Intelligent monitoring and precision control system for edible mushroom growth

Abstract

The application discloses an intelligent monitoring and precise regulation system for edible mushroom growth, and relates to the technical field of edible mushroom growth monitoring.The system comprises three modules, namely a multi-modal in-situ sensing module, a linkage prediction modeling module and a coupling decision module.The multi-modal in-situ sensing module receives information about edible mushroom varieties, calls corresponding reference and adaptive parameters, synchronously collects multi-dimensional data such as ion flow sources, and generates a data set.The linkage prediction modeling module constructs a three-in-one linkage prediction model based on the data set, accurately analyzes ion flow activity indexes and growth states, and predicts dynamic changes in the future 1-12 hours.The coupling decision module combines analysis, decides optimal environmental parameter combinations and multi-actuator collaborative regulation strategies.The system takes ion flow source signals as the core, breaks through the limitations of existing signal separation, low prediction accuracy and one-size-fits-all regulation, and improves the accuracy and intelligent level of monitoring and regulation, thereby guaranteeing the yield and quality of large-scale cultivation of edible mushrooms.

Description

Technical Field

[0001] This invention relates to the field of edible fungi growth monitoring technology, specifically to an intelligent monitoring and precise control system for edible fungi growth. Background Technology

[0002] Edible fungi, as a type of edible and medicinal microorganism with rich nutrition and unique flavor, have been cultivated in a large-scale and standardized industry. As the core place for large-scale cultivation of edible fungi, the stable control of the internal microenvironment parameters and the physiological metabolic state of edible fungi in the target cultivation room is the key to ensuring the yield and quality of edible fungi. Therefore, an intelligent monitoring and precise control system for the growth of edible fungi is needed.

[0003] Existing technologies, such as the invention application patent with announcement number CN121807072A, disclose a method for monitoring and making decisions on the growth of edible fungi, including: acquiring quantitative data of varieties, time-series data of growth environment, and image data of growth status, and generating multi-source structured features; completing feature fusion through a dynamic feature fusion model to perform feature fusion on the multi-source structured features and generate a fused feature vector; inputting the fused feature vector into an edible fungi growth model and outputting the growth status of edible fungi after a predetermined time; inputting the predicted value, stage growth target, stage-specific environmental response coefficient matrix, and regulation safety factor output by the edible fungi growth model into a quantitative decision-making model and outputting a quantitative regulation amount of environmental parameters; and regulating the environmental parameters of edible fungi according to the quantitative regulation amount of environmental parameters.

[0004] Regarding the above-mentioned solutions, the inventors of this application have discovered at least the following technical problems with the aforementioned technologies: 1. In the existing edible fungus cultivation monitoring process, signal acquisition and analysis have significant limitations, making it difficult to achieve a comprehensive and accurate characterization of the fungal growth state. This has become one of the core bottlenecks restricting the intelligent upgrading of cultivation. Currently, most existing monitoring methods only collect signals of a single type. Even if ion current signals and metabolic signals are collected simultaneously in some scenarios, the intrinsic correlation between the two has not been established, and cross-scale fusion analysis has not been achieved. It is impossible to comprehensively and accurately characterize the growth state of edible fungi by reflecting the synergistic effect of the microscopic electrophysiological activities of the fungi through ion current signals and the macroscopic growth and metabolic state of the fungi through metabolic signals. This results in insufficient correlation of monitoring data, making it difficult to capture subtle changes in the growth process of edible fungi from multiple dimensions, and failing to provide comprehensive data support for subsequent growth state analysis.

[0005] 2. Existing edible fungi growth prediction and parameter retrieval systems lack sufficient accuracy, failing to meet the refined requirements of large-scale cultivation and unable to provide reliable decision-making basis for cultivation regulation. Although database technology has been applied in the field of edible fungi cultivation, most databases have not formed a refined index structure of variety-growth stage-parameter. Parameters retrieved through RFID identification are mostly general benchmark parameters, unable to retrieve more personalized parameters based on the specific inoculation time, substrate formula, and other information of the current batch of edible fungi. At the same time, existing prediction models are mostly built based on single-type parameters, failing to fully integrate multi-dimensional sensing data. The prediction accuracy for the evolution of microenvironmental parameters and the growth status of edible fungi is low, unable to accurately predict the dynamic changes in the growth status of edible fungi in the future, and unable to meet the prediction needs of refined cultivation.

[0006] 3. Existing control strategies for edible fungi cultivation lack specificity and synergy, and their level of intelligence needs improvement. They cannot achieve precise control of fungi growth, making it difficult to guarantee the yield and quality of large-scale cultivation. Current control strategies mostly rely on preset fixed parameter combinations and actuator operation logic, without dynamically adjusting based on the real-time growth status and future growth trends of edible fungi. Even in scenarios where simple control is required, differentiated control schemes are not designed for different growth states and predicted trends, resulting in insufficient targeting of control strategies. Furthermore, there is a lack of collaborative linkage mechanisms between multiple actuators, making it impossible to achieve precise coordinated control of each actuator based on suitable environmental parameter combinations. This leads to poor control effects and makes it difficult to stably maintain the optimal environmental conditions required for fungi growth. Summary of the Invention

[0007] To address the aforementioned technical shortcomings, the present invention aims to provide an intelligent monitoring and precise control system for edible fungi growth.

[0008] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides an intelligent monitoring and precise control system for edible fungi growth, including the following modules: a multimodal in-situ sensing module: used to receive information on the variety of edible fungi in the current batch of cultivation in the target cultivation room, thereby calling up the corresponding physiological metabolic baseline parameters and growth environment adaptation parameters of the edible fungi, and then simultaneously collecting multi-dimensional data on the ion flow source, physiological metabolism, three-dimensional environment and volatile organic compound characteristics of the edible fungi, and generating a multimodal sensing dataset corresponding to each batch of cultivation edible fungi.

[0009] Linked Prediction Modeling Module: Based on the multimodal sensing dataset corresponding to edible fungi, it constructs a three-in-one linked prediction model of microenvironmental parameters, physiological metabolic indicators, and fungal growth status. Through the built-in cross-scale coupling sub-model of mycelial electrical signals and metabolic status, it realizes cross-scale fusion of electrical signals and metabolic signals, analyzes and obtains the ion current activity index corresponding to edible fungi, and then analyzes the growth status of edible fungi and predicts the dynamic changes of microenvironment and edible fungi growth status in the next 1-12 hours.

[0010] Coupled Decision and Precise Control Module: Based on the analysis and prediction results of edible fungi and the early warning trigger signal output by the mycelial electrical signal-metabolic state cross-scale coupling sub-model, combined with the corresponding growth benchmark parameters of edible fungi, it determines the optimal combination of environmental parameters and multi-actuator collaborative control strategy for the target cultivation room.

[0011] The beneficial effects of this invention are as follows: 1. This embodiment takes the ion current source signal of edible fungi as the core innovative entry point, achieving a breakthrough in signal fusion monitoring, and completely solving the pain points of existing technologies that ignore the source of ion current and have no correlation between signal separation. The system accurately collects the ion current source signal and ion current action potential signal of edible fungi through a multimodal in-situ sensing module, breaking through the limitations of existing technologies that only collect single electrical signals or do not pay attention to the characteristics of the ion current source; relying on the built-in mycelial electrical signal-metabolic state cross-scale coupling sub-model, after time synchronization, noise reduction and enhancement processing, it directly establishes the intrinsic correlation between the ion current source parameters and mycelial respiration rate, nutrient consumption rate and VOCs fingerprint spectrum, realizing the cross-scale fusion of microscopic ion current source signals and macroscopic metabolic signals without the need to build an additional mapping model, filling the gap in signal fusion monitoring technology with ion current source as the core in the field of edible fungi, making the monitoring more in line with the microscopic physiological activity of fungi, and providing accurate and reliable data support for subsequent analysis and prediction.

[0012] 2. The innovation of this solution in parameter calling and growth prediction lies in the precise application of ion current source signals, which significantly improves the precision of cultivation. By analyzing the information on the mushroom stick tags using an RFID reader and combining it with a refined database index to call personalized parameters, and by using ion current source parameters as the core input, a three-in-one linkage prediction model is integrated. This overcomes the limitations of existing models that do not incorporate ion current source signals and have low prediction accuracy. The model integrates ion current source data, multi-dimensional sensing data, and historical time-series data, which can accurately capture subtle changes in ion current source signals, thereby accurately predicting the dynamic changes in the microenvironment and growth status in the next 1-12 hours. The prediction accuracy is far superior to existing single-parameter models, highlighting the creative application value of ion current source signals in the prediction process and providing a reliable basis for regulatory decisions.

[0013] 3. The core innovation in the precision of growth state analysis in this embodiment lies in constructing a judgment system based on the ion current source signal, thus solving the problem of vague and one-sided judgments in existing technologies. It abandons the existing single-indicator judgment model, using ion current source parameters as a foundation and a linear mapping algorithm to accurately generate an ion current activity index, truly reflecting the microscopic electrophysiological activity of the bacteria. Combined with metabolic parameters and VOCs fingerprinting and other multi-dimensional indicators, it clarifies the precise parameter ranges for the three growth states. The introduction of ion current source parameters allows growth state judgment to penetrate from the macroscopic level to the microscopic physiological level, breaking through the limitations of existing judgments that only focus on macroscopic indicators, significantly improving the scientific rigor of the judgment, providing clear guidance for differentiated regulation, and demonstrating the creativity of ion current source analysis applications.

[0014] 4. The key innovation of this embodiment in terms of synergistic and differentiated regulation strategies lies in achieving precise regulation based on the ion current source signal, significantly improving the level of intelligent cultivation and the regulation effect. Combining the growth status analysis results, trend predictions, and early warning signals derived from the ion current source signal, and targeting three predicted trends of weak growth status, it matches the corresponding optimal environmental parameter combinations and multi-actuator synergistic regulation strategies, breaking through the limitations of existing fixed regulation that does not incorporate the ion current source signal. Through real-time feedback of bacterial microscopic activity via the ion current source signal, it achieves coordinated linkage among various actuators, precisely adjusting the regulation rhythm, stably maintaining the optimal growth environment, effectively alleviating the problem of weak growth, ensuring the yield and quality of large-scale cultivation, fully demonstrating the creative application of the ion current source in the regulation process, and possessing extremely high industrial application value. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the system structure connection of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Examples of embodiments of the present invention Figure 1As shown, an intelligent monitoring and precise control system for edible fungi growth includes the following modules: a multimodal in-situ sensing module, a linkage prediction modeling module, and a coupled decision-making and precise control module.

[0019] The linkage prediction modeling module is connected to the multimodal in-situ sensing module and the coupled decision and precise control module, respectively.

[0020] Multimodal in-situ sensing module: Used to receive information on the current batch of edible fungi in the target cultivation room, thereby calling up the corresponding physiological and metabolic baseline parameters and growth environment adaptation parameters of the edible fungi, and then simultaneously collecting multi-dimensional data on the ion flow source, physiological metabolism, three-dimensional environment and volatile organic compound characteristics of the edible fungi, generating multimodal sensing datasets for each batch of edible fungi.

[0021] In a specific embodiment, the process of calling the physiological metabolic baseline parameters and growth environment adaptation parameters corresponding to the edible fungi is as follows: When the current batch of edible fungi spawn enters the designated cultivation area of ​​the target mushroom house, the unique identification tag on the surface of the spawn is first read by an RFID reader deployed at the entrance of the mushroom house. The variety, inoculation time, and substrate formula information corresponding to each batch of edible fungi are then parsed to obtain the information. The database is automatically sent to the associated database. The database is indexed by variety-growth stage-parameter classification. The physiological metabolic baseline parameters and growth environment adaptation parameters corresponding to different growth stages of the edible fungi are extracted. The physiological metabolic baseline parameters include the mycelial respiration rate threshold, nutrient consumption rate range, and VOCs characteristic spectrum. The growth environment adaptation parameters include the optimal range of temperature and humidity, CO2 concentration, light spectrum, and substrate EC value.

[0022] In a specific embodiment, the generation of the multimodal sensing dataset corresponding to each batch of cultivated edible fungi is carried out as follows: First, the mycelial electrophysiological microelectrode array is embedded in the matrix during the preparation or inoculation stage of the fungi sticks. The in-situ physiological microsensor array is uniformly inserted into the fungi sticks of each shelf. The three-dimensional gridded environmental sensing module is arranged in the target cultivation room according to a spatial grid of 50cm×50cm×30cm. The VOCs electronic nose sensing unit, the near-infrared spectroscopy non-invasive detection unit, and the high-definition visual acquisition unit are deployed in each cultivation section.

[0023] Secondly, the mycelial electrophysiological microelectrode array collects the action potential signal generated by the transmembrane ion current of the mycelium and extracts the original characteristic parameters of the ion current; the in-situ physiological microsensor array collects the physiological metabolic characteristic parameters of the mycelium; the three-dimensional gridded environmental perception module collects the three-dimensional environmental characteristic parameters in the mushroom house; the VOCs electronic nose sensing module collects the volatile organic compound fingerprint spectrum during the growth process of edible fungi; the near-infrared spectroscopy non-invasive detection module non-invasively detects the metabolic parameters of soluble sugar, nitrogen and lignin degradation rate in the matrix; and the high-definition visual acquisition module captures images of changes in fungal morphology.

[0024] Finally, all collected data are format-converted and stored in a directory structure of variety-cultivation batch-data type to form a multimodal sensing dataset corresponding to each cultivation batch of edible fungi.

[0025] Linked Prediction Modeling Module: Based on the multimodal sensing dataset corresponding to edible fungi, it constructs a three-in-one linked prediction model of microenvironmental parameters, physiological metabolic indicators, and fungal growth status. Through the built-in cross-scale coupling sub-model of mycelial electrical signals and metabolic status, it realizes cross-scale fusion of electrical signals and metabolic signals, analyzes and obtains the ion current activity index corresponding to edible fungi, and then analyzes the growth status of edible fungi and predicts the dynamic changes of microenvironment and edible fungi growth status in the next 1-12 hours.

[0026] In a specific embodiment, the construction process of the three-in-one linkage prediction model of microenvironment parameters, physiological metabolic indicators and bacterial growth status is as follows: adopting the fusion mode of mechanism modeling + attention mechanism neural network, a three-in-one linkage prediction model is constructed, which includes a dynamic evolution sub-model of microenvironment parameters, a response sub-model of physiological metabolic indicators and a prediction sub-model of bacterial growth status, and a built-in cross-scale coupling sub-model of mycelial electrical signal and metabolic status.

[0027] Among them, the microenvironment parameter dynamic evolution sub-model is constructed based on the multi-physics field coupling theory and is used to simulate the spatial distribution and dynamic evolution law of microenvironment parameters; the physiological metabolic index response sub-model is constructed based on the edible fungus growth kinetic equation and is used to characterize the correlation response relationship between physiological metabolic parameters and the ion current source signal; the cell growth state prediction sub-model is constructed based on the improved attention mechanism neural network, with the comprehensive metabolic activity index output by the mycelial electrical signal-metabolic state cross-scale coupling sub-model as the core input, strengthening the dynamic attention to the ion current source signal, and realizing the accurate training and dynamic calibration of the model.

[0028] It should be noted that the specific implementation of the fusion mode of mechanism modeling + attention mechanism neural network is as follows: The mechanism modeling part is based on the growth physiology of edible fungi, the coupling mechanism between microenvironment parameters and fungal growth, to build the basic framework of the model, and to determine the core parameter range of each sub-model and the correlation logic between parameters; The attention mechanism neural network part, for multi-dimensional input data, focuses on the ion flow source signal and comprehensive metabolic activity indicators through the attention weight allocation algorithm, and calibrates and optimizes the output of mechanism modeling to finally form the fused prediction model, which takes into account both the theoretical rationality and prediction accuracy of the model.

[0029] The specific implementation of the sub-model for the dynamic evolution of microenvironmental parameters is as follows: Based on the multi-physics coupling theory, the coupling effects of temperature field, humidity field, CO2 concentration field and light field are integrated. Numerical simulation algorithm is used to simulate the spatial distribution differences of various microenvironmental parameters within a 50cm×50cm×30cm spatial grid in the target culture room. At the same time, the historical time series data of the microenvironment in the past 24 hours are input. Through model calculation, the fluctuation amplitude, rate of change and spatial distribution trend of various microenvironmental parameters in the next 1-12 hours are predicted, and the evolution law data of microenvironmental parameters are output.

[0030] The specific implementation of the physiological metabolic index response sub-model is as follows: Based on the edible fungus growth kinetic equation, input the ion current source parameters (ion current peak half-width, ion current frequency drift coefficient, ion current memory effect decay rate, and ion current burst interval variation coefficient) and physiological metabolic parameters (mycelial respiration rate, nutrient consumption rate, etc.) to establish a quantitative correlation between the two; at the same time, input the fusion results of the mycelial electrical signal-metabolic state cross-scale coupling sub-model to optimize and calibrate the correlation, so as to achieve accurate characterization of the correlation response between physiological metabolic parameters and ion current source signals.

[0031] The specific implementation of the improved attention mechanism neural network is as follows: The attention weight allocation algorithm is optimized, and the comprehensive metabolic activity index output by the mycelial electrical signal-metabolic state cross-scale coupling sub-model is set as a high-weight input term, with its weight ratio higher than other input parameters; real-time fluctuation characteristics of the ion current source signal are introduced, and when the fluctuation of the ion current source parameter exceeds the preset range, its attention weight is automatically increased; during model training and calibration, the fusion characteristics of the ion current source signal and metabolic signal are used as the core supervision term to dynamically adjust the network parameters, complete the model training and calibration, and ensure accurate matching between model input and output.

[0032] In a specific embodiment, the cross-scale fusion of electrical signals and metabolic signals is achieved as follows: the action potential signal of transmembrane ion flow of mycelium is collected by the mycelium electrophysiological microelectrode array of the multimodal in-situ sensing module, and the mycelium respiration rate, nutrient consumption rate and VOCs fingerprint spectrum are collected simultaneously by the in-situ physiological microsensor array and the VOCs electronic nose sensing module.

[0033] By using the timing synchronization unit of the coupled sub-model, the metabolic signal is calibrated at the millisecond level to eliminate misalignment based on the acquisition timing of the ion flow electrical signal. Wavelet denoising algorithm is used to process the electrical signal to remove interference. The metabolic signal drift is corrected and core features are extracted by the adaptive baseline calibration algorithm. The intrinsic correlation between the ion flow action potential signal, the ion flow source parameters and the mycelial respiration rate, nutrient consumption rate and VOC fingerprint spectrum is directly established.

[0034] It should be noted that the synchronous acquisition of each sensing unit in the multimodal in-situ sensing module is implemented as follows: After the mycelial electrophysiological microelectrode array is embedded inside the substrate of the mushroom stick, it acquires the action potential signal generated by the transmembrane ion current of the mycelium in real time, with the sampling frequency set to 1kHz to ensure accurate capture of subtle fluctuations in the ion current signal; the in-situ physiological microsensor array is inserted into the mushroom stick to synchronously acquire the mycelial respiration rate and nutrient consumption rate, with the sampling frequency set to 0.5kHz; the VOCs electronic nose sensing module is deployed in each cultivation section to collect the volatile organic compounds released during the growth of edible fungi in real time and generate a VOCs fingerprint spectrum, with the sampling frequency set to 0.1kHz. The acquisition actions of the three are started synchronously to ensure the consistency of the signal acquisition time sequence, laying the foundation for subsequent signal fusion.

[0035] The millisecond-level calibration implementation of the timing synchronization unit of the coupled sub-model is as follows: Based on the acquisition timing of the ion current signal, the acquisition timestamp of each group of ion current signals is recorded. At the same time, the acquisition timestamps corresponding to the metabolic signals (mycelial respiration rate, nutrient consumption rate, VOCs fingerprint spectrum) are obtained. The acquisition time difference between the two types of signals is calculated through the timing synchronization algorithm. The metabolic signal is calibrated at the millisecond level to accurately align the timestamp of the metabolic signal with the timestamp of the corresponding ion current signal, completely eliminating the signal misalignment caused by the difference in acquisition rate of different sensing units, and ensuring accurate matching of the two types of signals in the time dimension.

[0036] The specific implementation of the wavelet denoising algorithm for processing electrical signals is as follows: The existing wavelet denoising algorithm (a conventional technique in the field of signal processing; this solution does not modify the algorithm itself, but only applies it to ion flow electrical signal processing) is adopted. Specifically, the db4 wavelet basis is selected, and the acquired ion flow action potential signal is decomposed into three layers of wavelets. After decomposition, low-frequency signal components (corresponding to the original characteristics such as the half-width of the ion flow peak and the frequency drift coefficient) are retained, while high-frequency interference components (mainly electromagnetic interference from the mushroom house and sensor noise) are removed. The signal is then reconstructed through inverse wavelet transform to obtain the denoised ion flow electrical signal, ensuring the purity of the ion flow signal.

[0037] The specific implementation of the adaptive baseline calibration algorithm to correct metabolic signal drift is as follows: An existing adaptive baseline calibration algorithm (a conventional technique in the field of signal processing; this solution does not modify the algorithm itself, but only applies it to metabolic signal processing) is used to capture the baseline change trend of metabolic signals (mycelial respiration rate, nutrient consumption rate, VOCs fingerprint spectrum) in real time. A baseline drift threshold of 5% is set. When the baseline shift of the metabolic signal is detected to exceed this threshold, the baseline calibration program is automatically started. The signal shift is corrected through a linear fitting algorithm, while core features such as characteristic peaks and slopes of the metabolic signal are extracted to ensure the stability and effectiveness of the metabolic signal.

[0038] The specific implementation of establishing the intrinsic correlation between ion current-related signals and metabolic signals is as follows: Based on the physiological laws of edible fungi growth, and combined with the intrinsic connection between mycelial electrophysiological activities and metabolic activities, the noise-reduced and calibrated ion current action potential signals and ion current source parameters are correlated one-to-one with the corrected mycelial respiration rate, nutrient consumption rate, and VOCs fingerprint spectrum. A parameter mapping table is established to clarify the corresponding change law between parameters such as ion current peak half-width and frequency drift coefficient and metabolic parameters. Without the need to construct an additional mapping model, the intrinsic correlation between the two can be accurately established directly, providing support for the subsequent generation of comprehensive metabolic activity indicators.

[0039] In a specific embodiment, the analysis yields the ion current activity index corresponding to the edible fungi. The specific analysis process is as follows: First, parameter extraction: using a cross-scale coupling sub-model of mycelial electrical signal-metabolic state, the intrinsic parameters of ion current in edible fungi are extracted. The intrinsic parameters of ion current include the half-width at the ion current peak. Ion current frequency drift coefficient ion current memory effect attenuation rate and the coefficient of variation of the ion burst interval The extracted ion current source parameters are preprocessed and normalized to the 0-1 range.

[0040] It should be noted that, firstly, the action potential signal of the mycelial transmembrane ion current after wavelet denoising is input into the coupled sub-model. The model first performs feature recognition on the signal, captures the peak waveform in the ion current signal, and measures the time span from the peak value to half the peak value for each peak, thereby obtaining the half-width of the ion current peak. Then, the frequency offset of the ion current signal per unit time is calculated, and the ion current frequency drift coefficient is determined by combining the signal acquisition time and the initial frequency value. At the same time, the time when the amplitude of the ion current signal decays to 1 / e of the initial amplitude after the burst is recorded, and the ion current memory effect decay rate is calculated by combining the signal decay law. Finally, the time interval of multiple consecutive ion current bursts is statistically analyzed, and the coefficient of variation of the ion current burst interval is obtained by calculating the ratio of the standard deviation of the interval to the average value, thereby completing the accurate extraction of the four types of ion current source parameters.

[0041] Then, the ion current activity index is generated: using a linear mapping algorithm, the ion current activity index is generated based on the preprocessed ion current source parameters, as shown in the formula: The ion current activity index was obtained. .

[0042] In a specific embodiment, the analysis of the growth status of edible fungi is carried out as follows: ion current activity index, mycelial respiration rate, nutrient consumption rate, soluble sugar, nitrogen, lignin degradation rate and VOC fingerprint spectrum in the matrix are retrieved from the multimodal sensing dataset.

[0043] If the ion current activity index is between 2.2 and 3.0, the mycelial respiration rate is between 0.8 and 1.2, the nutrient consumption rate is between 0.5 and 0.8, the degradation rate of soluble sugars in the substrate is 15%-20%, the degradation rate of nitrogen is 10%-15%, and the VOCs fingerprint spectrum completely matches the VOCs characteristic spectrum in the physiological metabolic baseline parameters, then the edible fungi are considered to be in normal growth status.

[0044] If the ion current activity index is between 3.0 and 4.0, the mycelial respiration rate is between 1.2 and 1.5, the nutrient consumption rate is between 0.8 and 1.2, the degradation rate of soluble sugars in the substrate is 21%-28%, the degradation rate of nitrogen is 16%-22%, and the intensity of the characteristic peaks in the VOCs fingerprint spectrum that characterize vigorous mycelial growth is higher than the benchmark value, then the edible fungi are judged to be in a vigorous growth state.

[0045] If the ion current activity index is between 1.0 and 2.2, the mycelial respiration rate is between 0.5 and 0.8, the nutrient consumption rate is between 0.3 and 0.5, the degradation rate of soluble sugars in the substrate is 8%-14%, the degradation rate of nitrogen is 6%-9%, and there are no abnormal characteristic peaks in the VOCs fingerprint spectrum, then the growth status of edible fungi is judged to be weak.

[0046] In a specific embodiment, the prediction of the dynamic changes in the microenvironment and the growth status of edible fungi in the next 1-12 hours is carried out as follows: First, the ion flow source parameters, physiological metabolic parameters, microenvironment parameters, historical time series data of the past 24 hours, the physiological metabolic baseline parameters of the edible fungi, and the growth environment adaptation parameters of the multimodal sensing dataset are uniformly input into the three-in-one linkage prediction model, and the output microenvironment parameter evolution law, physiological metabolic response mapping relationship, ion flow activity time series change trend, and coupling correlation law of environment-metabolism-fungal growth are displayed.

[0047] If all four intermediate results are stable, the microenvironment parameters evolve smoothly and the fluctuation range meets the requirements of the growth environment adaptation parameters, the physiological metabolic response is stable and the parameters do not deviate, the ion current activity is gradual and the IFAI fluctuation range is ≤0.3, and the multi-parameter coupling is synergistic and there is no mismatch, then it is predicted that the growth status of edible fungi will remain normal for the next 1-12 hours.

[0048] If the four intermediate results show positive evolution, the microenvironment parameters gradually tend to the high value of the growth environment adaptation range and the fluctuation range is ≤ ±0.3℃, ±2%RH, ±30ppm, the physiological metabolism shows a positive response and the parameters gradually approach the vigorous growth threshold, the ion current activity time sequence shows a continuous upward trend and the IFAI increases by 0.1 to 0.2 per day, and the multi-parameter coupling shows vigorous growth characteristics without any imbalance, then it is predicted that the growth status of edible fungi will remain vigorous in the next 1-12 hours.

[0049] If the four intermediate results show mild mismatch, with microenvironment parameters slightly deviating from the optimal range for growth environment with fluctuations of ±0.3~0.5℃, ±2%~3%RH, ±30~50ppm, weak physiological metabolic response with parameters deviating from the baseline threshold by 5%~10%, ion current activity showing a slow downward trend with IFAI decreasing by 0.05~0.1 per hour, and mild imbalance of multi-parameter coupling without serious mismatch, then the growth status of edible fungi is predicted to gradually weaken in the next 1-12 hours.

[0050] Coupled Decision and Precise Control Module: Based on the analysis and prediction results of edible fungi and the early warning trigger signal output by the mycelial electrical signal-metabolic state cross-scale coupling sub-model, combined with the corresponding growth benchmark parameters of edible fungi, it determines the optimal combination of environmental parameters and multi-actuator collaborative control strategy for the target cultivation room.

[0051] In a specific embodiment, the decision-making process for determining the optimal combination of environmental parameters corresponding to the target cultivation room is as follows: if it is determined that the growth status of edible fungi is weak, and it is predicted that the growth status of edible fungi will remain normal for the next 1-12 hours, then the optimal combination of environmental parameters is the weak steady-state adaptation combination.

[0052] It should be noted that the slightly weak steady-state adaptation combination is as follows: the temperature is maintained at the median of the growth environment adaptation range ±0.3℃, the humidity is maintained at the median of the growth environment adaptation range ±2%RH, the CO2 concentration is controlled at 350-450ppm, the light spectrum is stably output according to the adaptation value of the current growth stage, and the matrix EC value is maintained at 1.2-1.8mS / cm. This combination can stabilize the ion current activity (keeping IFAI in the range of 2.2 to 3.0), avoid the decline of the growth state, and adapt to the trend of microenvironment stability.

[0053] If the growth status of edible fungi is determined to be weak, and it is predicted that the growth status of edible fungi will continue to be vigorous within the next 1-12 hours, then the optimal combination of environmental parameters is the weak-to-vigorous adaptation combination.

[0054] It should be noted that the slightly weak rising state adaptation combination is as follows: the temperature is gradually increased to the upper limit of the growth environment adaptation range (increase ≤0.3℃ / h), the humidity is maintained at the upper limit of the growth environment adaptation range ±2%RH, the CO2 concentration is controlled at 400-500ppm, the light intensity is gradually increased to the upper limit of the growth environment adaptation range (increase ≤5% / h), and the substrate EC value is maintained at 1.5-2.0mS / cm. This adaptation growth state shows a positive evolution trend, which promotes the enhancement of mycelial metabolic activity.

[0055] If the growth status of edible fungi is determined to be weak, and it is predicted that the growth status of edible fungi will gradually weaken in the next 1-12 hours, then the optimal combination of environmental parameters is the weak and slow-decreasing adaptation combination.

[0056] It should be noted that the slightly weaker, gradual reduction adaptation combination is as follows: the temperature is adjusted back to the midpoint of the growth environment adaptation range (adjustment range ≤ 0.2℃ / h), the humidity is adjusted back to the midpoint of the growth environment adaptation range ± 1%RH, the CO2 concentration is controlled at 300-400ppm, the light spectrum is maintained at the current adaptation value, and the matrix EC value is adjusted to 1.0-1.5mS / cm. This inhibits the expansion of microenvironment parameter deviations, alleviates the trend of weakening growth status, and stabilizes ion current activity.

[0057] In a specific embodiment, the decision-making process for the multi-actuator coordinated regulation strategy corresponding to the target cultivation room is as follows: if it is determined that the growth state of edible fungi is weak and it is predicted that the growth state of edible fungi will remain normal for the next 1-12 hours, then the multi-actuator coordinated regulation strategy is a weak steady-state coordinated regulation strategy.

[0058] It should be noted that the weak steady-state coordinated control strategy involves activating the zoned temperature control module, zoned humidity control module, and circulating ventilation module, using a fine-tuning mode (environmental parameter adjustment range of ±3%), and fine-tuning the temperature, humidity, and CO2 concentration at a frequency of 10 minutes per cycle to maintain the stability of the adapted environmental parameter combination; the adjustable spectral illumination module maintains the current output parameters without adjustment; and a parameter verification mechanism is set up at 20 minutes per cycle, using the ion current source parameters and IFAI real-time feedback to ensure the control effect and avoid over-control.

[0059] If the growth state of edible fungi is determined to be weak, and it is predicted that the growth state of edible fungi will remain vigorous within the next 1-12 hours, then the multi-actuator coordinated regulation strategy is a weak-to-rising-state coordinated regulation strategy.

[0060] It should be noted that the weak-to-high-level synergistic regulation strategy involves activating the zoned temperature control module, zoned humidity control module, zoned atmosphere control module, and adjustable spectral illumination module, adopting a moderate regulation mode (environmental parameter adjustment range of ±5%, spectral ratio adjustment range of ±10%), and gradually adjusting temperature, humidity, CO2 concentration, and light intensity at a frequency of 5 minutes / time to approach the upper limit of the suitable environmental parameter combination; prioritizing the activation of the circulation ventilation module to ensure the uniformity of microenvironmental parameters; and setting a parameter verification mechanism at 15 minutes / time to track changes in ion current activity and dynamically adjust the regulation range.

[0061] If the growth status of edible fungi is determined to be weak, and it is predicted that the growth status of edible fungi will gradually weaken in the next 1-12 hours, then the multi-actuator coordinated regulation strategy is a weak and gradual decline coordinated regulation strategy.

[0062] It should be noted that the mildly slow-reduction synergistic control strategy involves: activating the zoned temperature control module, zoned humidity control module, and zoned atmosphere control module, adopting a moderate control mode (environmental parameter adjustment range of ±5%), and adjusting temperature, humidity, and CO2 concentration at a frequency of 5 minutes / time to suppress the expansion of parameter deviations; maintaining the current output of the adjustable spectral illumination module to avoid exacerbating growth weakening due to light fluctuations; activating the matrix EC value adjustment module to gradually adjust the matrix EC value to the suitable range; and setting a parameter verification mechanism at 10 minutes / time to focus on tracking the temporal changes in ion current activity and adjust the control strategy in a timely manner to alleviate the weakening of growth status.

[0063] The examples described in this invention are not limited to the specific embodiments listed above. The examples are merely illustrative to facilitate understanding of the invention and do not constitute a limitation on the scope of protection of this invention. Any modifications, equivalent substitutions, etc., made within the spirit and principles of this invention should be included within the scope of protection.

[0064] The above description is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the protection scope of the present invention.

Claims

1. A smart monitoring and precise control system for edible fungi growth, characterized in that, Includes the following modules: Multimodal in-situ sensing module: Used to receive information on the current batch of edible fungi in the target cultivation room, thereby calling up the corresponding physiological and metabolic baseline parameters and growth environment adaptation parameters of the edible fungi, and then simultaneously collecting multi-dimensional data on the ion flow source, physiological metabolism, three-dimensional environment and volatile organic compound characteristics of the edible fungi, generating multimodal sensing datasets for each batch of edible fungi. Linked Prediction Modeling Module: Based on the multimodal sensing dataset corresponding to edible fungi, it constructs a three-in-one linked prediction model of microenvironment parameters, physiological metabolic indicators, and fungal growth status. Through the built-in cross-scale coupling sub-model of mycelial electrical signals and metabolic status, it realizes cross-scale fusion of electrical signals and metabolic signals, analyzes and obtains the ion current activity index corresponding to edible fungi, and then analyzes the growth status of edible fungi and predicts the dynamic changes of microenvironment and growth status of edible fungi in the next 1-12 hours. Coupled Decision and Precise Control Module: Based on the analysis and prediction results of edible fungi and the early warning trigger signal output by the mycelial electrical signal-metabolic state cross-scale coupling sub-model, combined with the corresponding growth benchmark parameters of edible fungi, it determines the optimal combination of environmental parameters and multi-actuator collaborative control strategy for the target cultivation room.

2. The intelligent monitoring and precise control system for edible fungi growth according to claim 1, characterized in that, The specific process for calling the physiological metabolic baseline parameters and growth environment adaptation parameters corresponding to edible fungi is as follows: When the current batch of edible mushroom spawn enters the designated cultivation area of ​​the target mushroom house, the unique identification tag on the surface of the spawn is first read by an RFID reader deployed at the entrance of the mushroom house. The reader then analyzes the information to obtain the variety, inoculation time, and substrate formula of each batch of edible mushrooms. The reader then automatically sends a request to the associated database. The database is indexed by variety-growth stage-parameter classification. The database extracts the physiological and metabolic baseline parameters and growth environment adaptation parameters corresponding to different growth stages of edible mushrooms. The physiological and metabolic baseline parameters include the mycelial respiration rate threshold, nutrient consumption rate range, and VOCs characteristic spectrum. The growth environment adaptation parameters include the optimal range of temperature and humidity, CO2 concentration, light spectrum, and substrate EC value.

3. The intelligent monitoring and precise control system for edible fungi growth according to claim 2, characterized in that, The specific process for generating the multimodal sensing dataset corresponding to each batch of cultivated edible fungi is as follows: First, the mycelial electrophysiological microelectrode array is embedded in the matrix during the preparation or inoculation stage of the substrate. The in-situ physiological microsensor array is evenly inserted into the substrate of each shelf. The three-dimensional gridded environmental sensing module is arranged in the target cultivation room according to a 50cm×50cm×30cm spatial grid. The VOCs electronic nose sensing unit, the near-infrared spectroscopy non-invasive detection unit, and the high-definition visual acquisition unit are deployed in each cultivation section. Secondly, the mycelial electrophysiological microelectrode array collects the action potential signal generated by the transmembrane ion current of the mycelium and extracts the original characteristic parameters of the ion current; the in-situ physiological microsensor array collects the physiological metabolic characteristic parameters of the mycelium; the three-dimensional gridded environmental sensing module collects the three-dimensional environmental characteristic parameters in the mushroom house; the VOCs electronic nose sensing module collects the volatile organic compound fingerprint spectrum during the growth process of edible fungi; and the near-infrared spectroscopy non-invasive detection module non-invasively detects the metabolic parameters of soluble sugar, nitrogen and lignin degradation rate in the matrix. The high-definition visual acquisition module captures images of bacterial cell morphological changes; Finally, all collected data are format-converted and stored in a directory structure of variety-cultivation batch-data type to form a multimodal sensing dataset corresponding to each cultivation batch of edible fungi.

4. The intelligent monitoring and precise control system for edible fungi growth according to claim 3, characterized in that, The specific construction process of the integrated predictive model combining microenvironment parameters, physiological metabolic indicators, and bacterial growth status is as follows: A three-in-one linkage prediction model is constructed by adopting a fusion mode of mechanism modeling and attention mechanism neural network, which includes a dynamic evolution sub-model of microenvironment parameters, a physiological metabolic index response sub-model and a cell growth state prediction sub-model, and a built-in cross-scale coupling sub-model of mycelial electrical signal-metabolic state. Among them, the microenvironment parameter dynamic evolution sub-model is constructed based on the multi-physics field coupling theory and is used to simulate the spatial distribution and dynamic evolution law of microenvironment parameters; the physiological metabolic index response sub-model is constructed based on the edible fungus growth kinetic equation and is used to characterize the correlation response relationship between physiological metabolic parameters and the ion current source signal; the cell growth state prediction sub-model is constructed based on the improved attention mechanism neural network, with the comprehensive metabolic activity index output by the mycelial electrical signal-metabolic state cross-scale coupling sub-model as the core input, strengthening the dynamic attention to the ion current source signal, and realizing the accurate training and dynamic calibration of the model.

5. The intelligent monitoring and precise control system for edible fungi growth according to claim 4, characterized in that, The specific implementation process for achieving cross-scale fusion of electrical and metabolic signals is as follows: The mycelial transmembrane ion current action potential signal is collected by the mycelial electrophysiological microelectrode array of the multimodal in-situ sensing module, and the mycelial respiration rate, nutrient consumption rate and VOCs fingerprint spectrum are collected simultaneously by the in-situ physiological microsensor array and VOCs electronic nose sensing module. By using the timing synchronization unit of the coupled sub-model, the metabolic signal is calibrated at the millisecond level to eliminate misalignment based on the acquisition timing of the ion flow electrical signal. Wavelet denoising algorithm is used to process the electrical signal to remove interference. The metabolic signal drift is corrected and core features are extracted by the adaptive baseline calibration algorithm. The intrinsic correlation between the ion flow action potential signal, the ion flow source parameters and the mycelial respiration rate, nutrient consumption rate and VOC fingerprint spectrum is directly established.

6. The intelligent monitoring and precise control system for edible fungi growth according to claim 5, characterized in that, The analysis yielded the ion current activity index corresponding to the edible fungi. The specific analysis process is as follows: First, parameter extraction: using a cross-scale coupling sub-model of mycelial electrical signal-metabolic state, the intrinsic parameters of ion current in edible fungi were extracted. These intrinsic parameters include the half-width at half-maximum (WWHM) of the ion current peak. Ion current frequency drift coefficient ion current memory effect attenuation rate and the coefficient of variation of the ion burst interval The extracted ion current source parameters are preprocessed and normalized to the 0-1 range. Then, the ion current activity index is generated: using a linear mapping algorithm, the ion current activity index is generated based on the preprocessed ion current source parameters, as shown in the formula: The ion current activity index was obtained. .

7. The intelligent monitoring and precise control system for edible fungi growth according to claim 6, characterized in that, The analysis of the growth status of edible fungi is conducted in the following specific process: Ion current activity index, mycelial respiration rate, nutrient consumption rate, degradation rate of soluble sugars, nitrogen, lignin and VOCs fingerprints in the matrix were retrieved from the multimodal sensing dataset. If the ion current activity index is between 2.2 and 3.0, the mycelial respiration rate is between 0.8 and 1.2, the nutrient consumption rate is between 0.5 and 0.8, the degradation rate of soluble sugars in the substrate is 15%-20%, the degradation rate of nitrogen is 10%-15%, and the VOCs fingerprint spectrum is completely matched with the VOCs characteristic spectrum in the physiological metabolic baseline parameters, then the edible fungi are judged to be in normal growth status. If the ion current activity index is between 3.0 and 4.0, the mycelial respiration rate is between 1.2 and 1.5, the nutrient consumption rate is between 0.8 and 1.2, the degradation rate of soluble sugars in the substrate is 21%-28%, the degradation rate of nitrogen is 16%-22%, and the intensity of the characteristic peaks in the VOCs fingerprint spectrum that characterize vigorous mycelial growth is higher than the benchmark value, then the edible fungi are judged to be in a vigorous growth state. If the ion current activity index is between 1.0 and 2.2, the mycelial respiration rate is between 0.5 and 0.8, the nutrient consumption rate is between 0.3 and 0.5, the degradation rate of soluble sugars in the substrate is 8%-14%, the degradation rate of nitrogen is 6%-9%, and there are no abnormal characteristic peaks in the VOCs fingerprint spectrum, then the growth status of edible fungi is judged to be weak.

8. The intelligent monitoring and precise control system for edible fungi growth according to claim 7, characterized in that, The specific prediction process for forecasting the dynamic changes in the microenvironment and growth status of edible fungi over the next 1-12 hours is as follows: First, the ion flow source parameters, physiological metabolic parameters, microenvironment parameters, historical time series data of the past 24 hours, physiological metabolic baseline parameters and growth environment adaptation parameters of edible fungi from the multimodal sensing dataset are uniformly input into the three-in-one linkage prediction model, and the output is the evolution law of microenvironment parameters, the mapping relationship of physiological metabolic response, the temporal change trend of ion flow activity, and the coupling relationship law of environment-metabolism-cell growth. If all four intermediate results are stable, the microenvironment parameters evolve smoothly and the fluctuation range meets the requirements of the growth environment adaptation parameters, the physiological metabolic response is stable and the parameters do not deviate, the ion current activity is slow and the IFAI fluctuation range is ≤0.3, and the multi-parameter coupling is synergistic and there is no mismatch, then it is predicted that the growth status of edible fungi will remain normal in the next 1-12 hours. If the four intermediate results show positive evolution, the microenvironment parameters gradually tend to the high value of the growth environment adaptation range and the fluctuation range is ≤ ±0.3℃, ±2%RH, ±30ppm, the physiological metabolism shows a positive response and the parameters gradually approach the vigorous growth threshold, the ion current activity time sequence shows a continuous upward trend and IFAI increases by 0.1 to 0.2 per day, and the multi-parameter coupling shows vigorous growth characteristics without any imbalance, then it is predicted that the growth status of edible fungi will remain vigorous in the next 1-12 hours. If the four intermediate results show mild mismatch, with microenvironment parameters slightly deviating from the optimal range for growth environment with fluctuations of ±0.3~0.5℃, ±2%~3%RH, ±30~50ppm, weak physiological metabolic response with parameters deviating from the baseline threshold by 5%~10%, ion current activity showing a slow downward trend with IFAI decreasing by 0.05~0.1 per hour, and mild imbalance of multi-parameter coupling without serious mismatch, then the growth status of edible fungi is predicted to gradually weaken in the next 1-12 hours.

9. The intelligent monitoring and precise control system for edible fungi growth according to claim 8, characterized in that, The decision-making process for determining the optimal combination of environmental parameters for the target culture chamber is as follows: If the growth status of edible fungi is determined to be weak, and it is predicted that the growth status of edible fungi will remain normal for the next 1-12 hours, then the optimal combination of environmental parameters is the weak steady-state adaptation combination. If the growth status of edible fungi is determined to be weak, and it is predicted that the growth status of edible fungi will continue to be vigorous within the next 1-12 hours, then the optimal combination of environmental parameters is the weak rising state adaptation combination. If the growth status of edible fungi is determined to be weak, and it is predicted that the growth status of edible fungi will gradually weaken in the next 1-12 hours, then the optimal combination of environmental parameters is the weak and slow-decreasing adaptation combination.

10. The intelligent monitoring and precise control system for edible fungi growth according to claim 9, characterized in that, The decision-making process for the multi-actuator coordinated control strategy corresponding to the target culture chamber is as follows: If the growth status of edible fungi is determined to be weak, and the growth status of edible fungi is predicted to remain normal for the next 1-12 hours, then the multi-actuator coordinated regulation strategy is a weak steady-state coordinated regulation strategy. If the growth status of edible fungi is determined to be weak, and the growth status of edible fungi is predicted to remain vigorous within the next 1-12 hours, then the multi-actuator coordinated regulation strategy is a weak-to-rising-state coordinated regulation strategy. If the growth status of edible fungi is determined to be weak, and it is predicted that the growth status of edible fungi will gradually weaken in the next 1-12 hours, then the multi-actuator coordinated regulation strategy is a weak and gradual decline coordinated regulation strategy.

Images