center wavelength of 450nm onto the leaves of the virus-free strawberry seedlings. At the moment of the falling edge of the pulse cutoff of the multi-frequency excitation drive signal, the timing logic control unit initiates a 20ns backoff delay window. During the yield delay window During the duration, the timing logic control unit calculates the ground state steady-state diffuse reflection electrical signal based on the acquired ground state steady-state signal. The real-time amplitude and variance changes are used to reconstruct the impedance distribution parameters of the bias charge injection path, and a set of heterogeneous reverse pump charge pulses flowing to the input of the photodiode are output. The transient level and amplitude of the heterogeneous reverse pump charge pulses are compared with the ground-state and steady-state diffuse reflection signals. The variance values ​​are positively correlated, and the amplitude and duration of the reverse-modulated heterogeneous reverse pump charge pulse are related to the current duty cycle parameter of the multi-frequency excitation driving signal.

[0031] By introducing a heterogeneous reverse pump charge pulse to the input terminal of the photodiode to pull and neutralize the residual photogenerated charge retained inside the junction capacitance in situ, the fluctuations in the parasitic charge dissipation rate caused by optical path drift are blocked. The constant ground-state photon flow provided by the subthreshold pre-saturation bias current essentially refers to, under completely dark conditions without external excitation light, forcibly maintaining a constant and extremely low-density non-equilibrium carrier substrate state inside the depletion layer by injecting a weak current of 15 μA in reverse to the anode of the photodiode. The thermally excited carrier transport characteristics in this state are... In terms of performance, it is completely equivalent to a weak, normalized bias photon injection, hence it is called a constant ground state photon current. This constant current allows the deep energy level charge traps inside the depletion layer to capture and fill a fixed number of charge carriers before the arrival of external high-frequency pulsed strong light radiation, thus achieving dynamic saturation. As a result, when the subsequent high-frequency excitation beam is cut off, the trap energy levels inside the depletion layer have been locked in advance, and no more deep delayed charge carriers generated by strong light injection can be released. This eliminates the physical source of long tail overshoot of junction capacitance and solidifies the zero-point dark current noise substrate.

[0032] In the yield delay window At the moment the count is full, the timing logic control unit controls the internal high-speed analog switch to disconnect the first charge channel to cut off the bias charge injection path, and simultaneously turns on the second charge channel to close the signal acquisition and holding circuit. This opens the electrical signal transmission path between the output of the time-resolved photodetector module and the capacitor input of the sample-and-hold circuit, receiving the reconstructed and purified nonlinear fluorescence decay waveform signal. The central processing chip adopts a hardware architecture of coordinated control by a microcontroller and a field-programmable gate array (FPGA). The microcontroller uses a core processor with a floating-point arithmetic unit to calculate the variance of the environmental diffuse scattering distortion and the values ​​of the timing sequence of the sliding window shift register. The FPGA is connected to the microcontroller via a data bus to receive control commands from the microcontroller and generate an internal reference clock with a frequency of 500 MHz through an internal high-frequency phase-locked loop, providing time stepping at the hardware level. For 2-nanosecond deterministic timing control, the falling edge of the high-frequency excitation drive signal pulse is directly controlled to initiate a 20-nanosecond backoff delay window. At the 30-nanosecond node after the pulse turn-off corresponding to the first discrete time and at the 120-nanosecond node after the pulse turn-off corresponding to the second discrete time, sampling trigger pulses are sent to the dual-channel high-speed discrete sampling circuit to control the time-resolved photoelectric detection module to complete signal acquisition. In the hardware switching transient when the backoff delay window is full and the first charge channel is disconnected while the second charge channel is simultaneously turned on, the analog front-end circuit of the signal acquisition and holding loop adopts a fully differential operational amplifier dynamic counter-current circuit and a dumb switch compensation structure to eliminate the transient switching noise caused by parasitic capacitance and charge injection effect during the switching process of the high-speed analog switch. An inverting control signal is connected in parallel at the high-speed analog switch control terminal of the second charge channel to drive the compensation field-effect transistor with the drain and source shorted.

[0033] When the high-speed analog switch is turned on and generates positive parasitic injection charge, the inverting control signal drives the compensation field-effect transistor to instantaneously absorb an equal amount of reverse charge. This controls the overshoot charge generated by channel switching to dissipate and converge the baseline within 4 nanoseconds. Before the 30-nanosecond node after pulse turn-off, the input of the signal acquisition and holding loop is restored to the ground-state baseline level, providing the hardware foundation for subsequent acquisition of the nonlinear fluorescence decay waveform level. The timing logic control unit controls the dual-channel high-speed discrete sampling circuit, at the first discrete moment after pulse turn-off... The corresponding 30ns node after pulse turn-off and the second discrete time. The corresponding radiation intensity level is sampled and read at 120ns after the corresponding pulse is turned off. and The central processing chip uses a logarithmic difference operator to calculate the relative decay rate characteristic scalar. Its calculation formula The system continuously acquires the relative decay rate characteristic scalar with a cycle period of 0.2s. The central processing unit (CPU) performs a first-order difference operation on the time axis to calculate the first-order difference slope of the time-domain scalar time series. Based on the first-order difference slope, a second-order difference operation is performed to obtain the second-order difference rate of change, thereby extracting the differential acceleration scalar. When the differential acceleration scalar When the threshold of nutrient stress (1.2) determined by prior engineering experience is exceeded for three consecutive cycles, a physiological obstruction of the photosynthetic electron transport chain in the strawberry virus-free seedling is established. The output port of the central processing chip directly generates a high-level transition command to drive the external track adjustment component to output the corresponding nutrient stress warning signal. This avoids the overlapping interference of depletion layer trap charges and junction capacitance parasitic charges on the observation window of the fast fluorescence decay front, realizing the acquisition of the fast fluorescence decay front signal. Under the condition of avoiding the mechanical adjustment and alignment of the external optical network geometry, the physical dependence on the absolute light intensity value is removed by utilizing the time-domain relative decay rate characteristics. This filters out the occasional data jump noise caused by sudden vibration of the inspection line track, improving the accuracy of identifying the early latent nutrient stress state of strawberry virus-free seedlings. For the first discrete time, For the second discrete time, The intensity level of the radiation light collected at the first discrete moment. The intensity level of the radiation light collected at the second discrete time. The relative decay rate is a characteristic scalar. The dimension origin and cross-scale calculation logic of the reciprocal of the nanosecond square of the differential acceleration scalar are explained as follows: Since the relative decay rate characteristic scalar is calculated at the bottom layer using the difference in the nanosecond-level time domain window between the first and second discrete moments as the denominator, the physical dimension of the relative decay rate characteristic scalar is the reciprocal of the nanosecond. After the central processing chip continuously reads this characteristic scalar and constructs the time-domain scalar time sequence with an overall inspection cycle of 0.2 seconds, the first-order and second-order difference operations it performs are essentially extracting the dimensionless pure digital rate of change of the relative evolution trend between discrete periods. In order to maintain the surface electrical... The dynamic dimension continuity of the sub-composite relaxation feature is achieved by the central processing chip internally forcibly incorporating a fixed time scaling correction factor when executing the differential operator. This factor is numerically equal to the time scale conversion coefficient corresponding to the conversion of a unit nanosecond to a 0.2-second overall period, thereby directly eliminating the influence of the overall step size. This allows the final output second-order differential rate of change to be directly mapped back to the surface time domain scale, characterized as a differential acceleration scalar whose physical dimension corresponds to the reciprocal of the square of the nanosecond. After the external orbit adjustment component outputs a nutrient stress warning level signal, the external automated fertilization system receives the warning level signal and adjusts the nutrient solution spray flow rate in the corresponding orbital position area, thus completing the quantitative control of the nutrient supply status of the strawberry virus-free seedlings.

[0034] Example 2: When the time-resolved photoelectric detection system encounters background light scattering and floating due to spatial temperature and humidity changes, as well as irregular mechanical displacement of the inspection device, in order to verify the anti-interference performance and parameter adaptive adjustment properties of an online non-destructive monitoring method for nutrient stress in strawberry virus-free seedlings under set conditions, the test was conducted on an optical detection line composed of a simulated track inspection suspension bracket, a multi-frequency semiconductor laser source with a center wavelength of 450nm, and a time-resolved photoelectric detection module containing a needle-shaped positive and negative polarity interlayer, i.e., a PIN silicon photodiode; in order to simulate the seedling temperature... The diffuse light field interference in the room superimposed a Gaussian white noise disturbance with a signal-to-noise ratio of 20dB on the receiving path of the scattered light path. At the same time, the carrier trapping effect at the moment of the strong pulse light cutoff was identified as the physical interference source of the signal baseline displacement. The test group using the method of this invention and the comparison sample group using the conventional photoelectric detection method were both operated in an environment with a patrol movement speed of 0.5m / s and the above-mentioned noise background. At the same time, the initial unsaturated original analog voltage signal output by the time-resolved photoelectric detection module was collected, and its amplitude fluctuation range of 10mV to 500mV was identified as the test reference data.

[0035] Based on the optical detection timing logic, the driving pulse width parameter The technical balance between controlling the population saturation of excited-state electrons and the initial signal level of the fast fluorescence radiation window lies in the mutual constraint between the excitation photon capture density per sampling cycle and the thermal power loss of the detector chip; when the time-resolved photodetector module monitors the operating condition fluctuation variance reflecting the variation of environmental diffuse scattering... When deviating from the preset stable range, the pulse width parameter is adjusted to compensate for the light intensity attenuation caused by disordered scattering of particles in high fog. The test process tends towards the upper limit of the value range, so three parameter gradients are set for this purpose: 50μs, 65μs, and 80μs, corresponding to the endpoints and median of the limited range. Specifically, the adaptive conversion rule for increasing the pulse width of the high-frequency excitation drive signal based on the deviation of the environmental diffuse scattering distortion variance exceeding the upper limit is manifested as a piecewise linear control strategy: the central processing chip performs real-time subtraction between the calculated dynamic operating condition fluctuation variance and the preset upper limit value of 0.05 to obtain the deviation. When it is greater than 0 and less than or equal to 0.10, the timing logic control unit... The pulse width is linearly widened by 3 microseconds for every 0.01% variance deviation, and is cumulatively modulated based on an initial reference pulse width of 50 microseconds. When the environmental conditions deteriorate to the point that the variance deviation exceeds 0.10, the system automatically triggers hard-limiting protection to lock the pulse width at the maximum threshold of 80 microseconds to avoid excessive heat dissipation of the detector chip. This improves the photon capture density of a single sample while maintaining system thermal stability. In tests with large-scale fluctuations in diffuse scattering, the pulse width parameters are compared with those of the fixed driving pulse width parameter of the sample group. With a duration of 50 μs and no reverse current or reverse pulse intervention, the fast fluorescence signal acquired by it suffers waveform attenuation and distortion due to overshoot charge in the junction capacitance.

[0036] In this invention's test group, a subthreshold pre-saturation bias current with an amplitude of 15 μA was reverse-injected into the anode of the photodiode during a 5 μs lead-in window period before the excitation drive pulse was applied to fill the energy level trap. A 20 ns backoff delay window was initiated at the instant the excitation drive pulse was cut off. The input is a ground-state steady-state diffuse reflection electrical signal. The heterogeneous reverse pump charge pulse generated by reverse modulation neutralizes and clears residual parasitic charges. At this point, the measured voltage baseline deviation variance decreases from 0.184 in the initial state to 0.012, confirming the physical offsetting effect of the heterogeneous charge pulse on the carriers residing in the junction capacitance. For the driving pulse width parameter, For the variance of operating conditions, To accommodate the delay window, For the ground-state steady-state diffuse reflection electrical signal, specifically, the deep physical synergistic mechanism between the environmental diffuse scattering distortion variance and the removal of residual photogenerated charge within the junction capacitance is as follows: When the diffuse scattering in the greenhouse environment increases, the total energy of transient background light leaking through the greenhouse fog and mechanical vibrations will cause random fluctuations. This directly changes the initial boundary conditions of the depletion layer and the recombination rate of the ground-state dark carriers in the unexcited state of the photodiode. Since the dissipation hysteresis of residual parasitic charge in the junction capacitance generated by the high-frequency turn-off transient is highly dependent on this initial ground-state carrier concentration, by continuously reading multiple sets of ground-state steady-state diffuse reflection signals... By reflecting the electrical signal and calculating its operating condition fluctuation variance, the degree of baseline drift caused by environmental fluctuations can be indirectly quantified and extracted. The central processing chip uses this variance value as the basis for closed-loop control, dynamically reconstructs the impedance distribution parameters of the bias charge injection path, thereby generating a set of heterogeneous reverse pump charge pulses with amplitude positively correlated with variance. This pump charge pulse introduces a precisely balanced reverse driving force to extract and actively neutralize the dissipated hysteresis charge introduced by the random drift of the ambient optical path, thus purifying the signal that finally enters the fast fluorescence acquisition window and reducing the impact of environmental fluctuations on transient overshoot clearance.

[0037] At the first discrete moment after the excitation pulse is turned off The corresponding pulse cutoff node 30ns and the second discrete time. At the 120ns node after the corresponding pulse cutoff, the high-speed discrete sampling circuit completes voltage extraction for the signal level conversion path; in the driving pulse width parameter With an upper limit of 80 μs, the radiation intensity level measured by the test group of this invention at the first discrete moment. The radiation intensity level at the second discrete moment is 342.5mV. The value is 46.2 mV. The relative decay rate characteristic scalar, which characterizes the electronic recombination relaxation properties of chlorophyll, was calculated using the logarithmic difference operator. It is 0.0222; while when the driving pulse width parameter When broadened to the control group outside the 100 μs boundary beyond the range, the test data show that the relative decay rate characteristic scalar At 0.0223, the growth curve tends to flatten. At this point, the number of excited electrons in the photosynthetic reaction center reaches a dynamic saturation state. Continuing to widen the pulse cannot change the signal output ratio. Instead, it increases thermal noise due to the continuous heat dissipation of the bias current, thus confirming the technical basis for limiting the upper limit of the parameter window to 80μs. As the severity of nutrient stress caused by nitrogen deficiency in strawberry virus-free seedlings increases from mild deficiency to severe stress deficiency in a gradient manner.

[0038] The differential acceleration scalar solved by the test group of this invention The value monotonically increases from 1.45 to 3.82, showing a regular response as the intensity of physiological lesions intensifies, exhibiting a unique quantitative correlation. In contrast, the control group, due to residual overshoot carriers at the tail edge covering the microsecond-level attenuation signal acquisition window, experienced random and disordered fluctuations in its calculated differential acceleration scalar between 0.85 and 2.14, making it impossible to establish corresponding monitoring indicators. Under continuous 48-hour operating stress test, the nutrient stress graded early warning level signal calculated by the test group of this invention maintained a stable online monitoring and identification accuracy of 96.5% for the nutrient stress status of strawberry virus-free seedlings. In contrast, the identification index of the control group decreased to 54.2% with the increase of transient noise caused by track vibration, quantitatively confirming the technical contribution of the method for clearing depletion layer trap charges and junction capacitance parasitic charges in suppressing swaying noise. For the first discrete time, For the second discrete time, The intensity level of the radiation light collected at the first discrete moment. The intensity level of the radiation light collected at the second discrete time. The relative decay rate is a characteristic scalar. For the driving pulse width parameter, It is a differential acceleration scalar.

[0039] Test data confirms that a subthreshold pre-saturation bias current is injected into the photodiode in reverse within the time window before the application of the multi-frequency excitation drive signal to fill the energy level trap, in conjunction with a yield delay window. The method involves internally inputting a heterogeneous reverse pump charge pulse to neutralize residual charge in the junction capacitance. This method stably acquires a fast fluorescence decay waveform signal unaffected by optical path drift and environmental diffuse scattering contamination, maintaining the monitoring accuracy of the relative decay rate characteristic scalar and differential acceleration scalar for the latent nutrient stress state of strawberry virus-free seedlings. Simultaneously, a high-level transition instruction from the central processing chip drives the external track adjustment component to output a corresponding nutrient stress graded early warning level signal. The external automatic fertilizer pump adjusts the opening of the nutrient solution fertilization valve in the corresponding track position seedling area according to the early warning level signal to balance plant height and functional leaf spread. This achieves differentiated and precise supply control of strawberry virus-free seedlings in specific stress areas within a factory-style closed seedling greenhouse. Specifically, the pulse amplitude and duration of the reverse neutralization charge pulse are reverse-modulated by the current duty cycle parameter of the high-frequency excitation drive signal. The specific control law and hardware execution steps are as follows: When the timing logic control unit detects that the duty cycle parameter of the high-frequency excitation drive signal tends to decrease in the current cycle, it indicates that the duration of a single excitation is shortening, and the total amount of delayed parasitic charge accumulated in the depletion layer of the photodiode under strong light is reduced synchronously. At this time, in order to prevent excessive reverse neutralization charge from causing reverse potential overshoot noise, the central processing chip controls the reverse pulse source to proportionally reduce the amplitude of its output reverse bias current to 5 mA, and reverse modulates the duration of the reverse neutralization pulse to shorten it to 15 nanoseconds; conversely, if it is detected that the current duty cycle parameter is widened and increased due to environmental compensation, it indicates that the charge trapping effect in the depletion layer is aggravated, and the reverse pulse source will synchronously and automatically reverse modulate the pulse amplitude to increase to 25 mA and extend the duration to 40 nanoseconds to achieve in-situ counterbalancing.

[0040] Example 3: This example combines Figures 1 to 2 This document describes an online, non-destructive monitoring method for nutrient stress in virus-free strawberry seedlings. Figure 1As shown, in step S101, in the absence of external excitation light, the time-resolved photoelectric detection module collects diffuse reflection light signals from the greenhouse environment. The central processing chip calculates the statistical discrete variance based on multiple sets of diffuse reflection light signals from the greenhouse environment and identifies the statistical discrete variance as the environmental diffuse scattering distortion variance. In step S102, the central processing chip increases the driving pulse width of the pulse period based on the environmental diffuse scattering distortion variance to generate a high-frequency excitation driving signal, driving the excitation light source to radiate a multi-frequency excitation beam onto the leaves of the virus-free strawberry seedlings. In step S103, at the instant the high-frequency excitation driving signal pulse cuts off, the central processing chip... The processing chip controls the reverse pulse source to inject reverse neutralizing charge pulses into the junction capacitance of the time-resolved photodetector module to neutralize the residual photogenerated charge in the junction capacitance. At the same time, the bias charge injection path is disconnected and the signal acquisition and holding circuit is turned on. In step S104, the time-resolved photodetector module acquires the chlorophyll fluorescence decay level waveform at the first discrete time and the second discrete time after the high-frequency excitation drive signal is turned off. The central processing chip uses the logarithmic difference operator to calculate the relative decay rate characteristic scalar. The first-order difference slope and the second-order difference change rate of the relative decay rate characteristic scalar are calculated to determine the nutritional stress status of the strawberry virus-free seedlings.

[0041] like Figure 2 As shown, the system is introduced from the initial node and is in a state of no external excitation detection. In the state of no external excitation light, the time-resolved photoelectric detection module continuously collects the diffuse reflection light signal of the greenhouse environment. The central processing chip calculates the variance of the environmental diffuse scattering distortion. After the environmental diffuse reflection light signal is collected and confirmed to be the variance of the environmental diffuse scattering distortion, the system enters the high-frequency excitation radiation state. According to the variance of the environmental diffuse scattering distortion, the driving pulse width of the pulse period is increased to generate a high-frequency excitation driving signal and radiate a multi-frequency excitation beam. At the moment when the high-frequency excitation driving signal is generated and at the falling edge of the pulse cutoff, the system switches to the charge neutralization and loop switching state, and controls the reverse pulse source to inject into the junction capacitance. A reverse neutralization charge pulse neutralizes the residual photogenerated charge and simultaneously disconnects the bias path and turns on the acquisition circuit. When the yield delay window is full and the residual photogenerated charge inside the junction capacitance is neutralized and cleared in situ, the system enters the discrete time domain signal acquisition state. At the first and second discrete moments after the drive signal is turned off, the time-resolved photodetector module acquires the chlorophyll fluorescence decay level waveform. After extracting the chlorophyll fluorescence decay level waveform at the first and second discrete moments, the system enters the feature calculation and stress determination state. The logarithmic difference operator is used to calculate the relative decay rate feature scalar, solve the first-order difference slope and the second-order difference rate of change, and perform threshold comparison.

[0042] If the second-order difference rate of change is within the preset safety threshold range, the plant is in a healthy supply state. The system resets the detection cycle and returns to the state without external stimulation detection. If the second-order difference rate of change exceeds the preset safety threshold range, it is determined that there is a nutrient stress state and enters the monitoring result output state. Finally, the nutrient stress monitoring results of the strawberry virus-free seedlings are output, and the central processing chip outputs the corresponding alarm signal and provides precise supply regulation.

[0043] Example 4: When the system faces physiological tolerance variations and uneven chlorophyll distribution in different batches of virus-free strawberry seedlings, the rate of change of the acquired fast fluorescence decay waveform fluctuates on the discrete time axis. This causes the characteristics of the first-order difference slope and the second-order difference rate of change to vary across different physiological stages. To establish the discrete calibration basis for the nutrient stress judgment threshold of 1.2 in the method and to open up the operator path, a judgment benchmark was configured through calibration experiments before signal acquisition started. Normal virus-free strawberry seedlings with a stable chlorophyll relative content of 45.2 in functional leaves and no nutrient deficiency symptoms were selected as calibration specimens and placed in a closed artificial incubator with a temperature control accuracy of ±0.5℃ and an illuminance maintained at 12000lx. The dual-channel high-speed discrete sampling circuit was controlled to continuously acquire 100 cycles of standard fluorescence waveform levels at a cycle period of 0.2s. The central processing chip read the first discrete time. The corresponding 30ns node radiation intensity level With the second discrete time Corresponding 120ns node radiation intensity level The characteristic scalar of the relative decay rate under standard conditions is calculated by the internal logarithmic difference operator. The solution results are then written sequentially into adjacent memory cells of the first data temporary storage register to construct a standard time-domain scalar time series. The relative decay rate characteristic scalar of three adjacent periods in the first data temporary storage register is read by the finite difference computing chip. A backward finite difference operation was performed. The first-order difference slope was calculated by subtracting the previous sampling period value from the current sampling period value. Simultaneously, the second-order difference rate of change was calculated by subtracting the first-order difference slope of the previous period from the current first-order difference slope. This allowed the extraction of the baseline differential acceleration scalar. Low-nitrogen mild stress and nitrogen deficiency severe stress with gradually decreasing nitrogen concentrations were applied to observe the trend of differential acceleration changes. Test data showed that the mean value of the baseline differential acceleration scalar for the standard supplied specimen remained at 0.38 with fluctuations not exceeding 0.05, while the differential acceleration scalar under low-nitrogen mild stress... The differential acceleration scalar value rises to the range of 1.25 to 1.62 under severe nitrogen deficiency stress. The value then climbs to above 3.40. Since 1.2 is the dividing point between the upper limit of fluctuation in a healthy, stress-free state and the lower limit of response to mild latent physiological lesions, 1.2 is set as the threshold for determining nutritional stress.

[0044] When the online non-destructive monitoring method operates continuously for more than 90 days in a seedling greenhouse environment, the attenuation of light source radiation, coupled with the irreversible shift in the carrier migration state inside the photodiode in the time-resolved photodetector module over time, causes a slow upward shift in the zero-point dark current level, resulting in a decrease in the ground-state steady-state diffuse reflection signal. The initial potential undergoes a static drift towards a higher level, resulting in fluctuations and variance in operating conditions. With differential acceleration scalar The calculation generates temperature drift deviation, affecting the reproduction results under standard operating conditions. To eliminate baseline offset caused by time-varying photoelectric devices and provide timely calibration, a baseline full-cycle automatic calibration process is configured. Every 24 hours of continuous system operation, the external pulse excitation signal is cut off and converted to a 10-second zero-excitation online self-test sequence. In a dark room without external excitation beam radiation, a dual-channel high-speed discrete sampling circuit continuously reads the residual level of the photodiode static output terminal for 50 cycles to obtain the background zero-point static voltage sequence. The central processing chip calls the mean filtering program to calculate the average dark drift set voltage characteristic value of the current cycle. And set the average dark drift voltage characteristic value As a global zero-point dynamic correction factor, it is latched into the hardware latch of the preceding differential subtractor. When resuming online monitoring, the preceding differential subtractor will then receive the ground-state steady-state diffuse reflection electrical signal. Radiated light intensity level at the first discrete moment and the radiation intensity level at the second discrete time. Subtract the characteristic value of the average dark drift set voltage respectively This is to block dark current temperature drift and eliminate long-term drift residuals, reducing the baseline offset caused by device aging by more than 92.4% and maintaining the variance of operating conditions. With relative decay rate characteristic scalar It remains stable within the preset range.

[0045] Example 5: When the system faces the condition of initial detection optical path reference variation caused by differences in the layout of different seedling greenhouses, before the method is put into operation, the discrete sampling circuit is controlled to collect 100 cycles of lightless reference voltage and calculate its average value to obtain the initial hardware bias level. The signal is then latched into the central processing chip. A standard diffuse reflection ceramic sheet is placed 30cm below the inspection track. The timing logic unit drives the energy radiation source to emit a beam with a center wavelength of 450nm. The time-resolved detection module receives the returned ground-state steady-state diffuse reflection electrical signal. Based on this, the static initial gain coefficient is calculated. Its calculation formula is ,in, The standard reference constant has a value of 300mV; This is the initial hardware bias level. This is the ground state steady-state diffuse reflection electrical signal. This is the static initial gain coefficient.

[0046] static initial gain coefficient After latching, when the system switches to the online non-destructive monitoring state for virus-free strawberry seedlings and the first batch of plants enters the inspection area, in order to prevent the arithmetic jump caused by the initial state of the register being empty, the sequential logic unit will calculate the relative decay rate characteristic scalar within the first 5 cycles. The shift register is filled, and the subsequent level signal is multiplied by the static initial gain factor in the preceding differential subtractor. After compensating for optical path differences, the central processing chip calls the finite difference operator to calculate the differential acceleration scalar. When the threshold value of 1.2 is exceeded for three consecutive cycles, the central processing chip drives the external fertilizer pump to increase the opening of the fertilizer valve by 15.5%, so that the nutrient supply status of the corresponding strawberry virus-free seedling matches the preset physiological development property boundary.

[0047] The embodiments of this application have been described above with reference to the accompanying drawings. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. This application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit of this application and the scope of protection of this invention, and all of these forms are within the protection scope of this application.

Claims

1. A method for online non-destructive monitoring of nutrient stress in virus-free strawberry seedlings, characterized in that, Includes the following steps: In step S101, under the condition of no external excitation light, the time-resolved photoelectric detection module collects the diffuse reflection light signal of the greenhouse environment. The central processing chip calculates the statistical discrete variance based on multiple sets of diffuse reflection light signals of the greenhouse environment and confirms the statistical discrete variance as the environmental diffuse scattering distortion variance. In step S102, the central processing chip increases the driving pulse width of the pulse period according to the variance of the environmental diffuse scattering distortion to generate a high-frequency excitation driving signal, which drives the excitation light source to radiate a multi-frequency excitation beam to the leaves of the strawberry virus-free seedling. Step S103: At the instant the pulse of the high-frequency excitation driving signal is cut off, the central processing chip controls the reverse pulse source to inject a reverse neutralizing charge pulse into the junction capacitance of the time-resolved photodetector module to neutralize the residual photogenerated charge in the junction capacitance. At the same time, the bias charge injection path is disconnected and the signal acquisition and holding circuit is turned on. In step S104, the time-resolved photoelectric detection module acquires chlorophyll fluorescence decay level waveforms at the first and second discrete moments after the high-frequency excitation drive signal is turned off. The central processing chip uses the logarithmic difference operator to calculate the relative decay rate characteristic scalar, calculates the first-order difference slope and the second-order difference rate of change of the relative decay rate characteristic scalar, and determines the nutrient stress state of the virus-free strawberry seedlings based on the first-order difference slope and the second-order difference rate of change.

2. The online non-destructive monitoring method for nutrient stress in virus-free strawberry seedlings according to claim 1, characterized in that, Step S101 further includes: adjusting the closed state of the internal optical path in the absence of external excitation light, continuously collecting 30 to 50 sets of diffuse reflection light signals of the greenhouse environment through the time-resolved photoelectric detection module; the central processing chip performs mean filtering and normalization calculation on the 30 to 50 sets of diffuse reflection light signals of the greenhouse environment and solves the statistical discrete variance, and confirms the statistical discrete variance as the environmental diffuse scattering distortion variance.

3. The online non-destructive monitoring method for nutrient stress in virus-free strawberry seedlings according to claim 1, characterized in that, In step S104, the first-order difference slope and second-order difference rate of change of the relative decay rate feature scalar are calculated. Based on the first-order difference slope and second-order difference rate of change, the nutritional stress state of the virus-free strawberry seedlings is determined, which is further refined into the following sub-steps: Step S1041, the central processing chip stores the relative decay rate feature scalar of five consecutive cycles into the sliding window shift register in the order of time generation to construct a time-domain scalar time series; Step S1042, the central processing chip performs first-order difference calculation on the time-domain scalar time series to calculate the first-order difference slope, and obtains the second-order difference rate of change based on the first-order difference slope through second-order difference calculation. When the second-order difference rate of change exceeds the preset safety threshold range, the central processing chip outputs the nutrient stress monitoring results of the virus-free strawberry seedlings.

4. The online non-destructive monitoring method for nutrient stress in virus-free strawberry seedlings according to claim 1, characterized in that, Step S102, which increases the driving pulse width of subsequent pulse cycles based on the environmental diffuse scattering distortion variance, is further refined into the following sub-steps: Step S1021, the central processing chip compares the calculated environmental diffuse scattering distortion variance with the internally stored reference variance interval; Step S1022, when the environmental diffuse scattering distortion variance is greater than the upper limit of the reference variance interval, the central processing chip increases the pulse width of the high-frequency excitation driving signal based on the deviation of the environmental diffuse scattering distortion variance from the upper limit, thereby increasing the radiation energy of the excitation source in a single pulse cycle.

5. The online non-destructive monitoring method for nutrient stress in virus-free strawberry seedlings according to claim 1, characterized in that, Step S103, controlling the reverse pulse source to inject a reverse neutralizing charge pulse into the junction capacitance of the time-resolved photodetector module, is refined into the following sub-steps: Step S1031, when the falling edge of the pulse cutoff of the high-frequency excitation drive signal is captured, the reverse pulse source is triggered to turn on; Step S1032, the reverse pulse source injects a reverse bias current into the time-resolved photodetector module, and the pulse amplitude and duration of the reverse neutralizing charge pulse are reverse-modulated by the current duty cycle parameter of the high-frequency excitation drive signal.

6. The online non-destructive monitoring method for nutrient stress in virus-free strawberry seedlings according to claim 1, characterized in that, Step S103, which involves simultaneously disconnecting the bias charge injection path and turning on the signal acquisition and holding circuit, is further refined into the following sub-steps: Step S1033, disconnecting the first charge channel at the instant the high-frequency excitation drive signal is cut off to cut off the bias charge injection path; Step S1034, simultaneously turning on the second charge channel to close the signal acquisition and holding circuit, and resolving the electrical signal transmission path between the output terminal of the photoelectric detection module and the capacitor input terminal of the sample-and-hold circuit by the conduction time of the second charge channel.

7. The online non-destructive monitoring method for nutrient stress in virus-free strawberry seedlings according to claim 1, characterized in that, In step S104, the time-resolved photodetector module collects the chlorophyll fluorescence decay level waveform at the first and second discrete moments after the high-frequency excitation drive signal is turned off. Specifically, after the residual photogenerated charge is cleared by the reverse neutralization charge pulse, the time-resolved photodetector module collects the chlorophyll fluorescence decay level waveform at the first and second discrete moments within a time window of 10μs to 0.5ms after the high-frequency excitation drive signal is turned off.

8. The online non-destructive monitoring method for nutrient stress in virus-free strawberry seedlings according to claim 1, characterized in that, The modulation frequency of the multi-frequency excitation beam is 10kHz to 500kHz; the sampling rate of the time-resolved photoelectric detection module is 10MSPS to 100MSPS.

9. The online non-destructive monitoring method for nutrient stress in virus-free strawberry seedlings according to claim 1, characterized in that, The tracking and monitoring steps include the following: Step S105, the central processing chip continuously records the first-order difference slope and second-order difference rate of change of the same strawberry virus-free seedling at the same growth node during a continuous inspection cycle of 7 to 30 days, and constructs a historical dataset; Step S106, the central processing chip calculates the sliding variance of the historical dataset, and when the sliding variance continues to increase within 3 consecutive inspection cycles and exceeds the preset nutrient stress alarm threshold, it outputs a nutrient stress alarm signal.

Images