Intelligent edible mushroom incubator based on artificial intelligence Internet of Things

Abstract

The invention belongs to the field of agricultural engineering, and particularly relates to an intelligent edible mushroom incubator based on an artificial intelligence internet of things. Comprising an edible fungus culture box body, and an edible fungus growth frame module, a spray atomization module, a carbon dioxide module, a fresh air circulation module, a visual detection module, a growth light module, a temperature sensing module and an intelligent disinfection module which are arranged on the edible fungus culture box body in an assembling manner, the upper surface, the lower surface, the left surface, the right surface and the rear surface of the edible mushroom incubator body are provided with heat preservation interlayers, and the front side surface is of a double-open cold storage door structure. By means of the Internet of Things technology, the artificial intelligence technology and the like, real-time monitoring of the growth environment of the edible mushrooms and accurate mastering of the growth condition of the edible mushrooms are achieved, so that the production efficiency and yield of the edible mushrooms are improved, and meanwhile the production cost and risk are reduced.

Description

Technical Field

[0001] The present invention belongs to the field of agricultural engineering, and specifically relates to an intelligent edible fungus cultivation box based on artificial intelligence Internet of Things. Background Art

[0002] Edible fungi are a kind of green food, rich in a variety of nutrients, such as protein, minerals and vitamins, which can enhance the human immune system. Some edible fungi also have medicinal value and can be used to prevent and treat certain diseases. Therefore, the demand for edible fungi products in the world food market is increasing. Although there are many types of edible fungi and they vary in size, they are all composed of mycelium living inside the matrix and fruiting bodies growing on the surface of the matrix. Mycelium is the nutritional organ of edible fungi. It is a mesh body composed of many branched hyphae. Its main function is to decompose the matrix, absorb and transport nutrients. The fruiting body is interwoven with differentiated hyphae and is the edible part of edible fungi. In the production of edible fungi, pests and diseases may occur, causing the yield and quality of edible fungi to decline. Diseases can be divided into infectious diseases and physiological diseases. Infectious diseases are mainly divided into four categories according to pathogens: fungi, bacteria, viruses and nematodes. The main cause of infectious diseases and physiological diseases is improper management. For example, wilt disease is usually caused by too dense mushroom production, resulting in insufficient nutrition or too high carbon dioxide concentration; white rot is usually caused by high temperature and poor ventilation. Pests generally include insects, nematodes, mites and mollusks. Mites are tiny spider-like pests that are extremely harmful in the production of edible fungi. Common species include tyrophagous mites and tarsoneme mites. Mites mainly harm the mycelium with nymphs or adults. They bite off the mycelium and eat it, or bore the fruiting body into a cavity. Cooked pigs and cattle are very attractive to mites, especially when the temperature is high in the afternoon, the mites are particularly active and will gather on the bones in large numbers. At this time, take out the bones and soak them in boiling water to kill the mites, and the bones can still be reused. Continuous trapping for a week can greatly reduce the insect population. The prevention and control of edible fungi diseases and pests must adhere to the principle of prevention as the main focus and comprehensive prevention and control, mainly prevention.

[0003] Edible fungi are heterotrophic organisms that cannot produce nutrients themselves and need to obtain nutrients from the outside world to grow, develop and reproduce. The nutrients required by edible fungi include carbon sources, nitrogen sources, inorganic salts, etc. Carbon and nitrogen have a great influence on the nutritional growth of edible fungi, and timely supplementation of the required elements is the key to high yields of edible fungi.

[0004] Suitable environmental conditions such as temperature, humidity, air and light are the basic conditions for the growth of edible fungi. Different edible fungi have different requirements for the environment, and the same edible fungi have different requirements for environmental conditions during the growth process. Temperature is an important environmental parameter that affects edible fungi. Each edible fungi requires a certain optimum, minimum and maximum growth temperature; edible fungi have different requirements for air humidity at different growth stages, generally low at the beginning and high at the end; all edible fungi are aerobic fungi, and edible fungi are sensitive to CO2 during the fruiting body differentiation stage. Sufficient oxygen can promote the differentiation and growth of fruiting bodies. Excessive accumulation of CO2 makes it difficult for fruiting bodies to form and grow slowly; light is not required during the mycelial growth stage, but certain scattered stimulation is required during the fruiting body differentiation and development stage of some edible fungi. Since the key environmental parameters that affect the growth of edible fungi can be monitored and controlled, indoor industrial cultivation of edible fungi has become an alternative production method to natural cultivation. By monitoring and adjusting the key growth environment parameters of edible fungi, the yield and quality of edible fungi can be improved. Indoor industrial cultivation can create an adjustable environment for edible fungi, so that they can obtain the predetermined results of high yield, high quality and low production cost. Proper environmental control also helps to detect diseases and minimize their impact. However, since environmental parameters have a strong nonlinear relationship and exhibit dynamic characteristics, the problem to be solved is complicated. In addition, factors such as the climate in the incubator, the stage of the growth cycle, and the external weather will also impose certain restrictions on the cultivation. Although the application of incubators has taken shape, there are still many problems in the application technology. Due to the insufficient intelligence of the incubator, the efficiency and yield of the incubator depend on the engineering and technical level of the technicians, which makes the efficiency and yield of crops grown in different incubators vary greatly. The edible fungus production process is usually based on a complete program, and the climate parameters are selected from a specific range. A common way is to adjust the growth environment regularly according to the parameters preset by the technicians, and the technicians need to correct the climate parameters in real time during the growth of the edible fungi. This method is highly dependent on the engineering and technical level of a single technician. There is currently no mature theory to guide how to adjust and maintain the optimal growth environment for edible fungi at each stage. In addition, manually adjusting the climate of the edible fungus growth environment is a time-consuming process and there is a potential risk of human error. At the same time, the traditional cultivation method based on manual monitoring, control and harvesting requires a lot of labor costs, which forces an increase in production and operation investment. Therefore, it is necessary to provide timely, intelligent, cost-effective and environmentally friendly solutions for the edible fungi industry.

[0005] With the advent of the fourth industrial revolution (Industry 4.0), innovative technologies such as the Internet of Things, cloud computing, big data and artificial intelligence are widely used in industrial automation. The parallel revolution in agriculture (Agriculture 4.0) has formed the so-called smart agriculture concept by applying these new technologies to the agricultural field. The concept of the Internet of Things is to connect all objects to the Internet through information sensing devices to exchange information for intelligent identification and management. Artificial intelligence is the study of using computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.). The Internet of Things and artificial intelligence technologies can achieve automatic climate management and reduce labor expenses, thereby reducing production costs. Using the Internet of Things system to control the internal climate and transform traditional greenhouses into smart incubators can help create optimal growth conditions for edible fungi under external weather conditions.

[0006] Deep learning is a branch of artificial intelligence. Due to the rapid development of deep learning, visual recognition algorithms have made significant progress. As a result, many object detection algorithms have emerged that can identify and provide information such as the exact location and volume of objects in images or videos. Visual recognition algorithms can obtain the real-time growth status of edible fungi and promptly identify the poor growth status of edible fungi. Therefore, the application of IoT technology and artificial intelligence can bring solutions with low labor costs and high planting efficiency to the edible fungi planting industry. Summary of the invention

[0007] The purpose of the present invention is to provide an intelligent edible fungus incubator based on artificial intelligence Internet of Things, collect photos of edible fungi at different growth periods and pictures of poor growth status, perform model training using a target recognition algorithm after preliminary screening and appropriate labeling of the pictures. The recognition results are formed into nodes and input into a graphic database to form an association with the data read by sensors such as temperature and humidity according to the time axis, and the data is further mined and analyzed. The processed data simulates the growth process of edible fungi through a deep learning framework, and the model adjusts environmental factors in real time according to the real-time data of the sensors and the image recognition results. Furthermore, when the incubator recognizes that the edible fungi are in a poor state due to the lack of a certain element, the nutrient solution is automatically extracted from the nutrient solution tank to supplement the missing elements by irrigation or atomization. At the same time, the present invention has the characteristics of automatic adaptation, constant temperature and humidity, and a high degree of intelligence.

[0008] The present invention solves the above technical problems with the following technical solutions: an intelligent edible fungus incubator based on artificial intelligence Internet of Things, comprising:

[0009] The edible fungi incubator body and the edible fungi growth frame module, spray atomization module, carbon dioxide module, fresh air circulation module, visual detection module, growth light module, temperature sensor module and intelligent disinfection module which are assembled on the edible fungi incubator body. The edible fungi incubator body is provided with thermal insulation interlayers on the upper, lower, left, right and rear surfaces, and the front side is a double-opening cold storage door structure.

[0010] The edible fungus growth frame module consists of an assembly guide rail and a combined support frame; a plurality of sliders are arranged on the assembly guide rail, and the combined support frame is connected to the assembly guide rail through the sliders. The combined support frame can adjust the axial position of the assembly guide rail through the provided sliders. Holes are evenly opened on the left and right sides of the two side columns of the combined support frame, and the assembly is completed by bolt connection and accessories. The combined support frame can be assembled into a grid form or a layer plate form, and a hook is also provided on the combined support frame.

[0011] The spray atomization module is composed of a heat pump, a water tank, a nutrient solution tank, a spray atomization pipeline and a humidity sensor. The heat pump is equipped with a heat recovery device to recover the heat generated by the condenser during industrial air conditioning refrigeration, thereby reducing the energy consumption required for heating during operation. It is used to detect and adjust the real-time humidity inside the incubator and supplement the elements that edible fungi lack during the growth period.

[0012] The CO2 module consists of a CO2 compressor, a CO2 air pipe and a CO2 concentration sensor. The value of the CO2 concentration sensor is read in real time; the start and stop of the CO2 compressor is remotely controlled by an artificial intelligence model or the central control background, and the CO2 concentration inside the incubator is adjusted in real time by the length of the compressor start time and the CO2 concentration sensor feedback value.

[0013] The fresh air circulation module is composed of a purification fresh air fan, an air pressure detection sensor and a wind speed detector. The start and stop of the purification fresh air fan is remotely controlled by the artificial intelligence model or the central control background, and the carbon dioxide concentration inside the incubator is adjusted in real time through the feedback values ​​of the air pressure detection sensor and the wind speed detector;

[0014] The growth light module consists of several light strips evenly distributed on the top of the incubator. The spectrum frequency and brightness are remotely adjusted through an artificial intelligence model or a control center and control background to adapt to the lighting conditions of different types of edible fungi.

[0015] The visual recognition module is composed of a recognition big data system and an image processing module. The system uses a programmable wide-angle camera, which is evenly distributed inside the incubator (a single incubator is arranged in two rows with a horizontal spacing of 2.5 meters) to collect growth status images at different periods and identify the real-time growth status of edible fungi;

[0016] The temperature sensing module consists of an industrial air conditioner, an equipment mounting frame, a sunshade net, and a temperature sensor. The outdoor unit of the industrial air conditioner is installed on the equipment mounting frame on one side of the incubator. The air outlet of the industrial air conditioner is a double-outlet air outlet arranged at the top of the inside of the edible fungus incubator, and one of the air outlets is installed on the equipment mounting frame inside the incubator. The start and stop of the industrial air conditioner is remotely controlled by an artificial intelligence model or the central control background. The cold and warm gases are evenly delivered to all parts of the incubator through the sunshade film, and the internal temperature of the edible fungus incubator is dynamically adjusted by the temperature sensor;

[0017] Furthermore, all electronic instruments in a single incubator are controlled by a host computer, and all electronic instruments can be enabled individually or simultaneously.

[0018] Furthermore, the historical data of all electronic instruments will be collected and stored in the local storage card and uploaded to the cloud for backup at regular intervals.

[0019] Furthermore, the camera will take photos at regular intervals and upload them to the database. After appropriately selecting some photos and weighing the edible fungi in the images, the corresponding weight of the edible fungi is marked through a labeling tool.

[0020] Furthermore, a digital twin model is established through the historical data of each sensor in the database and the volume weight of the edible fungi in the identified incubator. The model can predict the yield of edible fungi in the incubator and alarm for abnormal conditions.

[0021] Furthermore, through a deep learning model based on the growth database and image recognition results, the growth period of edible fungi is divided into mycelium growth period and fruiting body formation period (different varieties of edible fungi are slightly different), and the internal environmental parameters of the incubator are dynamically adjusted according to different growth stages to ensure that the growth of edible fungi is always within an appropriate range and avoid cognitive errors caused by manual experience judgment.

[0022] Furthermore, when the camera matches the image of the poor growth state of edible fungi in the database, and according to the matching result, the corresponding nutrient solution is extracted from the nutrient solution tank to supplement the lacking elements for the edible fungi in the incubator.

[0023] Furthermore, when the incubator identifies an abnormal condition and cannot match the existing database or the match is low, the sensor data and real-time images will be uploaded to the backend central control platform through the cloud platform for manual judgment and the abnormality will be recorded.

[0024] Furthermore, the incubator will modify the suitable growth range and image recognition features according to the variety of edible fungi being grown.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] The present invention realizes real-time monitoring of the growth environment of edible fungi and accurate grasp of the growth status of edible fungi through means such as Internet of Things technology and artificial intelligence technology, thereby improving the production efficiency and output of edible fungi, while reducing production costs and risks. Its standardized production process ensures the high quality and safety of the product, not only reduces the labor intensity and labor costs of traditional greenhouse planting, but also reduces the technical threshold of edible fungi planting, and reduces the negative impact of insufficient engineering and technical experience of growers on edible fungi production. The edible fungi incubator reduces the technical threshold of edible fungi planting, promotes the upgrading of the agricultural industry, promotes the efficient use of resources, and conforms to the concept of sustainable development. More importantly, it simplifies the operating process, broadens the channels for farmers to increase their income, and injects new vitality into the prosperity and development of the edible fungi industry.

[0027] In addition, the present invention can automatically replenish the required nutrient solution to the edible fungi by identifying the poor growth of the edible fungi, thereby reducing the probability of poor development of the edible fungi, ensuring the yield of the edible fungi and lowering the technical threshold for growers.

[0028] Finally, by establishing and matching different databases, the types of edible fungi grown are diversified, the use areas of edible fungi incubators are broadened, and the market risks brought to growers by growing a single variety of edible fungi are reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is an isometric view of the present invention;

[0030] Figure 2 It is a left side view of the present invention;

[0031] Figure 3 This is a structural diagram of the edible fungus growth framework module of the present invention;

[0032] Figure 4 This is a structural diagram of the spray atomization module of the present invention;

[0033] Figure 5 This is a structural diagram of the carbon dioxide module of the present invention;

[0034] Figure 6 This is a structural diagram of the fresh air circulation module of the present invention;

[0035] Figure 7 This is a diagram showing the structure of the visual inspection module of the present invention;

[0036] Figure 8 This is a diagram showing the structure of the grow light module of the present invention;

[0037] Fig. 9 It is a structural diagram of the temperature module of the present invention;

[0038] In the figure: 1. Edible fungus culture box body; 2. Edible fungus growth frame module; 201. Assembly guide rail; 202. Combined support frame; 203. Slider; 3. Spray atomization module; 301. Heat pump; 302. Water tank; 303. Nutrient solution tank; 304. Spray atomization pipeline; 305. Humidity sensor; 4. Carbon dioxide module; 401. Carbon dioxide compressor; 402. Carbon dioxide air pipe; 403. Carbon dioxide concentration sensor; 5. Fresh air circulation module; 501. Purification fresh air fan; 502. Air pressure detection sensor; 503. Wind speed detector; 6. Growth light module; 7. Visual recognition module; 8. Temperature sensor module; 801. Industrial air conditioner; 802. Equipment mounting frame; 803. Shading net; 804. Temperature sensor; 805. Air outlet of industrial air conditioner. DETAILED DESCRIPTION

[0039] The present invention is further described below in conjunction with the accompanying drawings and embodiments, but they do not constitute a limitation on the protection scope of the claims of the present invention.

[0040] Embodiment 1:

[0041] like Figure 1-9 As shown, the present invention provides a technical solution, in which the edible fungus variety planted in the example is Pleurotus geesteranus, and the planting form is mushroom stick planting: an intelligent edible fungus culture box based on artificial intelligence Internet of Things, including: an edible fungus culture box body 1, an edible fungus growth frame module 2, a spray atomization module 3, a carbon dioxide module 4, a fresh air circulation module 5, a visual detection module 6, a growth light module 7, a temperature sensor module 8 and an intelligent disinfection module. The edible fungus growth frame module 2, the spray atomization module 3, the carbon dioxide module 4, the fresh air circulation module 5, the visual detection module 6, the growth light module 7, and the temperature sensor module 8 are assembled with the edible fungus culture box body. The corresponding space is reserved in the edible fungus culture box body for each module, and the startup, deactivation, and morphological change of a single module will not affect the normal use of other modules. The upper, lower, left, right and rear surfaces of the edible fungus culture box are all composed of thermal insulation interlayer (100mm thick polyurethane insulation material, thermal conductivity: 0.024; closed porosity: 99.5%; tensile strength: 10MPa), and the front side is a double-opening cold storage door (the inner layer is polyurethane insulation material; the outer layer is composed of SII, PVC color steel plate, and stainless steel plate; the total thickness is 100mm).

[0042] The edible fungus growth frame module 2 is composed of an assembly guide rail 201 and a combined support frame 202; a plurality of sliders 203 are arranged on the assembly guide rail 201, and the combined support frame 202 is connected to the assembly guide rail 201 through the sliders 203. The combined support frame 202 can adjust the axial position of the assembly guide rail 201 through the set sliders 203. Holes are evenly opened on the left and right sides of the two side columns of the combined support frame 202, and the assembly is completed by bolt connection and accessories. The combined support frame 202 can be assembled into a grid form or a layer plate form, and a hook is also provided on the combined support frame 202.

[0043] After the edible fungus growth frame module 2 is assembled, it is assembled with the edible fungus incubator body 1. At this time, the edible fungus growth frame module adopts a grid form. After the Pleurotus geesteranus sticks are placed in the grid, the edible fungus incubator body is closed and locked. At this time, the entry of carbon dioxide, the entry of external air and the exhaust of internal air are all one-way channels, reducing the impact of external factors on the growth of edible fungi. Select the Pleurotus geesteranus recipe in the central control background, and import the corresponding database in the visual detection module. After the incubator starts working, the artificial intelligence model initializes the internal environmental parameters of the incubator according to the preset parameters. The artificial intelligence model adjusts the humidity, carbon dioxide concentration, temperature and other parameters in real time according to the growth status of Pleurotus geesteranus, and replenishes the missing nutrients for Pleurotus geesteranus in time. When the visual recognition module recognizes that the edible fungus has entered another growth stage, the artificial intelligence model decides to adjust the environmental factors in the edible fungus incubator and sends a prompt to the background. When the visual recognition module recognizes that the edible fungus is in the mature stage, the artificial intelligence model sends a prompt to the background and adjusts the environment in the edible fungus incubator to slow down the growth of the edible fungus, avoiding economic losses caused by the aging of edible fungi due to untimely harvesting.

[0044] Edible fungus growth frame module: The assembly rail 201 is fixed to the floor of the incubator by riveting. The combined support frame 202 is a grid, and the number of mesh holes of a single combined support frame is 25x15. The combined support frame is assembled on the slider 203 provided with the assembly rail. After adjusting the spatial position, it is connected to the assembly rail through positioning pins; wherein, holes are evenly opened on the left and right sides of the columns on both sides of the combined support frame, and the vertical distance of a single hole is 200mm. The assembly is completed by bolting with accessories (Accessory 1: grid, flange frame with corresponding holes; Accessory 2: layer plate, flange frame with corresponding holes). The combined support frame is assembled into a grid form (using mushroom stick form) and a layer plate form (using soil culture form) according to the growth conditions of edible fungi. Among them, whether it is a grid form or a layer plate form, a hook is provided at a vertical distance of 200mm for hanging pig and cattle bones to lure and kill mites;

[0045] Temperature module: It is composed of industrial air conditioner, sunshade net and temperature sensor. The industrial air conditioner is installed in the incubator. The air outlet of the industrial air conditioner is set at the middle top of the edible fungus incubator for double-sided ventilation. The cold and warm gases are evenly delivered to all parts of the incubator through the sunshade net. The temperature sensor is located inside the incubator to identify the real-time temperature inside the incubator and adjust the temperature inside the incubator. The temperature sensing module is composed of industrial air conditioner (compressor power 4410W, cooling capacity 15550W, cooling efficiency 4900W, heating capacity 16300W, heating efficiency 4980W), equipment mounting frame, sunshade net (needle number range: 1-6 needles, layer number range: 1-2 layers), and temperature sensor. The temperature sensor 403 is configured to monitor and feedback the temperature data inside the incubator to an integrated control unit in real time and continuously; the integrated control unit is further connected to the artificial intelligence model or the remote central control background, and automatically controls the start and stop time of the industrial air conditioner through remote instructions according to the real-time temperature data, so as to accurately control and maintain the temperature inside the incubator within the preset ideal range.

[0046] Spray atomization module: It consists of a heat pump 301, a water tank 302, a nutrient solution tank 303, a spray atomization pipeline 304 and a humidity sensor 305. The heat pump is installed on the same side of the industrial air conditioner. The heat pump is equipped with a heat recovery device to recover the heat generated by the condenser during the refrigeration of the industrial air conditioner. The spray atomization module extracts liquid from the water tank and the nutrient solution tank according to different needs, and after pressurization, it is sent to the branch humidification pipeline through the main humidification pipeline, and humidification and drug administration are performed according to different signals; the nutrient solution tank is located at the bottom of the incubator, and the nutrient solution tank (formula 1: sucrose 1%, calcium carbonate 1%, magnesium sulfate 0.05%, potassium dihydrogen phosphate 0.1%) contains a certain concentration of calcium, potassium and other ions nutrient solution, which is used to supplement the elements lacking in edible fungi during the growth period; the humidity sensor is located inside the incubator to detect the real-time humidity inside the incubator. The heat pump machine 301 also includes a control unit for adjusting the operating state of the heat pump, and automatically adjusting the heat recovery and supply according to the water temperature, ambient temperature and heat output of the industrial air-conditioning condenser to achieve the best energy efficiency ratio. A water level sensor and a temperature sensor are set inside the water tank to monitor the water level and water temperature in real time to ensure the stability and safety of the water supply. A plurality of atomizing nozzles are set on the spray atomizing pipeline 304, and the density and coverage of the spray can be adjusted according to actual needs. The spray atomizing pipeline 304 is equipped with a flow control valve and a pressure regulating device to achieve precise control of the flow and pressure of the spray and improve the spraying efficiency.

[0047] Carbon dioxide module: It consists of a carbon dioxide compressor 401, a carbon dioxide gas pipe 402 and a carbon dioxide concentration sensor 403. The carbon dioxide compressor is located on one side of the incubator, and the amount of carbon dioxide gas charged is controlled by controlling the start and stop time of the carbon dioxide compressor. Due to the high density of carbon dioxide gas, the carbon dioxide concentration sensors are located at different heights inside the incubator, and the real-time carbon dioxide concentration inside the incubator is detected in combination with the amount of carbon dioxide gas charged. The carbon dioxide concentration sensor 403 is configured to monitor the carbon dioxide concentration data inside the incubator in real time and continuously and to feedback the carbon dioxide concentration data inside the incubator to an integrated control unit; the integrated control unit is further connected to the artificial intelligence model or the remote central control background communication, and according to the preset carbon dioxide concentration threshold range and real-time concentration data, the start and stop time of the carbon dioxide compressor 401 is automatically adjusted through remote commands, so as to accurately control and maintain the carbon dioxide concentration in the incubator within the preset ideal range.

[0048] Fresh air circulation module: composed of a purification fresh air fan 501, an air pressure detection sensor 502 and a wind speed detector 503. The purification fresh air fan is installed on one side of the edible fungus incubator body, used for internal and external environmental gas exchange, and purifies the air to reduce the impact of miscellaneous bacteria in the air on edible fungi. The air pressure detection sensor is located inside the incubator, used to detect the internal air pressure of the incubator to ensure that the internal air pressure is in a stable state. The wind speed identification instrument is located at the air outlet of the fan inside the incubator, used to identify the real-time air volume of the gas exchange inside the incubator. The start and stop of the purification fresh air fan is remotely operated by the artificial intelligence model or the central control background, and the gas concentration inside the incubator is adjusted in real time through the feedback values ​​of the air pressure detection sensor and the wind speed detector; the purification fresh air fan (triple filtration, filter particle size ≤2.5 microns) can effectively filter particles with a particle size of 2.5 microns or more, including but not limited to bacteria, viruses, dust and other harmful substances in the air, to ensure that the air introduced into the incubator is pure and pollution-free, thereby reducing the adverse effects on the growth of edible fungi. The air pressure detection sensor is used to monitor the gas pressure state inside the incubator in real time. The wind speed detector is installed at the outlet of the fresh air duct to measure and record the wind speed of the fresh air blower. In addition, the air inlet and outlet are both one-way. The pressure data and wind speed data are transmitted to the integrated control unit in real time. The integrated control unit is further connected to the artificial intelligence model or the remote central control background to adjust the output power or start and stop state of the purification fresh air blower through remote commands according to the changes in wind speed and pressure, so as to achieve precise control of the gas flow in the incubator, ensure uniform gas distribution, and improve the culture efficiency.

[0049] Growth light module: evenly distributed at the top of the incubator, mainly in the form of white light and purple light, the spectral frequency and brightness are remotely adjusted by the artificial intelligence model or the control center and control background; the spectral frequency and brightness are remotely adjusted by the artificial intelligence model or the control center and control background. These light strips are evenly arranged at the top of the incubator to ensure that all areas in the incubator receive uniform light. The light strip uses an energy-efficient LED light source, which is suitable for providing a suitable lighting environment for edible fungi. Through the integrated control unit, the growth light module can output light of different spectral frequencies, including but not limited to red light and blue light. This function is remotely controlled by the artificial intelligence model or the control center and control background, and the spectral frequency is automatically adjusted according to the growth stage and light requirements of the edible fungi; the growth light module also has the characteristic of adjustable brightness, which can realize remote stepless adjustment of the brightness of the light strip. The integrated control unit is further connected to the artificial intelligence model or the remote control center and intelligently adjusts the brightness of the light according to the real-time monitoring data of the light intensity in the incubator, combined with the light tolerance and growth rate of the edible fungi, to create the most suitable lighting conditions and promote the healthy growth of edible fungi.

[0050] Visual recognition module: The camera is mainly responsible for real-time monitoring and regular photography. The cameras used by the system are programmable wide-angle cameras, which are evenly distributed inside the incubator (two rows in a single incubator, with a horizontal spacing of 2.5 meters) to collect and identify the growth status images of edible fungi at different stages; the recognition big data system uses various parameters such as mycelium coverage and the number of edible fungi of different sizes to determine the growth dynamics of mycelium and edible fungi at different stages in the production process of edible fungi. The collected images are imported into the image processing module and converted into square grayscale images. Fourier analysis is used as the derived input feature to avoid the system using computing resources to identify objects in irrelevant areas of the image. When identifying image targets, the Faster R-CNN algorithm is used to identify the bounding box of edible fungi. The edible fungi in the target image are weighed and labeled with corresponding data, and the corresponding target mass is measured by the algorithm.

[0051] The neural network model is trained by collecting images of the growth status of edible fungi at different periods. The two-dimensional images of edible fungi at fixed angles are obtained and dense point clouds are generated. The point clouds are fused to reversely solve and generate the three-dimensional posture of the edible fungi. The expected total yield of edible fungi is calculated and output by combining historical data, environmental parameters (such as temperature, humidity, etc.) and the growth cycle characteristics of edible fungi.

[0052] By importing the image library of edible fungi in poor growth status, it is possible to identify and judge whether the edible fungi are in a poor state, adjust the environmental climate in the incubator in time, and report abnormal conditions based on the identification results.

[0053] When the real-time image of edible fungus growth cannot match the image in the database or the degree of match is low, the background real-time image and sensor basic data are uploaded to the cloud background for manual and expert diagnosis system judgment.

[0054] The temperature sensing module consists of an industrial air conditioner 801, an equipment mounting frame 802, a sunshade net 803, and a temperature sensor 804. The start and stop of the industrial air conditioner is remotely controlled by an artificial intelligence model or the central control background. Since the mushroom stick cultivation mode is adopted, the sunshade net uses 6 needles and 2 layers. The cold and warm gases are evenly delivered to all parts of the incubator through the sunshade net, and the internal temperature of the edible mushroom incubator is dynamically adjusted through the temperature sensor.

[0055] When the big data recognition system identifies that edible fungi are in a poor growth state, it will extract nutrient solution from the nutrient solution tank (Formula 1: 1% sucrose, 1% calcium carbonate, 0.05% magnesium sulfate, 0.1% potassium dihydrogen phosphate, 97.85% deionized water; Formula 2: 1% each of urea, sucrose, and citric acid, 93% deionized water) to supplement the lacking elements for the edible fungi based on the judgment results.

[0056] Artificial Intelligence Model: The database for training the model is based on a historical dataset consisting of changes made by technicians, known visual recognition result data, and historical sensor data. During the growth of edible mushrooms, there is a time range in which it is assumed that certain actions are suitable for execution. In the process of evaluating changes, technicians define the time range. Based on this data, the final dataset is created by iterating all growth processes every hour. In these datasets, the time difference values ​​are first regarded as input features to evaluate the dynamics of the parameters used in the decision tree model. Then, the decision tree is used to define whether climate change is needed at the current moment, and the k-nearest neighbor algorithm is combined to obtain the parameter values ​​that should be used after the change.

[0057] The actual operation data collected by engineering and technical personnel in the early stage is combined with the historical data of each sensor and the visual recognition results to roughly divide the growth of edible fungi into the mycelium growth period and the fruiting body formation period (different varieties of edible fungi are slightly different). The artificial intelligence algorithm model automatically identifies the current growth stage and intelligently adjusts environmental parameters such as light intensity, temperature, humidity and carbon dioxide concentration to the preset parameter range accordingly.

[0058] It should be noted that all sensor data and real-time monitoring images can be uploaded to the central control background, realizing one-to-many management by a single technician. When the equipment is abnormal, the recognition image cannot be matched, or the edible fungus growth curve deviates significantly from the historical data, the central control background receives the abnormal prompt, and the technician adjusts the equipment online or offline to solve the abnormality.

[0059] The intelligent disinfection module consists of a fumigation device, a drug delivery device, a drug tank, a water pump and a drug spraying pipeline. The fumigation device includes an electric heating element and a fumigant container. The electric heating element is used to heat the fumigant (active ingredient content 56% aluminum phosphide) in the fumigant container to generate fumigation gas to disinfect the planting box; a variety of drugs (carbendazim, thiophanate, carbolic acid, etc.) are placed in the drug delivery device, and the drugs are quantitatively delivered into the drug tank according to the integrated control unit signal; the start and stop of the water pump is controlled according to the integrated control unit signal to control the running time. The drug spraying pipeline is equipped with a flow control valve, a pressure regulating device and multiple atomizing nozzles, and the flow rate (2-12m 2 / h), pressure (0.4-2.bar) and equivalent nozzle hole diameter (1.3mm-2.8mm); the integrated control unit is further connected with the artificial intelligence model or the remote central control background communication to issue disinfection prompts and make medicine ratios and adjust the spray flow and range according to the growth of different varieties of edible fungi.

[0060] Embodiment 2:

[0061] The present invention provides a technical solution, in which the edible fungus variety planted in Example 2 is white mushroom, and the planting form is soil cultivation: an intelligent edible fungus incubator based on artificial intelligence Internet of Things, comprising: an edible fungus incubator body, an edible fungus growth frame module, a temperature module, a spray atomization module, a carbon dioxide module, a circulation module, a visual recognition module and a growth light module.

[0062] Usually, the entire cycle of white mushrooms from inoculation to harvesting takes about 30 days. The growth cycle of white mushrooms can be divided into the white mushroom mycelium growth period and the fruiting body formation period. By analyzing the area of ​​the mycelium and the volume of the fruiting body, the visual recognition module and the artificial intelligence model will match different databases.

[0063] The implementation method of Example 2 is substantially the same as that of Example 1, except that:

[0064] Edible fungus growth frame module: The combined support frame 202 is in the form of a layer plate, and a single combined support frame has 5 layers. The vertical heights of the combined support frame layers are staggered to obtain more light;

[0065] Spraying and atomizing module: more watering spray nozzles are added to the spraying and atomizing pipeline 304, and formula 2 (1% each of urea, sucrose, and citric acid) is added to the nutrient solution tank 303 to cope with the longer growth cycle of white mushrooms in the incubator;

[0066] Carbon dioxide module: The growth of white mushrooms is more sensitive to carbon dioxide concentration, and carbon dioxide concentration sensors need to be installed in more locations to ensure data accuracy.

[0067] Visual recognition module: The recognition big data system needs to add the database of white mushroom mycelium growth period and fruiting body formation period;

[0068] Temperature sensing module: The uniformity of cold and warm air in soil cultivation is relatively low. Therefore, the sunshade net uses 3 needles and 1 layer.

Claims

1. An intelligent edible fungus incubator based on artificial intelligence Internet of Things, characterized in that: include: An edible fungus incubator body (1) and an edible fungus growth frame module (2), a spray atomization module (3), a carbon dioxide module (4), a fresh air circulation module (5), a visual detection module (6), a growth light module (7), a temperature sensor module (8) and an intelligent disinfection module which are arranged on the edible fungus incubator body (1) in an assembled manner. The edible fungus incubator body (1) is provided with heat-insulating interlayers on the upper, lower, left, right and rear five surfaces, and the front side surface is a double-opening cold storage door structure.

2. The intelligent edible fungus incubator based on artificial intelligence Internet of Things according to claim 1, characterized in that: The edible fungus growth frame module (2) is composed of an assembly rail (201) and a combined support frame (202); the assembly rail (201) is provided with a plurality of sliders (203); the combined support frame (202) is connected to the assembly rail (201) via the sliders (203); the combined support frame (202) can adjust the axial orientation of the assembly rail (201) via the provided sliders (203); the two side columns of the combined support frame (202) are evenly provided with holes on the left and right sides; the assembly is completed by bolt connection with accessories; the combined support frame (202) can be assembled into a grid form or a layer form; and the combined support frame (202) is also provided with a hook.

3. The intelligent edible fungus incubator based on artificial intelligence Internet of Things according to claim 1, characterized in that: The spray atomization module (3) is composed of a heat pump (301), a water tank (302), a nutrient solution tank (303), a spray atomization pipeline (304) and a humidity sensor (305). The heat pump (301) is provided with a heat recovery device, which is used to recover the waste heat released by the condenser during the refrigeration process of the industrial air conditioner, and directly use the recovered heat to heat the water in the water tank (302). The heat pump (301) is also provided with a control unit, which is used to adjust the operating state of the heat pump. The control unit can automatically adjust the recovery and supply of heat according to the water temperature, the ambient temperature and the heat output of the industrial air conditioner condenser. A water level sensor and a temperature sensor are provided inside the water tank to monitor the water level and water temperature in real time to ensure the stability and safety of the water supply. The spray atomization pipeline (304) is provided with a plurality of atomization nozzles, and the spray atomization pipeline (304) is also provided with a flow control valve and a pressure regulating device.

4. The intelligent edible fungus incubator based on artificial intelligence Internet of Things according to claim 1, characterized in that: The carbon dioxide module (4) is composed of a carbon dioxide compressor (401), a carbon dioxide air pipe (402) and a carbon dioxide concentration sensor (403), wherein the carbon dioxide concentration sensor (403) is configured to monitor the carbon dioxide concentration data inside the incubator in real time and continuously and to feed back the carbon dioxide concentration data to an integrated control unit; the integrated control unit is further connected to the artificial intelligence model or the remote central control background for communication, and automatically adjusts the start and stop time of the carbon dioxide compressor (401) through remote instructions according to a preset carbon dioxide concentration threshold range and real-time concentration data, so as to accurately control and maintain the carbon dioxide concentration in the incubator within a preset ideal range.

5. The intelligent edible fungus incubator based on artificial intelligence Internet of Things according to claim 1, characterized in that: The fresh air circulation module (5) is composed of a purification fresh air fan, an air pressure detection sensor and a wind speed detector. The purification fresh air fan can effectively filter particles with a particle size of more than 2.5 microns. The air pressure detection sensor is used to monitor the gas pressure state inside the incubator in real time. The wind speed detector is installed at the outlet of the fresh air duct and is used to measure and record the wind speed of the fresh air fan. The pressure data and wind speed data are transmitted to the integrated control unit in real time. The integrated control unit is further connected to the artificial intelligence model or the remote central control background so as to adjust the output power or start and stop state of the purification fresh air fan through remote commands according to the changes in wind speed and pressure, so as to achieve precise control of the gas flow in the incubator, ensure uniform gas distribution, and improve the culture efficiency.

6. The intelligent edible fungus incubator based on artificial intelligence Internet of Things according to claim 1, characterized in that: The growth light module (7) is composed of a plurality of light strips evenly distributed at the top of the interior of the incubator. The spectral frequency and brightness are remotely adjusted through an artificial intelligence model or a central control background. These light strips are evenly arranged at the top of the interior of the incubator to ensure that each area in the incubator can receive uniform light. Through an integrated control unit, the growth light module (7) can output light of different spectral frequencies.

7. The intelligent edible fungus incubator based on artificial intelligence Internet of Things according to claim 1, characterized in that: The visual recognition module (6) is composed of a recognition big data system and an image processing module. The visual recognition module (6) collects images of the growth status of edible fungi at different periods to train a neural network model. The two-dimensional images of the edible fungi at a fixed angle are obtained and a dense point cloud is generated. The point cloud is fused to reversely solve and generate a three-dimensional posture of the edible fungi. The expected total yield of the edible fungi is calculated and output in combination with historical data, environmental parameters and edible fungi growth cycle characteristics. By importing the image library of edible fungi in poor growth status, it is possible to identify and judge whether the edible fungi are in poor condition, adjust the environmental climate in the incubator in time, and report abnormal conditions based on the identification results; When the real-time image of edible fungi growth cannot match the image in the database or the matching degree is low, the background real-time image and basic sensor data are uploaded to the cloud background for manual and expert diagnosis system judgment; When the big data recognition system identifies that edible fungi are in a poor growth state, it will extract nutrient solution from the nutrient solution tank based on the judgment results to supplement the lacking elements for the edible fungi.

8. The intelligent edible fungus incubator based on artificial intelligence Internet of Things according to claim 1, characterized in that: The temperature sensing module (8) is composed of an industrial air conditioner, an equipment mounting frame, a sunshade net, and a temperature sensor. The temperature sensor (403) is configured to monitor the temperature data inside the incubator in real time and continuously and to feed back the temperature data inside the incubator to an integrated control unit; the integrated control unit is further connected to the artificial intelligence model or the remote central control background for communication, and automatically controls the start and stop time of the industrial air conditioner through remote commands based on the real-time temperature data, so as to accurately control and maintain the temperature inside the incubator within a preset ideal range.

9. The intelligent edible fungus incubator based on artificial intelligence Internet of Things according to claim 8, characterized in that: By combining the actual operation data of engineering and technical personnel collected in the early stage with the historical data of each sensor and the visual recognition results, the growth of edible fungi is roughly divided into the mycelium growth period and the fruiting body formation period. The artificial intelligence algorithm model automatically identifies the current growth stage, and intelligently adjusts environmental parameters such as light intensity, temperature, humidity and carbon dioxide concentration to the preset parameter range accordingly; for different varieties of edible fungi, you can manually switch to the corresponding database, and the artificial intelligence algorithm model adaptively adjusts the growth characteristic conditions of the corresponding edible fungi.

10. The intelligent edible fungus incubator based on artificial intelligence Internet of Things according to claim 1, characterized in that: The intelligent disinfection module is composed of a fumigation and disinfection device, a drug delivery device, a medicine tank, a water pump and a medicine spraying pipeline. The fumigation and disinfection device includes an electric heating element and a fumigant container. The electric heating element is used to heat the fumigant in the fumigant container to generate fumigation gas to disinfect the planting box; a variety of drugs are placed in the drug delivery device, and the drugs are quantitatively delivered into the medicine tank according to the signal of the integrated control unit; the start and stop of the water pump controls the running time according to the signal of the integrated control unit. The medicine spraying pipeline is equipped with a flow control valve, a pressure regulating device and multiple atomizing nozzles, and the flow, pressure and equivalent spray hole diameter of the nozzle are adjusted according to the signal of the integrated control unit; the integrated control unit is further connected to the artificial intelligence model or the remote central control background communication, and sends out disinfection prompts and performs medicine ratio and adjusts the spray flow and range according to the growth of different varieties of edible fungi.

Images