Mushroom factory intelligent monitoring system based on Internet of Things

Abstract

The invention relates to the technical field of cross fusion of the Internet of Things technology and intelligent agriculture, and discloses a mushroom factory intelligent monitoring system based on the Internet of Things, and the system is internally provided with modules, a mushroom factory sensor monitoring module employs an agricultural sensor to monitor mushroom growth data, environment data and state data in real time, and the agricultural sensor is used for monitoring the mushroom growth data. The mushroom growth model online updating module updates mushroom growth model parameters and sensor calibration data online, the factory intelligent analysis module calculates a mushroom yield and weight prediction value, a disease risk index and a mushroom growth stage matching degree, and the fault alarm module performs three-level early warning prompt. The offline emergency control module can perform network offline switching when network interruption is detected, the personalized parameter configuration module formulates personalized environment control curves of mushrooms of different varieties for users, the mushroom yield prediction module performs yield cycle management on the mushrooms, and the user interaction module provides visual real-time data service for the users. And receiving user feedback information.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Things technology and smart agriculture cross-fusion technology, in particular to a mushroom factory intelligent monitoring system based on Internet of Things. BACKGROUND

[0002] With the development of facility agriculture, the demand for intelligent management and control of edible fungus factory production is increasingly urgent. However, the existing technology still has significant bottlenecks: first, the visual recognition problem caused by the dense layer of mushroom house shelves and the small difference in color and luster of mushroom cap, traditional RGB cameras are easily disturbed by light and difficult to penetrate the obstruction, resulting in high misjudgment rate of browning and mixed bacteria; second, the environmental control relies on general experience curve, lacking dynamic adaptation ability to different varieties characteristics and environmental fluctuations, resulting in waste of resources and quality fluctuations; third, the sensor drift is serious in high-humidity disinfection environment, the service life is short, and there is lack of local bottom control when the network is interrupted, the device reliability is low; fourth, data collection and decision-making are disconnected, single threshold alarm mode cannot effectively early warn multi-factor coupling risk, and the data value is not fully tapped. In view of the above problems, the present application provides a mushroom factory intelligent monitoring system based on Internet of Things. SUMMARY

[0003] (I) Technical problems solved In view of the shortcomings of the prior art, the present application provides a mushroom factory intelligent monitoring system based on Internet of Things, which has the advantages of high-precision visual recognition of three-dimensional point cloud and multi-spectrum fusion, dynamic growth modeling of multi-factor coupling, yield prediction of accurate production and sales matching, disease risk prevention and control of hierarchical response, and offline emergency control, solving the problems of inaccurate visual recognition, extensive environmental control, low device reliability and insufficient data utilization in edible fungus factory production.

[0004] To achieve the above purpose, the present application provides the following technical scheme: a mushroom factory intelligent monitoring system based on Internet of Things, comprising a mushroom factory sensor monitoring module, a mushroom growth model online updating module, an intelligent factory analysis module, a fault alarm module, an offline emergency control module, a personalized parameter configuration module, a yield prediction module and a user interaction module; The mushroom factory sensor monitoring module uses agricultural grade sensors to monitor mushroom growth data, environmental data and state data in real time, and transmits the data to the intelligent factory analysis module through the network; The mushroom growth model online updating module is provided with a multi-factor coupling growth model, and the mushroom growth model parameters and sensor calibration data are updated online through the connection of cloud platform or central database, and the data is transmitted to the intelligent factory analysis module through the network; The intelligent factory analysis module calculates the mushroom yield prediction value , disease risk index Matching degree with mushroom growth stage ; The fault alarm module is based on disease risk index The sensor itself health state data, three-level early warning and targeted measures are taken; The offline emergency control module automatically switches to the locally pre-stored control strategy library to perform closed-loop control when detecting network interruption; The personalized parameter configuration module is based on the matching degree of the multi-factor coupled growth model and the mushroom growth stage , provides personalized environmental control curves for different varieties of mushrooms for users; The mushroom yield prediction module is based on the mushroom yield prediction value , the yield cycle management is carried out; The user interaction module provides visual real-time data service for users through Web browser or mobile App, and receives user feedback information.

[0005] Preferably, the mushroom factory sensor monitoring module includes a mushroom state data monitoring unit, a factory environment data monitoring unit and a mushroom growth data monitoring unit.

[0006] Preferably, the mushroom state data monitoring unit collects visual data including images, video streams, 3D point cloud data, multispectral images, near-infrared images, high-resolution texture data and color LAB value quantitative data through high-humidity industrial cameras and edge computing visual AI boxes, for identifying cap state, browning, contamination and signs of insect pests.

[0007] Preferably, the factory environment data monitoring unit collects environmental sensor data including temperature, humidity, carbon dioxide, oxygen, VOC, light intensity / quality, air pressure and substrate moisture content through high-humidity, corrosion-resistant industrial sensor arrays deployed in each area of the mushroom house, for real-time monitoring of all-round environmental factors closely related to mushroom growth and quality formation.

[0008] Preferably, the mushroom growth data monitoring unit collects growth stage and yield estimation data including growth stage identification, biomass fresh weight estimation, cap diameter, mycelium growth rate and fruiting body number through visual AI analysis, embedded weighing sensors and RFID batch tags, for building personalized mushroom parameter configuration and providing core input for yield prediction and growth cycle optimization.

[0009] Preferably, the mushroom growth model online updating module is provided with a multi-factor coupled growth model, specifically including variety-specific sub-model, environmental response sub-model and stress adaptation sub-model.

[0010] Preferably, the factory intelligent analysis module calculates the mushroom yield prediction value based on the monitoring and updating data , and the calculation formula is

[0011] In the formula, represents the mushroom yield prediction value, i.e. the fresh weight of biomass per unit area, represents the number of bags per unit area, represents the average radius of the cap, represents the cap density, represents the average cross-sectional area of the stem, represents the average height of the stem, represents the stem density, represents the environmental adaptation coefficient.

[0012] Preferably, the factory intelligent analysis module calculates the disease risk index based on the monitoring and updating data , and the calculation formula is

[0013] In the formula, represents the disease risk index, i.e. the risk probability of bacterial spot disease / mold contamination, represents the maximum humidity drop within 24 hours, represents the maximum temperature rise within 24 hours, represents the actual VOC concentration, represents the actual oxygen concentration, , , , respectively represent the humidity mutation risk coefficient, the temperature mutation risk coefficient, the VOC concentration risk coefficient and the hypoxic environment risk coefficient.

[0014] Preferably, the factory intelligent analysis module calculates the mushroom growth stage matching degree based on the monitoring and updating data , and the calculation formula is

[0015] In the formula, represents the growth stage matching degree, and α, β, γ respectively represent the accumulated temperature matching weight coefficient, the cap diameter matching weight coefficient and the mycelium growth speed matching weight coefficient, represents the actual accumulated temperature, represents the target stage standard accumulated temperature, represents the actual average cap diameter, represents the target stage standard cap diameter, represents the actual mycelium growth speed, This indicates the standard mycelial growth rate at the target stage.

[0016] Preferably, the fault alarm module is based on the disease risk index. A three-level early warning system is implemented based on the sensor's own health status data. The specific determination method is as follows: (1) When the system detects a sudden drop in humidity and a rise in temperature, and the disease risk index is high When the threshold is exceeded, a high-risk alarm for bacterial spot disease is automatically triggered, which is determined to be a level one alarm. The local audible and visual alarm is automatically triggered, and intervention suggestions are generated to increase the humidifier RH to 85% and reduce the fan speed to reduce temperature fluctuations. (2) When the drift rate of the carbon dioxide and VOC sensor is >15% or the data interruption lasts for more than 1 hour, a medium-level equipment fault alarm will be triggered, which will be judged as a level 2 alarm. The local audible and visual alarm will be automatically triggered, and the sensor data will be marked as unreliable on the system interface. At the same time, the system will automatically call the latest calibration parameters in the mushroom growth model online update module to perform soft correction on the sensor data, and prompt the administrator to perform on-site calibration or replacement within 3 days. (3) Matching degree during mushroom growth stages When the concentration remains below 0.6 and the environmental data matches the preset curve, a primary warning for abnormal growth will be triggered, which will be classified as a level 3 alarm. The alarm will only be displayed as an orange flashing notification and recorded in the system management interface. A diagnostic suggestion will be generated indicating that growth lag has been detected and that the activity of the microbial strain or the substrate nutrients should be checked, providing management personnel with a basis for early intervention decisions.

[0017] Compared with existing technologies, this invention provides an IoT-based intelligent monitoring system for mushroom factories, which has the following advantages: 1. This invention effectively addresses the identification challenges caused by severe shading of the shelves in the mushroom house and small differences in the color of the mushroom caps by employing 3D point cloud and multispectral imaging technologies. The 3D point cloud provides three-dimensional spatial information to penetrate the shading, while the multispectral image can capture spectral data beyond the visible light range to distinguish minute color differences. The combination of the two significantly improves the online identification accuracy of the system and can effectively avoid the error of misjudging "browning" as "miscellaneous fungi".

[0018] 2. This invention incorporates a multi-factor coupled growth model within the online update module of the mushroom growth model. Through the collaborative work of three sub-models—variety specificity, environmental response, and stress adaptation—the system achieves a leap from general experience to variety customization, from static parameters to dynamic optimization, and from ideal growth to stress-resistant management. This enables the model to accurately adapt to the specific needs of different varieties and production areas, and to possess the self-learning ability to cope with environmental fluctuations.

[0019] 3. This invention predicts mushroom yield by calculating the predicted value. , when the actual demand yield of the manufacturer is within 90%-110% of the predicted value , the system can determine that it is "expected to match" and maintain the existing management strategy; when the demand yield is higher than 110% of the predicted value , the system will prompt in the personalized parameter configuration module that "the target yield is too high, it is recommended to optimize the environmental parameters or extend the growth cycle"; when the demand yield is lower than 90% of the predicted value , the system will prompt that "there is a risk of overcapacity, and production plans can be adjusted", and finally achieve the effect of precise matching of production and sales and resource input according to demand.

[0020] 4、The present application calculates the disease risk index above, which is used as the core evaluation standard for starting the hierarchical response of the fault alarm module; when the disease risk index is in the interval of 0.5-0.7, the system performs a yellow warning in the user interaction module and prompts the administrator to pay attention to the environmental trend; when the disease risk index is in the interval of 0.7-0.85, the system triggers a level one alarm and automatically executes the preset emergency control strategy (such as linking humidification and ventilation equipment); when the disease risk index is above 0.85, in addition to the above measures, the system will forcibly pop up an emergency treatment process in the user interface and suggest starting expert consultation to build a complete set of automated and hierarchical management processes from risk warning to emergency disposal; by calculating the mushroom growth stage matching degree , the personalized parameter configuration module provides accurate stage switching judgment basis, which is substituted into the environmental control curve to automatically switch and optimize the environmental parameters, so as to realize smooth and accurate transition of the mushroom growth stage, thereby avoiding environmental stress caused by stage misjudgment. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 The present application is a system flowchart. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0023] Please refer to Figure 1, the mushroom factory intelligent monitoring system based on the Internet of Things, comprising a mushroom factory sensor monitoring module, a mushroom growth model online updating module, a factory intelligent analysis module, a fault alarm module, an offline emergency control module, a personalized parameter configuration module, a yield prediction module and a user interaction module; The mushroom factory sensor monitoring module uses agricultural level sensors to monitor mushroom growth data, environmental data and state data in real time, and transmits the data to the factory intelligent analysis module through the network; The mushroom growth model online updating module is provided with a multi-factor coupled growth model, and the mushroom growth model parameters and sensor calibration data are updated online through the connection of a cloud platform or a central database, so that the model can adapt to new varieties and environmental changes, and the data is transmitted to the factory intelligent analysis module through the network; The factory intelligent analysis module calculates the mushroom yield prediction value , disease risk index and mushroom growth stage matching degree , providing decision basis for subsequent module analysis; The fault alarm module performs three-level early warning according to the disease risk index and the sensor health state data; when the system identifies that the humidity drops suddenly and the temperature rises, and the disease risk index exceeds the threshold value, it will trigger the high-risk alarm of bacterial spot disease, instead of waiting for a single parameter to exceed the limit, which not only discovers the problem (alarm), but also takes immediate action (linkage control) through the preset strategy to minimize the loss; The offline emergency control module automatically switches to the local pre-stored control strategy library to perform closed-loop control when detecting network interruption (public network or local area network), so as to ensure that the mushroom house environment is maintained at the latest optimized setting value or safety baseline before the network interruption, and provides reliable offline backup capability; The personalized parameter configuration module provides users with individualized environmental control curves (temperature, humidity, light, and air coupling curves) for different varieties of mushrooms according to the multi-factor coupled growth model and the mushroom growth stage matching degree ; The mushroom yield prediction module manages the yield cycle according to the mushroom yield prediction value ; The user interaction module provides visual real-time data services for users through a Web browser or a mobile App, and receives user feedback information, which is the centralized entrance for the entire system to interact with management personnel.

[0024] The mushroom factory sensor monitoring module comprises a mushroom state data monitoring unit, a factory environment data monitoring unit and a mushroom growth data monitoring unit.

[0025] The mushroom state data monitoring unit collects visual data including images, video streams, 3D point cloud data, multispectral images, near-infrared images, high-resolution texture data, and color LAB value quantitative data through a high-humidity-resistant industrial camera and an edge computing visual AI box, for identifying cap state, browning, contamination, and signs of insect damage. The advantage is that 3D point cloud and multispectral imaging technology are used to effectively solve the identification problem caused by serious shelf blocking in the mushroom house and small color difference of the cap, where 3D point cloud provides three-dimensional spatial information to penetrate the blocking, and multispectral images can capture spectral data beyond the visible light range to distinguish small color differences. The combination of the two greatly improves the online identification accuracy of the system and effectively avoids the error of misjudging "browning" as "contamination".

[0026] The factory environment data monitoring unit collects environmental sensor data including temperature, humidity, carbon dioxide, oxygen, VOC, light intensity / quality, air pressure, and substrate moisture content through a high-humidity-resistant, corrosion-resistant industrial sensor array deployed in each area of the mushroom house, for real-time monitoring of all-round environmental factors closely related to mushroom growth and quality formation, providing a data foundation for the precise control of subsequent modules. The advantage is that online monitoring of the most critical carbon dioxide, oxygen, and VOC for edible fungi is performed, and light, substrate moisture, and other key parameters are added to solve the problem of reduced monitoring dimensions in traditional systems. The sensors with high-humidity-resistant and corrosion-resistant features can cope with the high-humidity and disinfection environment in the mushroom factory.

[0027] The mushroom growth data monitoring unit collects growth stage and yield estimation data including growth stage identification, biomass fresh weight estimation, cap diameter, mycelium growth rate, and fruiting body number through visual AI analysis, embedded weighing sensors, and RFID batch tags, for building personalized parameter configurations for mushrooms and providing core inputs for yield prediction and growth cycle optimization. The advantage is that visual data (growth stage, cap size) and physical data (weight) are collected to provide direct data support for personalized growth models, making the collected parameters multi-sourced and solving the problem of one-size-fits-all in traditional data.

[0028] The mushroom growth model online updating module has a multi-factor coupled growth model, which specifically includes variety-specific sub-models (responsible for storing genotype-phenotype correlation parameters of more than 20 varieties such as shiitake mushrooms, oyster mushrooms, and golden needle mushrooms, such as 400 for oyster mushrooms and 320 , new varieties can be quickly imported parameters), environmental response sub-model (responsible for quantifying the influence coefficient of temperature, humidity, carbon dioxide, and light on growth rate, such as RH every 5% reduction, 1.2d extension of pleurotus swelling period), and dynamic matching of environmental changes) and stress adaptation sub-model (responsible for recording the growth compensation mechanism under stress such as bacterial contamination and sudden humidity drop, such as after VOC short-term exceeds the standard, the oxygen concentration needs to be increased to 21% to restore growth).

[0029] The advantages are: through the cooperative work of the above three sub-models of variety specificity, environmental response and stress adaptation, the system realizes the leap from general experience to variety customization, from static parameters to dynamic optimization, and from ideal growth to stress management, so that the model can accurately adapt to the specific needs of different varieties and different locations, and has the ability to learn from environmental fluctuations.

[0030] The factory intelligent analysis module calculates the mushroom yield prediction value based on monitoring and updating data , the calculation formula is:

[0031] In the formula, represents the mushroom yield prediction value, i.e. the fresh weight of biomass per unit area (kg / m²), including the weight of the cap and the stem, represents the number of mushroom bags per unit area (pieces / m², such as 6 pieces / m², system default), represents the average radius of the cap cm, calculated by visual AI box 3D point cloud data, represents the cap density (g / cm³, Pleurotus 1.15, Lentinus 1.22, provided by the mushroom growth model online updating module), represents the average cross-sectional area of the stem (cm², obtained by multi-spectral image edge detection), represents the average height of the stem cm, measured by 3D point cloud data, represents the stem density (g / cm³, Pleurotus 0.98, Lentinus 1.05, provided by the model updating module), represents the environmental adaptation coefficient, ranging from 0.8 to 1.0, data from the factory environment data monitoring unit.

[0032] The advantages are: through the calculation of the mushroom yield prediction value , used in the yield prediction module, when the actual demand yield of the factory is within the 90%-110% range of the prediction value , the system can determine that it is "expected to match" and maintain the existing management strategy; when the demand yield is higher than 110%, the system will prompt "target yield is high, suggest optimizing environmental parameters or extending growth cycle" in the personalized parameter configuration module; when the demand yield is lower than When the disease risk index reaches 90%, the system will prompt "there is a risk of overcapacity, and the production plan can be adjusted", and finally achieve the effect of precise matching of production and sales and resource input on demand.

[0033] The factory intelligent analysis module calculates the disease risk index based on monitoring and updating data The calculation formula is:

[0034] In the formula, represents the disease risk index, i.e. the risk probability of bacterial spot disease / mold contamination (0-1, When ≥0.7, trigger high alert, represents the maximum humidity drop in 24 hours (%, represents the maximum temperature rise in 24 hours (℃, represents the actual concentration of VOC (ppm, collected by the factory environment data monitoring unit, standard threshold 50 ppm), represents the actual oxygen concentration (%, , , , respectively represent the humidity mutation risk coefficient, the temperature mutation risk coefficient, the VOC concentration risk coefficient and the hypoxic environment risk coefficient (0.008 / 0.012 / 0.02 / 0.05 for bacterial spot disease, 0.005 / 0.006 / 0.03 / 0.08 for mold contamination, provided by the mushroom growth model online updating module).

[0035] The advantage is: through the calculation of the above disease risk index , it is used as the core evaluation standard for the startup of the fault alarm module to respond to the disease risk index When the disease risk index is in the range of 0.5-0.7, the system performs yellow warning in the user interaction module, prompting the administrator to pay attention to the environmental trend; when the disease risk index is in the range of 0.7-0.85, the system triggers a first-level alarm and automatically executes the preset emergency control strategy (such as linkage humidification and ventilation equipment); when the disease risk index is above 0.85, in addition to the above measures, the system will forcibly pop up an emergency treatment process in the user interface and suggest starting expert consultation to build a complete set of automated and hierarchical management process from risk warning to emergency disposal.

[0036] The factory intelligent analysis module calculates the mushroom growth stage matching degree based on monitoring and updating data The calculation formula is:

[0037] In the formula, represents the growth stage matching degree (0-1, ≥0.9, it is determined to enter the target stage), α, β, γ respectively represent the accumulated temperature matching weight coefficient, the cap diameter matching weight coefficient and the mycelium growth speed matching weight coefficient (mycelium stage 0.6 / 0.2 / 0.2, expansion stage 0.3 / 0.5 / 0.2, provided by the mushroom growth model online updating module), represents the actual accumulated temperature (°C), , , from the factory environment data monitoring unit, 5℃ for Pleurotus ostreatus and 3℃ for Flammulina velutipes, represents the target stage standard accumulated temperature (°C), such as 200 for Pleurotus ostreatus expansion stage, provided by the mushroom growth model online updating module), represents the actual average cap diameter (cm), collected by visual AI box, represents the target stage standard cap diameter (cm), such as 2.5 cm for Pleurotus ostreatus expansion stage, provided by the mushroom growth model online updating module), represents the actual mycelium growth speed (mm / d, calculated by RFID batch tag record growth period and mycelium packet radial growth), represents the target stage standard mycelium growth speed (mm / d), such as 3.2 mm / d for Pleurotus ostreatus mycelium stage, provided by the mushroom growth model online updating module).

[0038] The advantage is that the above mushroom growth stage matching degree is calculated to provide accurate stage switching judgment basis for the individualized parameter configuration module, which is substituted into the environmental control curve to automatically switch and optimize the environmental parameters, so as to realize smooth and accurate transition of mushroom growth stage, thereby avoiding environmental stress caused by stage misjudgment.

[0039] The fault alarm module performs three-level early warning according to the disease risk index and sensor self-health state data (such as drift rate Drift, data interruption time), specifically: (1) When the system identifies that the humidity suddenly drops and the temperature rises, and the disease risk index exceeds the threshold value (such as When the drift rate of the key sensors (such as carbon dioxide, VOC) is greater than 15% or the data interruption duration lasts more than 1 hour, a medium-level device fault warning (secondary warning) will be triggered, triggering local audible and light alarms (70dB buzzer + yellow flashing) and APP pop-up windows, and marking the sensor data as untrusted on the system interface. At the same time, the system will automatically call the latest calibration parameters in the mushroom growth model online update module to soft correct the sensor data, and prompt the administrator to perform on-site calibration or replacement within 3 days. When the drift rate of the key sensors (such as carbon dioxide, VOC) is greater than 15% or the data interruption duration lasts more than 1 hour, a medium-level device fault warning (secondary warning) will be triggered, triggering local audible and light alarms (70dB buzzer + yellow flashing) and APP pop-up windows, and marking the sensor data as untrusted on the system interface. At the same time, the system will automatically call the latest calibration parameters in the mushroom growth model online update module to soft correct the sensor data, and prompt the administrator to perform on-site calibration or replacement within 3 days. When the growth phase matching degree is continuously less than 0.6, and the environmental data matches the preset curve, a primary growth anomaly warning (tertiary warning) will be triggered, with only orange flashing prompts and log records on the system management interface, and a diagnostic suggestion of detecting growth lag and suggesting checking strain viability or substrate nutrients is generated to provide early intervention decision-making basis for management personnel.

[0040] Advantages: Through the establishment of the above-mentioned three-level warning mechanisms for business risk, device failure, and growth anomaly, a comprehensive and dead-angle-free monitoring network covering "biological growth - hardware state - data quality" is formed, realizing the upgrade from passive response to active warning and from single-point alarm to systematic diagnosis.

[0041] Summary: The present application builds a mushroom factory sensor monitoring module, a mushroom growth model online update module, a factory intelligent analysis module, a fault alarm module, an offline emergency control module, a personalized parameter configuration module, a yield prediction module, and a user interaction module. In the sensor monitoring module, accurate and reliable collection of multi-source heterogeneous data is realized. In the growth model online update module, continuous evolution and self-adaptation of core algorithms are realized. In the factory intelligent analysis module, accurate quantification and prediction of yield, disease, and growth phase are realized. In the fault alarm module, multi-level identification and automatic disposal of risks are realized. In the offline emergency control module, control is uninterrupted in extreme network conditions. In the personalized parameter configuration module, precise cultivation from "one-size-fits-all" to "one-product-one-strategy" is realized. In the yield prediction module, data is connected from production management to business decision-making. In the user interaction module, unified management and visual management of all functions are realized.

[0042] While embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely exemplary of the principles and application of the present application. Numerous modifications and adaptions can be effected without departing from the spirit and scope of the present application, which is not limited to the exact construction and arrangement described. It is intended, therefore, to cover all modifications and adaptions that fall within the scope of the claims and their equivalents.

Claims

1. An intelligent monitoring system for mushroom factory based on Internet of Things, characterized in that, The mushroom factory sensor monitoring module, the mushroom growth model online updating module, the factory intelligent analysis module, the fault alarm module, the offline emergency control module, the personalized parameter configuration module, the yield prediction module and the user interaction module are included. The mushroom factory sensor monitoring module adopts agricultural level sensors to monitor mushroom growth data, environmental data and state data in real time, and transmits the data to the factory intelligent analysis module through the network. The mushroom growth model online updating module is provided with a multi-factor coupled growth model, and the mushroom growth model parameters and sensor calibration data are updated online through the connection of a cloud platform or a central database, and the data is transmitted to the factory intelligent analysis module through the network. The factory intelligence analysis module calculates a mushroom yield prediction value based on monitoring and updating data , disease risk index , and mushroom growth stage matching degree ; The fault alarm module conducts three-level early warning and takes targeted measures according to the disease risk index and sensor self-health state data. When the offline emergency control module detects network interruption, it automatically switches to the locally pre-stored control strategy library to execute closed-loop control. The personalized parameter configuration module matches the multi-factor coupling growth model with the mushroom growth stage , and provides the user with a personalized environment control curve of different varieties of mushrooms. The mushroom yield prediction module performs yield cycle management according to the mushroom yield weight prediction value . The user interaction module provides visual real-time data service for users through Web browser or mobile App, and receives user feedback information. 2.The IoT-based intelligent monitoring system for mushroom factory according to claim 1, characterized in that: The mushroom factory sensor monitoring module includes a mushroom state data monitoring unit, a factory environment data monitoring unit and a mushroom growth data monitoring unit. 3.The IoT-based intelligent monitoring system for mushroom factory according to claim 2, characterized in that: The mushroom state data monitoring unit collects visual data including images, video streams, 3D point cloud data, multispectral images, near-infrared images, high-resolution texture data and color LAB value quantitative data through high-humidity industrial cameras and edge computing visual AI boxes, for identifying cap state, browning, contamination and signs of insect pests. 4.The IoT-based intelligent monitoring system for mushroom factory according to claim 2, characterized in that: The factory environment data monitoring unit collects environmental sensor data including temperature, humidity, carbon dioxide, oxygen, VOC, light intensity / quality, air pressure and substrate moisture content through high-humidity, corrosion-resistant industrial sensors deployed in each area of the mushroom house, for real-time monitoring of all-round environmental factors closely related to mushroom growth and quality formation. 5.The IoT-based intelligent monitoring system for mushroom factory according to claim 2, characterized in that: The mushroom growth data monitoring unit collects growth stage and yield estimation data including growth stage identification, fresh weight estimation value, cap diameter, mycelium growth rate and fruiting body number through visual AI analysis, embedded weighing sensors and RFID batch tags, for building mushroom personalized parameter configuration and providing core input for yield prediction and growth cycle optimization. 6.The IoT-based intelligent monitoring system for mushroom factory according to claim 1, wherein: The mushroom growth model online updating module is provided with a multi-factor coupled growth model, including variety-specific sub-model, environmental response sub-model and stress adaptation sub-model. 7.The IoT-based intelligent mushroom factory monitoring system according to claim 1, wherein: The factory intelligence analysis module calculates the mushroom yield prediction value based on the monitoring and updating data The calculation formula is: ； In the formula, represents the predicted value of mushroom yield, i.e. fresh weight of biomass per unit area, represents the number of bags per unit area, represents the average radius of the cap, represents the cap density, represents the average cross-sectional area of the stem, represents the average height of the stem, represents the stem density, represents the environmental adaptation coefficient. 8.The IoT-based intelligent mushroom factory monitoring system according to claim 1, characterized in that: The factory intelligence analysis module calculates the disease risk index based on the monitoring and updating data The calculation formula is: ； In the formula, represents the disease risk index, i.e. the bacterial spot disease / mold contamination risk probability, represents the maximum humidity drop within 24 hours, represents the maximum temperature rise within 24 hours, represents the actual VOC concentration, represents the actual oxygen concentration, , , , respectively represent the humidity mutation risk coefficient, the temperature mutation risk coefficient, the VOC concentration risk coefficient and the hypoxic environment risk coefficient. 9.The IoT-based intelligent mushroom factory monitoring system according to claim 1, wherein: The factory intelligent analysis module calculates the mushroom growth stage matching degree based on the monitoring and updating data The calculation formula is: ； In the formula, represents the growth stage matching degree, and α, β, and γ respectively represent the accumulated temperature matching weight coefficient, the cap diameter matching weight coefficient, and the mycelium growth speed matching weight coefficient, represents the actual accumulated temperature, represents the target stage standard accumulated temperature, represents the actual cap average diameter, represents the target stage standard cap diameter, represents the actual mycelium growth speed, represents the target stage standard mycelium growth speed. 10.The IoT-based intelligent mushroom factory monitoring system according to claim 1, characterized in that: The fault alarm module determines the disease risk index according to the disease risk index The third level early warning is performed according to the sensor self-health state data, and the specific determination method is: (1) When the system identifies a sudden drop in humidity + temperature rise, and the disease risk index exceeds the threshold, automatically trigger the bacterial spot disease high-risk alarm, determine it as a first-level alarm, automatically trigger the local audible and light alarm, and generate an intervention suggestion to increase the humidifier RH to 85% and reduce the fan speed to reduce temperature fluctuations; (2) When the drift rate of the carbon dioxide and VOC sensors is greater than 15% or the data interruption duration exceeds 1 hour, a medium-level device fault alarm will be triggered, determining a secondary alarm, automatically triggering a local audible and visual alarm, marking the sensor data as unreliable on the system interface, and at the same time, the system will automatically call the latest calibration parameters in the mushroom growth model online updating module to soft correct the sensor data, and prompt the administrator to perform on-site calibration or replacement within 3 days. (3) When the mushroom growth stage matching degree When the matching degree is less than 0.6 and the environmental data matches the preset curve, the growth anomaly preliminary warning is triggered, the third level warning is determined, the orange flashing prompt and log record are only performed in the system management interface, and the detection of growth lag is generated. The diagnostic suggestion of checking the strain activity or substrate nutrient is proposed to provide the early intervention decision basis for the management personnel.

Images