Intelligent temperature control method for edible mushroom cultivation growth chamber

Abstract

The application belongs to the technical field of temperature control, and specifically relates to an intelligent temperature control method for edible mushroom cultivation growth chambers. The method determines temperature abnormalities through a comprehensive evaluation function combining instantaneous over-limit proportion and standard deviation, and determines the control priority accordingly, realizes the transformation from average temperature control to abnormal area priority control, constructs a two-dimensional temperature distribution map of the layer and carries out gradient calculation, accurately locates the high-temperature area and its direction, directly controls the deflection direction and angle of the deflector through coordinate conversion and nonlinear mapping, makes the cold air flow preferentially flow to the high-temperature area, effectively solves the horizontal and vertical temperature uneven problems caused by the three-dimensional cultivation frame, intelligently fuses the temperature control effect verification with the mycelium growth speed compliance and the out-mushroom uniformity, quantifies the growth compliance through image analysis, and carries out hierarchical early warning based on the secondary verification result, and forms a complete intelligent closed-loop control system.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of temperature control, and specifically relates to an intelligent temperature control method for a mushroom cultivation growth chamber. BACKGROUND

[0002] Industrial cultivation of edible mushrooms is an important part of modern agricultural production, and the uniformity and stability of the temperature of the growth environment are key factors in determining yield and quality.

[0003] At present, the mushroom cultivation growth chamber mostly uses a three-dimensional layered cultivation rack to improve space utilization, but the layer rack structure easily leads to uneven temperature distribution in different levels and different areas. The traditional temperature control method mainly relies on the readings of a limited number of temperature sensors in the growth chamber to regulate the total output power of the air conditioning unit through a PID algorithm.

[0004] For example, a mushroom fruiting temperature control system and its control method disclosed in Chinese Patent Publication No. CN115968718A, by combining time information, temperature information and real-time growth condition information, reasonably control the mushroom fruiting temperature.

[0005] The above-mentioned scheme is more accurate in temperature regulation, but still has the following limitations: 1. For modern mushroom houses using three-dimensional layered cultivation racks, due to the effects of hot air rising, cold air sinking and rack blocking, there are complex temperature gradients at different vertical heights and different positions in the same horizontal layer. The control strategy based on point or regional average temperature cannot perceive and eliminate the temperature unevenness in three-dimensional space, resulting in abnormal temperature in local areas and affecting the consistency of mushroom growth.

[0006] 2. The existing method usually only takes the temperature set value as the control target, lacks the ability to accurately locate and prioritize the temperature abnormal area, and cannot dynamically adjust the air supply strategy according to the two-dimensional distribution characteristics of the temperature field, resulting in blind distribution of cooling air flow, low control efficiency and high energy consumption.

[0007] 3. The traditional control method forms an open-loop system, which lacks effective verification of the control effect. After the temperature returns to normal, whether the agricultural goal of promoting uniform mycelium growth and uniform mushroom production has been truly achieved still depends on manual experience, and cannot realize reverse verification and closed-loop optimization based on intelligent analysis of growth images, making it difficult to discover and warn of hidden growth abnormalities in a timely manner.

[0008] Therefore, there is an urgent need for an intelligent temperature control method that can realize accurate monitoring and abnormal diagnosis of three-dimensional layered temperature, directional guiding control based on spatial temperature field analysis, and quantitative evaluation and early warning of mushroom growth conformity by integrating image processing technology. SUMMARY

[0009] In order to overcome the shortcomings in the background art, the embodiment of the present application provides an intelligent temperature control method for edible mushroom cultivation growth chamber, which can effectively solve the problems involved in the above background art.

[0010] The purpose of the present application can be achieved by the following technical solutions: an intelligent temperature control method for edible mushroom cultivation growth chamber, comprising: acquiring temperature data of each layer of a three-dimensional layered cultivation rack and edible mushroom growth images.

[0011] Analyzing whether the temperature of each layer of the cultivation rack is abnormal, screening the corresponding layer rack with abnormal temperature, and determining the temperature control priority.

[0012] Analyzing the temperature difference in the up-down direction and the front-back direction of the temperature abnormality corresponding layer rack, and adjusting the deflector angle of the cold air input port according to the temperature difference according to the preset temperature adjustment rule.

[0013] After the temperature control is completed, the set temperature is maintained for a preset period, and during the preset period, the edible mushroom growth images of each cultivation unit of the corresponding layer rack are collected, the mycelium growth speed compliance and the mushroom uniformity of the edible mushrooms in each cultivation unit are analyzed based on the images from left to right, and the edible mushroom growth compliance is calculated based on the mycelium growth speed compliance and the mushroom uniformity.

[0014] Screening the cultivation unit whose edible mushroom growth compliance does not reach the preset compliance and recording its corresponding position, triggering the image acquisition device to move to the corresponding position for secondary image acquisition and again calculating the edible mushroom growth compliance.

[0015] If the edible mushroom growth compliance calculated again does not reach the preset compliance, a warning operation is triggered.

[0016] Compared with the prior art, the embodiment of the present application has at least the following advantages or beneficial effects: (1) The present application precisely locates the high temperature area and its direction by constructing a two-dimensional temperature distribution map of the layer rack and performing gradient calculation, directly controls the deflection direction and angle of the deflector through coordinate conversion and nonlinear mapping, makes the cold air flow preferentially flow to the high temperature area, effectively solves the horizontal and vertical temperature unevenness problem caused by the three-dimensional cultivation rack, and significantly improves the temperature uniformity.

[0017] (2) The present application determines whether the temperature is abnormal by combining the comprehensive evaluation function of instantaneous over-limit ratio and standard deviation, and determines the control priority accordingly, realizes the change from average temperature control to abnormal area priority control, avoids blind distribution of refrigeration capacity, improves the control efficiency, and reduces the energy consumption.

[0018] (3) The temperature regulation effect verification is intelligently fused with the mycelium growth speed compliance and the spore output uniformity of the edible fungi, the growth compliance is quantified through image analysis, and grading early warning is performed based on the secondary verification result, so that a complete intelligent closed-loop control system is formed, hidden growth abnormalities can be found in time, and the yield and quality are ensured. BRIEF DESCRIPTION OF DRAWINGS

[0019] The application will be further described by using the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the application, and other drawings can be obtained by those skilled in the art without creative labor on the premise of not paying creative labor.

[0020] Figure 1 The schematic diagram of the step flow of the method of the application is shown.

[0021] Figure 2 The schematic diagram of the flow of analyzing whether the temperature of each layer of the cultivation frame is abnormal is shown.

[0022] Figure 3 The schematic diagram of the flow of the early warning operation is shown. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application.

[0024] Please refer to Figure 1 As shown in the drawings, the application provides an intelligent temperature control method for edible fungi cultivation growth chamber, comprising: acquiring temperature data of each layer of a three-dimensional layered cultivation frame and edible fungi growth images.

[0025] Analyzing whether the temperature of each layer of the cultivation frame is abnormal, screening the corresponding layer frame with abnormal temperature, and determining the temperature regulation priority.

[0026] Analyzing the temperature difference in the up-down direction and the front-back direction of the corresponding layer frame with abnormal temperature, and adjusting the angle of the deflector of the cold air input port according to the temperature difference according to the preset temperature adjustment rule.

[0027] After the temperature regulation is completed, the set temperature is maintained for a preset period, and during the preset period, the edible fungi growth images of each cultivation unit of the corresponding layer frame are collected, the mycelium growth speed compliance and the spore output uniformity of the edible fungi in each cultivation unit are analyzed based on the images from left to right, and the edible fungi growth compliance is calculated based on the mycelium growth speed compliance and the spore output uniformity.

[0028] The cultivation unit with the edible mushroom growth compliance not reaching the preset compliance is screened out, and a corresponding position is recorded.

[0029] If the edible mushroom growth compliance calculated again does not reach the preset compliance, a pre-warning operation is triggered.

[0030] The embodiment of the application realizes the transition from average temperature control to abnormal area priority regulation by temperature regulation priority, intelligently fuses temperature regulation effect verification and edible mushroom mycelium growth speed compliance and mushroom output uniformity, quantifies the growth compliance, and performs hierarchical pre-warning based on the secondary verification result, forms a complete intelligent closed-loop control system, can timely find hidden growth abnormalities, and guarantees yield and quality.

[0031] In a preferred embodiment of the application, temperature sensors are arranged at the four corners of the upper and lower end faces of each layer of the three-dimensional layered cultivation rack, and the temperature sensors collect temperature data of the corresponding positions in real time at a preset sampling frequency.

[0032] Each layer is equipped with an image acquisition device that can slide back and forth in the horizontal direction, which is used to obtain edible mushroom growth images of each cultivation unit.

[0033] It should be noted that the temperature distribution in the layer space of the three-dimensional layered cultivation rack is easily affected by factors such as interlayer shielding and airflow flow, and the temperature at different positions may differ. Arranging temperature sensors at the four corners of the upper and lower end faces of the layer can collect temperature data from different directions, which can more comprehensively capture the temperature information in the layer compared to single-point sampling, reduce monitoring distortion caused by local temperature deviation, and provide more reliable basic data for subsequent temperature anomaly analysis.

[0034] The three-dimensional layered cultivation rack includes multiple cultivation units, and the image acquisition device that slides back and forth in the horizontal direction can traverse all cultivation units on the layer, avoiding the omission of edge cultivation units caused by fixed position sampling, ensuring the acquisition of complete growth images of each cultivation unit, and providing a reliable basis for comprehensive evaluation of the growth state of each cultivation unit. When the cultivation unit that does not reach the preset compliance is screened out, the device can accurately move to the corresponding position for secondary image acquisition, reducing image acquisition errors through repeated verification.

[0035] Referring to Figure 2 In a preferred embodiment of the application, the specific way of analyzing whether the temperature of each layer of the cultivation rack is abnormal includes: obtaining all temperature sampling data of the target layer within a preset time window to form a temperature time series data sequence of the layer.

[0036] It should be noted that the preset time window is obtained by experimental calibration, specifically: collect multiple sets of temperature fluctuation characteristic data of target strains at each cultivation stage, each data set contains temperature data of multiple repeated cultivation cycles, the sampling frequency is once every 10 minutes, the control environment humidity, illumination and other variables remain unchanged, in the mycelium growth stage, take different lengths of time window as algorithm input, monitor the recognition accuracy of temperature anomaly in each window, take the artificially marked temperature anomaly period as the benchmark, when the coincidence rate of the abnormal period identified by the algorithm under a certain time window and the benchmark is ≥ 90%, it is determined that the window is effective, finally, from all the effective windows, select the window length that performs stably on most data sets and has the highest calculation efficiency, and determine it as the preset time window.

[0037] The temperature time series data sequence is filtered and denoised, and a smoothed temperature change curve is calculated by using a moving average method.

[0038] The average temperature value of the smoothed temperature change curve is calculated, and the instantaneous offset amplitude of each sampling point temperature is calculated based on the average temperature value.

[0039] The number of sampling points with an instantaneous offset amplitude exceeding a preset amplitude threshold in the time window is counted, and the proportion of the number of sampling points with an instantaneous offset amplitude exceeding a preset amplitude threshold in the total number of sampling points is calculated, and is recorded as the instantaneous overrun ratio.

[0040] It should be noted that the instantaneous offset amplitude reflects the deviation of the temperature of a single sampling point from the average temperature, and the proportion of the number of sampling points exceeding the preset amplitude threshold quantifies the frequency of short-term and violent fluctuations of the temperature in the preset time window. The higher the instantaneous overrun ratio, the worse the short-term stability of the temperature. The instantaneous overrun ratio, as one of the core inputs of the comprehensive evaluation function, complements the standard deviation reflecting the overall dispersion of the temperature. A single instantaneous offset amplitude can only reflect the abnormality of an individual sampling point, while its proportion can reflect the universality of the abnormality, providing a quantitative basis for the high-frequency and short-term fluctuation dimension for the comprehensive judgment of whether the temperature in the layer is abnormal, and avoiding misjudgment caused by isolated abnormal points.

[0041] The preset amplitude threshold is the critical temperature amplitude at which short-term temperature deviation has a significant adverse effect on mycelium activity and fruiting body morphology under suitable growth temperature of the target strain. For example, when the target strain is golden needle mushroom, a short-term deviation temperature of 0.4°C, 0.6°C and 0.8°C is applied to the golden needle mushroom under a set temperature of 20°C, and the deviation lasts for 1 hour. Multiple repeated experiments show that when the deviation is 0.8°C, the daily growth rate of mycelium decreases by 25% from the standard speed, and the preset amplitude threshold can be set to 0.8°C.

[0042] Calculate the standard deviation of the temperature data in the time window, and establish a comprehensive evaluation function with the instantaneous overrun ratio and the standard deviation as inputs, and output a comprehensive abnormal index value.

[0043] It should be noted that the comprehensive evaluation function is constructed by weighted linear combination, the instantaneous overrun ratio reflects the frequency of short-term sudden abnormality of temperature, and the standard deviation reflects the overall dispersion degree of temperature, both of which characterize the temperature abnormality from different dimensions, and the weighted combination can avoid the one-sidedness of a single index, wherein the weight coefficients of the instantaneous overrun ratio and the standard deviation can be set according to the sensitivity of the edible fungus variety to short-term fluctuation and long-term stability of temperature, and the sensitivity setting is more experimental calibration, specifically, collecting multiple groups of mycelium daily growth rate and spore uniformity decay rate under short-term temperature fluctuation and long-term temperature dispersion, when the decay rate caused by short-term fluctuation is greater than long-term dispersion, it is determined that it is more sensitive to short-term sudden change, otherwise it is more sensitive to long-term stability, for example, if the golden needle mushroom is sensitive to short-term temperature sudden change, the weight coefficient of the instantaneous overrun ratio is greater than that of the standard deviation, the weight coefficient of the exemplary instantaneous overrun ratio is 0.6, and the weight coefficient of the standard deviation is 0.4, the flat mushroom requires higher long-term stability, and the weight coefficient of the standard deviation is greater than that of the instantaneous overrun ratio, the weight coefficient of the exemplary instantaneous overrun ratio is 0.3, and the weight coefficient of the standard deviation is 0.7.

[0044] The comprehensive abnormal index value is compared with the preset threshold value, and if it exceeds the preset threshold value, it is determined that the temperature of the layer is abnormal.

[0045] It should be noted that the preset threshold value is derived by combining the biological characteristics of edible fungus cultivation, historical temperature data and actual regulation and control requirements, specifically: the temperature fluctuation limit of the target fungus under the suitable growth temperature is determined by experiment, that is, when the temperature abnormality degree exceeds a certain value, it will significantly affect the mycelium growth rate or spore uniformity, and when the mycelium growth rate decreases by ≥20% or the spore uniformity dispersion coefficient increases by ≥30%, it is determined that it is significantly affected, which is used as the biological basis for threshold setting, and multiple groups of layer temperature data in the normal cultivation period are collected, the distribution range of the comprehensive abnormal index value is calculated, and the maximum comprehensive abnormal index value is taken as the preset threshold value.

[0046] The comprehensive evaluation function combining the instantaneous overrun ratio and the standard deviation is used to determine the temperature abnormality, and the regulation and control priority is determined accordingly, which realizes the change from average temperature control to abnormal area priority regulation and control, avoids blind distribution of refrigeration capacity, improves the regulation and control efficiency, and reduces the energy consumption.

[0047] All layers determined to be temperature abnormality are screened out, and sorted according to the comprehensive abnormal index value of each layer to determine the temperature regulation and control priority.

[0048] In a preferred embodiment of the present application, the temperature regulation priority determination method comprises: sorting all the temperature abnormal shelves screened out in descending order according to their comprehensive temperature abnormal index values to generate a first priority sequence.

[0049] In the first priority sequence, if there are shelves with the same comprehensive abnormal index value, then the shelves are further sorted according to their heights, and the shelves with higher heights in the three-dimensional layered cultivation shelf are preferentially regulated.

[0050] It should be noted that the comprehensive temperature abnormal index value is a core quantitative index reflecting the severity of temperature abnormality of the shelf. The higher the value, the more significant the deviation of the temperature from the normal state, and the greater the potential impact on the growth of edible fungi. Therefore, it needs to be preferentially regulated. In the three-dimensional layered cultivation shelf, the shelves with different heights are affected by environmental factors such as airflow and heat distribution. Generally, the high shelves may have a higher risk of temperature abnormality spreading due to weaker airflow circulation, which indirectly affects the temperature stability of the lower shelves. Therefore, when the comprehensive abnormal index values are the same, preferentially regulating the high shelves can reduce the chain effect of abnormality, which meets the regulation requirements of the three-dimensional cultivation environment.

[0051] In a preferred embodiment of the present application, the preset temperature adjustment rule comprises: obtaining real-time temperature values of monitoring points in the front-back direction and the up-down direction of the temperature abnormal shelf to construct a two-dimensional temperature distribution map of the shelf.

[0052] Gradient calculation is performed on the two-dimensional temperature distribution map to obtain the maximum direction and maximum amplitude of the temperature spatial variation rate. The maximum direction indicates the direction of the high-temperature area in the shelf plane.

[0053] The maximum direction is converted from the shelf coordinate system to the deflection coordinate system of the guide vane to calculate the target deflection direction of the guide vane for eliminating the temperature unevenness.

[0054] It should be noted that the shelf coordinate system takes the shelf plane as the reference, and can set the horizontal front-back direction as the X-axis, the vertical up-down direction as the Y-axis, and a fixed vertex such as the lower left corner as the origin for quantifying the maximum direction of the temperature gradient in the shelf plane.

[0055] The guide vane deflection coordinate system takes the cold air input port as the reference, sets the initial vertical state of the guide vane as 0°, and sets the left deflection angle as positive and the right deflection angle as negative for defining the mechanical action direction of the guide vane.

[0056] The maximum direction of the temperature gradient in the shelf coordinate system is converted to the corresponding angle in the guide vane coordinate system through a geometric transformation matrix. The conversion relationship between the shelf coordinate system and the guide vane deflection coordinate system is determined based on the relative position of the cold air input port and the shelf, and the angle from the origin to the maximum gradient direction in the shelf coordinate system is defined as , the target deflection angle of the deflector is , and the conversion relationship is: , wherein, is a scaling coefficient, is a bias compensation angle, and a specific value thereof is determined by a relative installation position of the cold air input port and the shelf and airflow field simulation experiments.

[0057] The simulation experiment specifically includes: for different relative installation positions of the cold air input port and the shelf, simulating a diffusion path of cold air in a shelf plane, temperature field changes, and effects on a high-temperature region, using ANSYS Fluent software, constructing a three-dimensional model of the shelf space, setting parameters such as cold air speed, initial temperature, and high-temperature region position, simulating airflow distribution rules under different deflector deflection angles through numerical calculation, observing whether the cold air flow can accurately cover the high-temperature region in the shelf through iterative testing of deflector actions corresponding to different values and values, and finally screening out optimal values and values that make the temperature gradient amplitude decrease to a target range.

[0058] For example, when the cold air input port is directly opposite the center of the shelf, , , if , then .

[0059] The shelf coordinate system is used to accurately describe the temperature distribution characteristics in the shelf plane, and the deflection of the deflector depends on its own mechanical coordinate system. The two belong to different spatial dimensions. Through coordinate system conversion, the deflection direction of the deflector is strictly corresponding to the spatial position of the high-temperature region in the shelf, avoiding the cold air flow to meaningless areas, ensuring that the cold quantity accurately acts on the temperature abnormal point, and rapidly balancing the temperature of the shelf.

[0060] The maximum amplitude of the temperature space change rate is input into a predefined nonlinear function, which defines the mapping relationship between the gradient amplitude and the deflection angle of the deflector, and the output is the target deflection angle of the deflector.

[0061] The predefined nonlinear function is calibrated through experiments, specifically: for the target strain cultivation shelf structure and the cold air input system, under different maximum amplitudes of the temperature space change rate, test the improvement effect of different deflector deflection angles on temperature unevenness, record the optimal deflection angle that makes the temperature gradient amplitude decrease to the target range, for example, the deflector deflection angle test range is-40° to +40°, every 5° is an interval, each group of parameters is tested 3 times, after eliminating abnormal values exceeding the average value ±10%, the average value is taken, and the minimum deflection angle that makes the temperature gradient amplitude decrease to the target range is recorded as the optimal angle.

[0062] The maximum amplitude of the temperature spatial change rate in the experiment is taken as input, and the corresponding optimal deflection angle is taken as output, and a mapping relationship between the two is established by polynomial fitting to form a predefined nonlinear function, for example, when the gradient amplitude is small, the function outputs a small deflection angle, and when the gradient amplitude increases significantly, the function outputs an angle that increases rapidly, so as to enhance the cold air guiding effect.

[0063] The maximum amplitude of the temperature spatial change rate directly reflects the severity of temperature unevenness, and the corresponding optimal deflection angle is a key parameter for eliminating the unevenness, and the deflection angle output by the function can effectively guide the cold air flow to the high-temperature area through experimental verification, weaken the temperature gradient by enhancing local cold supply, and finally realize temperature distribution uniformization.

[0064] The target deflection direction and the target deflection angle are combined to generate a guide vane control instruction to drive the guide vane to perform corresponding actions, so that the cold air flow preferentially flows to the high-temperature area.

[0065] It should be noted that the target deflection direction indicates the spatial orientation of the high-temperature area to which the cold air flow needs to be directed, and the target deflection angle quantifies the specific amplitude by which the guide vane needs to be deflected, and the combination of the two can accurately control the flow direction and intensity of the cold air flow, directly targeting the temperature unevenness problem in the shelf for directional regulation and control. The control instruction generated by combining the target deflection direction and the angle can accurately guide the cold air flow to the high-temperature area through the guide vane, avoid wasting cold energy in non-high-temperature areas, quickly balance the temperature of the shelf, shorten the duration of temperature abnormalities, and reduce the adverse effects on the growth of edible fungi.

[0066] By constructing a two-dimensional temperature distribution map of the shelf and calculating the gradient, the high-temperature area and its direction are accurately located, and the deflection direction and angle of the guide vane are directly controlled through coordinate conversion and nonlinear mapping, so that the cold air flow preferentially flows to the high-temperature area, effectively solving the horizontal and vertical temperature unevenness problems caused by the vertical cultivation shelf, and significantly improving the temperature uniformity.

[0067] In a preferred embodiment of the present application, the mycelium growth speed conformity includes: acquiring multiple edible fungus growth images of the same cultivation unit collected in time sequence to form an image time sequence of the cultivation unit.

[0068] The images in the image time sequence are preprocessed, including grayscale, noise reduction and image enhancement, to improve the feature recognition of the images.

[0069] Edge detection is used to accurately extract the mycelium area from the preprocessed image, distinguish the mycelium from the non-target area, and obtain a binary profile image of the mycelium area.

[0070] Feature reference points are selected in the binarized contour image. By comparing images at different time points using image registration technology, the displacement distance of the feature reference points is calculated. Combined with the acquisition time interval, the growth rate of mycelium is obtained.

[0071] It should be noted that the feature reference point is a stable and recognizable point in the binary contour image of the mycelial region, such as the inflection point of the mycelial edge. Through the feature registration algorithm, images of the same cultivation unit collected at different time points are aligned to eliminate background offset interference caused by slight displacement or vibration of the equipment during the image acquisition process.

[0072] In the image after interference is eliminated, the pixel coordinates of the same feature reference point are obtained in the images at two different time points. The pixel displacement distance of the point is calculated using the Euclidean distance formula. Then, combined with the physical scale of the image, the pixel distance is converted into the actual physical displacement distance. The Euclidean distance formula is an existing calculation formula and will not be elaborated here.

[0073] Dividing the calculated actual physical displacement distance of the feature reference point by the time interval between the two image acquisitions yields the mycelial growth rate. For example, if the displacement distance of a certain feature reference point within 24 hours is... The mycelial growth rate is .

[0074] By selecting stable feature points and combining them with image registration technology, the spatial position changes of mycelia in the time dimension can be accurately captured, eliminating the interference of non-growth factors. The growth rate can be calculated by correlating the displacement distance with the time interval, directly quantifying the dynamic growth state of mycelia. This provides an objective basis for subsequent comparison with standard growth rates and calculation of the growth rate conformity, ensuring that the assessment of the growth status of edible fungi is accurate and reliable.

[0075] The actual measured mycelial growth rate is compared with the standard growth rate of the edible fungus variety at a set temperature, and the growth rate conformity is calculated using a preset formula.

[0076] It should be noted that the preset formula is: ,in For growth rate compliance, This represents the actual growth rate of the mycelium. The formula is the standard growth rate at a set temperature. It is directly related to the actual growth rate of the mycelium and the standard growth rate at the set temperature. The core logic is that the closer the growth rate is to the standard growth rate at the set temperature, the higher the degree of conformity.

[0077] The standard growth rate at the set temperature is obtained by referring to the growth rate data of the target strain at the set temperature published in academic research literature and industry technical specifications, combining small-scale cultivation experiments under ideal conditions, comparing and calibrating the measured rate with the literature value, and finally obtaining the standard growth rate at the set temperature.

[0078] In a preferred embodiment of the present application, the uniformity of the out-of-fungus includes: obtaining multiple edible mushroom growth images of the same cultivation unit collected in time sequence, and the collection range needs to cover the entire cultivation unit to avoid missing the edge area.

[0079] The collected edible mushroom growth images are subjected to denoising, background separation and image enhancement processing.

[0080] A pre-trained target detection neural network model is used to recognize the processed edible mushroom growth images, locate each sub-entity in the images, and generate the minimum bounding rectangle boundary box thereof.

[0081] It should be noted that the target detection neural network model uses a YOLOv5 model, and the processed edible mushroom growth images are input into the YOLOv5 model. Through network structures such as convolution layers and pooling layers, the feature information such as edges, textures, shapes, etc. of the sub-entities is extracted from the input images layer by layer, and high-order features representing the sub-entities are gradually abstracted.

[0082] Based on the extracted features, the model outputs the regions where the sub-entities may exist as candidate boxes and calculates the confidence of each candidate box belonging to the sub-entity category. At the same time, the coordinates of the candidate boxes are optimized by a boundary box regression algorithm to make them more accurately surround the sub-entities.

[0083] According to the confidence threshold, high-confidence candidate boxes are selected and low-confidence candidate boxes that are false detections are removed. For each sub-entity after screening, the model generates a minimum bounding rectangle boundary box that can completely surround the sub-entity. The boundary box is output in the form of coordinates, realizing accurate positioning of each sub-entity.

[0084] An instance segmentation algorithm is used to segment the pixel area of each sub-entity from the processed edible mushroom growth images.

[0085] For each detected sub-entity, its morphological features are calculated, including calculating its pixel area, calculating the aspect ratio of its minimum bounding rectangle boundary box, fitting its cap with a minimum circumscribed circle and calculating the diameter.

[0086] It should be noted that the pixel area is a sub-entity pixel-level mask output based on the instance segmentation algorithm, and the total number of pixels in the mask is directly counted, which is the pixel area of the sub-entity. Combined with the physical scale of the image, the pixel area is converted into the actual physical area.

[0087] The minimum circumscribed circle diameter of the cap is obtained by first extracting the contour of the cap through edge detection, and then fitting the cap contour using a minimum circumscribed circle algorithm. The diameter of the circumscribed circle can be converted into an actual physical length through the conversion relationship between pixels and actual scales. Using the minimum circumscribed circle algorithm to fit the cap contour can more stably and accurately reflect the actual size of the cap compared to directly using the side length of the minimum circumscribed rectangle, avoiding measurement deviations caused by random directions of the cap, thereby making the evaluation results of the uniformity of the outgrowth more objective and reliable.

[0088] The statistical dispersion degrees of all sub-entities on the above morphological characteristics are calculated to obtain an area dispersion coefficient, a length-width ratio dispersion coefficient, and a cap diameter dispersion coefficient.

[0089] It should be noted that the dispersion coefficient is a standardized index for measuring the dispersion degree of data. The dispersion coefficient calculation formula is: wherein is the dispersion coefficient, is the standard deviation of the morphological characteristic data, is the average value of the morphological characteristic data, and the area dispersion coefficient, the length-width ratio dispersion coefficient, and the cap diameter dispersion coefficient can be calculated from the dispersion coefficient calculation formula.

[0090] For example, a certain cultivation unit detects 5 sub-entities, with cap diameter measurement values of 5.0 cm, 5.2 cm, 4.7 cm, 5.3 cm, and 4.8 cm. The average value is calculated to be , and the standard deviation is . Therefore, the cap diameter dispersion coefficient is , The smaller the value, the higher the uniformity.

[0091] The dispersion coefficient eliminates the influence of the dimension and magnitude of the characteristic data, and can objectively reflect the consistency of the morphological characteristics of the same batch of sub-entities. For example, the smaller the area dispersion coefficient, the more uniform the size of the sub-entities. A large length-width ratio dispersion coefficient indicates that the morphological differences of the sub-entities are significant, and there may be problems with uneven growth environment. These coefficients provide a quantitative basis for evaluating the uniformity of the growth of the edible mushroom population and are important indicators for subsequent comprehensive judgment of the growth state.

[0092] The spatial distribution uniformity of all sub-entities in the cultivation unit image is calculated. The image is divided into a plurality of grids, the number of sub-entities in each grid is counted, and the standard deviation of the number of sub-entities in all grids is calculated. The standard deviation is used to represent the uniformity of the spatial distribution.

[0093] The area dispersion coefficient, the length-width ratio dispersion coefficient, the cap diameter dispersion coefficient, and the spatial distribution standard deviation are normalized.

[0094] It should be noted that the normalization processing adopts a linear normalization processing mode, and the area dispersion coefficient, the length-width ratio dispersion coefficient, the cap diameter dispersion coefficient and the spatial distribution standard deviation calculated above are respectively mapped to the 0-1 interval, and the specific calculation formula is: wherein: is the characteristic value of the normalization result, is the original characteristic value to be normalized, is the minimum value, is the maximum value.

[0095] The minimum value and the maximum value can be determined based on the normal range of the target fungus and the same growth stage in the historical cultivation data, and the original value characteristic value is the dispersion coefficient of the morphological feature calculated above.

[0096] The above four normalized characteristic values are fused into a comprehensive score by a weighted fusion algorithm, and the comprehensive score is the fruiting body uniformity.

[0097] It should be noted that the weighted fusion algorithm refers to an algorithm for linearly weighting and summing a plurality of according to a preset weight to obtain a comprehensive score, wherein the preset weight can be calibrated according to historical data combined with expert experience, and specifically: a plurality of cultivation data are collected, including the original data of the above four characteristic values and the corresponding final cultivation quality score in a normal batch, the above four characteristic values are combined by two by two based on the experience of experts, the importance of each characteristic to the uniformity is compared and judged, a judgment matrix is constructed, a subjective weight is calculated, the information entropy of each characteristic value is calculated based on the original data of the four characteristic values, a set of objective weights is calculated according to the variation degree, and the subjective weight and the objective weight are calculated by linear combination to obtain the final weight. The final weight is applied to a new cultivation batch for verification and fine-tuning, and is finally fixed as the preset weight used by the system.

[0098] The influence weight of different characteristics on the uniformity is different, and the weighted mode can flexibly reflect this difference.

[0099] The core of the fruiting body uniformity is the consistency of the morphological characteristics and the spatial distribution of the fruiting bodies in the same batch. The more uniform the morphological characteristics and the more uniform the distribution, the higher the uniformity, the smaller the dispersion coefficient, the smaller the difference between the fruiting bodies in the area, shape and other dimensions, and the smaller the spatial distribution standard deviation, the more uniform the distribution of the fruiting bodies in the growth area. The comprehensive score after weighted fusion integrates these dispersed consistency indicators into a single numerical value. The smaller the numerical value, the higher the overall uniformity, and vice versa. Therefore, the fruiting body uniformity can be directly represented.

[0100] In a preferred embodiment of the present application, the calculation of the growth compliance of edible fungi based on the mycelium growth speed compliance and the spore emission uniformity comprises: normalizing the calculated mycelium growth speed compliance and spore emission uniformity to make them in the same numerical dimension range, and assigning weight values to them respectively.

[0101] The normalized mycelium growth speed compliance, spore emission uniformity and their corresponding weight values are fused by using a weighted geometric mean algorithm to calculate the final growth compliance of edible fungi.

[0102] It should be noted that the mycelium growth speed compliance and the spore emission uniformity are core indicators for measuring the growth state of edible fungi, and both of them determine the overall growth compliance. The mycelium growth speed compliance is the basis, and the spore emission uniformity is a direct manifestation of the growth quality, which needs to be evaluated comprehensively.

[0103] The original values of the calculated mycelium growth speed compliance and spore emission uniformity are processed by using the same linear normalization method as the above morphological characteristics to make them in the same numerical dimension range, and weight values are assigned to them respectively. The weight values can be determined according to historical data combined with expert experience, and the determination logic is consistent with the determination method of the internal weight of the spore emission uniformity.

[0104] The normalized mycelium growth speed compliance and spore emission uniformity and their corresponding weight values are fused by using a weighted geometric mean algorithm, and the calculation formula of the growth compliance of edible fungi is: , wherein , is the weight value, is the normalized spore emission uniformity, is the normalized mycelium growth speed compliance. The geometric mean algorithm is more stable to extreme values than the simple arithmetic mean, and the final growth compliance result is more reliable. The fusion of the two key indicators avoids the one-sidedness of single indicator evaluation and more comprehensively reflects the growth state of edible fungi.

[0105] In a preferred embodiment of the present application, the two-dimensional temperature distribution map is gradient calculated to obtain the maximum direction and maximum amplitude of the temperature spatial variation rate, comprising: discretizing the layer plane into a two-dimensional temperature matrix composed of grid points.

[0106] The temperature gradient components of each grid point in the two-dimensional temperature matrix in the direction and the direction are calculated.

[0107] According to the gradient components, the gradient amplitude and gradient direction of each grid point are calculated.

[0108] Iterate through the gradient magnitudes of all grid points to find the maximum gradient magnitude and the coordinates of the grid point where it is located.

[0109] Extract the gradient direction corresponding to the coordinate point, and use it as the direction of the maximum rate of change of the temperature space.

[0110] It should be noted that the temperature distribution on the shelf plane is a continuous spatial field, but in actual measurement and calculation, it is impossible to obtain the temperature of every point. It is necessary to discretize the continuous space into a set of finite grid points, i.e., a two-dimensional temperature matrix. This process is consistent with the computer's sampling and storage logic for spatial data, and it is convenient to approximate the overall temperature distribution through the temperature values ​​of discrete points.

[0111] Gradient is a mathematical concept describing the local rate of change of a multivariate function. In two-dimensional space, the temperature gradient... direction and The directions reflect the temperature variation trends in the horizontal and vertical directions, respectively. The central difference method is used to calculate the gradient components. The gradient magnitude is obtained by taking the square root of the sum of the squares of the two gradient components. The calculation of the gradient magnitude is based on the mathematical definition of the vector magnitude and direction, which can quantify the strength and direction of temperature changes. By traversing all grid points and filtering the maximum gradient magnitude and the grid point coordinates where it is located, the gradient direction corresponding to the coordinate point is extracted. This can accurately pinpoint the direction of the fastest temperature rise in the region. This direction directly reflects the dominant distribution trend of temperature spatial differences, providing a clear direction for analyzing the influencing factors such as the location of heat sources and ventilation paths in the environment.

[0112] Please see Figure 3 As shown, in a preferred embodiment of the present invention, the triggering of the early warning operation includes: comparing the recalculated edible fungus growth compliance with a preset compliance threshold.

[0113] If the recalculated edible fungus growth compliance is much lower than the preset compliance threshold, an advanced warning operation is triggered.

[0114] If the recalculated edible fungus growth compliance is close to but does not reach the preset compliance threshold, a low-level warning operation is triggered.

[0115] The early warning operation includes an early warning report containing the location number of the abnormal cultivation unit, image data, and the calculation results of the growth compliance.

[0116] The warning report is sent to the remote monitoring terminal.

[0117] It should be noted that the preset compliance threshold is determined through historical data experiments. Specifically, multiple sets of growth compliance samples of known target bacterial species at different growth stages are collected, the optimal growth range that can guarantee their normal growth and development is statistically determined, and the minimum compliance value corresponding to the growth state within this range is used as the preset compliance threshold.

[0118] The edible mushroom growth compliance is far less than the preset compliance threshold, and the specific determination method is that if the recalculated compliance value is significantly lower than the preset threshold and has exceeded the lower limit value that the edible mushroom can tolerate under normal growth conditions, the lower limit value is set based on experimental data, and an exemplary lower limit value is 70% lower than the preset threshold. For example, when the target mushroom species is Flammulina velutipes and the preset compliance threshold is 80 points, the simulated growth conditions of 75 points, 65 points, and 55 points of growth compliance are applied to Flammulina velutipes, respectively, each group is repeated for 3 times, and the monitoring is continued for 3 days. The results show that when the growth compliance is 55 points, the mycelium growth stagnation rate reaches 30%, and the fruiting body deformity rate rises to 40%, which is significantly lower than the normal growth state. Therefore, it can be determined that the growth compliance of 55 points and below is far less than the preset threshold, which means that the current environmental conditions have deviated significantly from the suitable range, which may lead to growth stagnation, serious quality decline, or even death. At this time, it is determined that the growth compliance is far less than the preset threshold, and a high-level early warning is triggered for emergency intervention.

[0119] The edible mushroom growth compliance is close to but does not reach the preset compliance threshold, and the specific determination method is that the recalculated compliance value is close to but does not reach the preset compliance threshold, but the compliance value is within the critical interval of the preset threshold. The critical interval is set based on experimental data, that is, by determining the influence of the target mushroom species on mycelium activity and fruiting body development when the growth compliance is near the threshold, the numerical range that will not cause serious growth problems but has potential risks is determined.

[0120] The critical interval is set based on experimental data, and an exemplary critical interval is 80%-99% of the preset threshold. When the target mushroom species is Flammulina velutipes and the preset compliance threshold is 80 points, the simulated growth conditions of 78 points, 70 points, and 65 points of growth compliance are applied to Flammulina velutipes, respectively, each group is repeated for 5 times, and the monitoring is continued for 2 days. The results show that when the growth compliance is 78 points, the mycelium growth rate decreases by 5% compared to the normal state of 80 points, the fruiting body uniformity decreases slightly but the deformity rate is still less than 8%. When the compliance is 70 points, the growth rate decreases by 10%, and the deformity rate rises to 12%. When the compliance is 65 points, the growth rate decreases by 15%, and the deformity rate is 15%. None of them has growth stagnation or serious quality decline, but there is a significant difference from the normal state. Therefore, it can be determined that the growth compliance of 65-78 points, which is 81.25%-97.5% of the preset threshold critical interval, is close to but does not reach the preset compliance threshold.

[0121] The temperature regulation effect verification is intelligently integrated with the mycelium growth rate compliance and the fruiting body uniformity of edible mushrooms. The growth compliance is quantified through image analysis, and the secondary verification results are used for hierarchical early warning, forming a complete intelligent closed-loop control system that can timely detect hidden growth abnormalities and ensure yield and quality.

[0122] The above formulas are all dimensionless values calculated, and the formulas are obtained by collecting a large amount of data to simulate a recent real situation, and the preset parameters in the formulas are set by a person skilled in the art according to actual conditions.

[0123] The above embodiments can be realized wholly or partially by software, hardware, firmware or any other combination. When realized by software, the above embodiments can be realized wholly or partially in the form of a computer program product.

[0124] Those skilled in the art can realize that the modules and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solutions. A person skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0125] In addition, each functional module in each embodiment of the present application can be integrated in one processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0126] The above is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0127] Finally, the above is merely preferred embodiments of the present application, and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for intelligent temperature control in an edible fungus cultivation and growth chamber, characterized in that, The method comprises the following steps: acquiring temperature data and edible mushroom growth images of each layer of the stereoscopic layered cultivation frame; analyzing whether the temperature of each layer of the cultivation frame is abnormal, screening the corresponding layer frame with abnormal temperature, and determining the temperature control priority; analyzing the temperature difference in the up-down direction and the front-back direction of the corresponding layer frame with abnormal temperature, and adjusting the angle of the deflector of the cold air input port according to the temperature difference and the preset temperature adjustment rule; after the temperature control is completed, maintaining the set temperature for a preset period, and during the preset period, collecting edible mushroom growth images of each cultivation unit of the corresponding layer frame, analyzing the mycelium growth speed compliance and the mushroom uniformity of the edible mushrooms in each cultivation unit from left to right based on the images, and calculating the edible mushroom growth compliance based on the mycelium growth speed compliance and the mushroom uniformity; screening the cultivation unit whose edible mushroom growth compliance does not reach the preset compliance and recording the corresponding position, triggering the image acquisition device to move to the corresponding position for secondary image acquisition and again calculating the edible mushroom growth compliance; if the edible mushroom growth compliance calculated again does not reach the preset compliance, triggering the warning operation; the mycelium growth speed compliance comprises: acquiring multiple edible mushroom growth images of the same cultivation unit collected in time sequence to form an image time sequence of the cultivation unit; preprocessing each image in the image time sequence, including grayscale, noise reduction and image enhancement, to improve the feature recognition degree of the image; using edge detection to accurately extract the mycelium area from the preprocessed image, distinguishing the mycelium from the non-target area, and obtaining a binary contour image of the mycelium area; selecting a feature reference point in the binary contour image, comparing images at different time points through image registration technology, calculating the displacement distance of the feature reference point, and combining the time interval of acquisition to obtain the growth speed of the mycelium; comparing the actually measured mycelium growth speed with the standard growth speed of the edible mushroom variety at the set temperature, and calculating the growth speed compliance through a preset formula; The out-mushroom uniformity comprises: acquiring multiple edible mushroom growth images of the same cultivation unit collected in time sequence, the collection range needs to cover the entire cultivation unit to avoid missing the edge area; using a pre-trained target detection neural network model to identify the processed edible mushroom growth image, locate each fruiting body in the image, and generate its minimum bounding box; using an instance segmentation algorithm to segment the pixel area of each fruiting body from the processed edible mushroom growth image; for each detected fruiting body, calculate its morphological characteristics, including: calculating its pixel area, calculating the aspect ratio of its minimum bounding box, fitting the minimum circumscribed circle of its cap and calculating the diameter; calculate the statistical dispersion degree of all fruiting bodies in the above morphological characteristics, respectively obtaining the area dispersion coefficient, the aspect ratio dispersion coefficient and the cap diameter dispersion coefficient; calculate the spatial distribution uniformity of all fruiting bodies in the cultivation unit image, divide the image into several grids, count the number of fruiting bodies in each grid, and calculate the standard deviation of the number of fruiting bodies in all grids, which is used to represent the uniformity of spatial distribution; normalize the area dispersion coefficient, the aspect ratio dispersion coefficient, the cap diameter dispersion coefficient and the spatial distribution standard deviation; through a weighted fusion algorithm, the above four normalized characteristic values are fused into a comprehensive score, which is the out-mushroom uniformity; The edible mushroom growth compliance degree based on the mycelium growth speed compliance degree and the out-mushroom uniformity comprises: normalizing the calculated mycelium growth speed compliance degree and out-mushroom uniformity, so that they are in the same numerical dimension range, and respectively assigning weight values; using a weighted geometric mean algorithm, the normalized mycelium growth speed compliance degree, the out-mushroom uniformity and their corresponding weight values are fused to calculate the final edible mushroom growth compliance degree.

2. The intelligent temperature control method for edible mushroom cultivation growth chamber according to claim 1, characterized in that: temperature sensors are arranged at the four corners of the upper and lower end faces of each layer of the three-dimensional layered cultivation shelf, and the temperature sensors collect temperature data at the corresponding positions in real time at a preset sampling frequency; each layer is equipped with an image acquisition device that can slide back and forth in the horizontal direction to obtain edible mushroom growth images of each cultivation unit.

3. The intelligent temperature control method for an edible mushroom cultivation growth chamber according to claim 1, wherein: The specific way of analyzing whether the temperature of each layer of cultivation shelf is abnormal comprises: obtaining all temperature sampling data of the target layer within a preset time window to form a temperature time series data sequence of the layer; performing filtering and denoising processing on the temperature time series data sequence, and calculating a smoothed temperature change curve using a moving average method; calculating the average temperature value of the smoothed temperature change curve, and taking the average temperature value as a reference to calculate the instantaneous offset amplitude of each sampling point temperature; counting the number of sampling points whose instantaneous offset amplitude exceeds a preset amplitude threshold within the time window, and calculating the proportion of the number of sampling points in the total number of sampling points, denoted as the instantaneous overrun proportion; Calculate the standard deviation of the temperature data in the time window, and establish a comprehensive evaluation function with the instantaneous over-limit ratio and the standard deviation as inputs, and output a comprehensive abnormality index value; Compare the comprehensive abnormality index value with a preset threshold value, and if it exceeds the preset threshold value, determine that the layer frame temperature is abnormal; Screen out all layer frames determined to be temperature abnormal, and sort them according to their comprehensive abnormality index values to determine the temperature control priority.

4. The intelligent temperature control method of claim 3, wherein: The method for determining the temperature control priority comprises: Sort all the screened temperature abnormal layer frames in descending order according to their comprehensive temperature abnormality index values to generate a first priority sequence; If there are layer frames with the same comprehensive abnormality index value in the first priority sequence, then sort them again according to their heights in the vertical layered cultivation frame, and preferentially control the layer frames with higher heights.

5. The intelligent temperature control method of claim 1, wherein: The preset temperature adjustment rule comprises: Obtain the real-time temperature values of the monitoring points in the up-down and front-back directions of the temperature abnormal layer frame, and construct a two-dimensional temperature distribution map of the layer frame; Calculate the gradient of the two-dimensional temperature distribution map to obtain the maximum direction and maximum amplitude of the temperature spatial variation rate, and the maximum direction indicates the direction of the high-temperature area in the layer frame plane; Convert the maximum direction from the layer frame coordinate system to the deflection coordinate system of the guide vane, and calculate the target deflection direction of the guide vane for eliminating the temperature unevenness; Input the maximum amplitude of the temperature spatial variation rate into a predefined nonlinear function, which defines the mapping relationship between the gradient amplitude and the deflection angle of the guide vane, and the output is the target deflection angle of the guide vane; Combine the target deflection direction and the target deflection angle to generate a guide vane control instruction to drive the guide vane to perform corresponding actions to make the cold air flow preferentially to the high-temperature area.

6. The intelligent temperature control method of claim 5, wherein: The gradient calculation of the two-dimensional temperature distribution map to obtain the maximum direction and maximum amplitude of the temperature spatial variation rate comprises: The layer plane is discretized into a two-dimensional temperature matrix consisting of grid points; calculating the temperature gradient component in the direction of the line connecting the two grid points in the two-dimensional temperature matrix for each grid point in the two-dimensional temperature matrix the direction of the line connecting the two grid points in the two-dimensional temperature matrix the direction of the line connecting the two grid points in the two-dimensional temperature matrix According to the gradient components, calculate the gradient amplitude and gradient direction of each grid point; Iterate through the gradient amplitudes of all grid points to find the maximum gradient amplitude and its grid point coordinates; Extract the gradient direction corresponding to the coordinate point as the maximum direction of the temperature spatial variation rate.

7. The intelligent temperature control method of claim 1, wherein: The triggering of the warning operation comprises: Compare the recalculated edible mushroom growth compliance value with a preset compliance threshold value; If the recalculated compliance value is lower than the preset compliance threshold value and has exceeded the lower limit value that the edible mushrooms can tolerate during normal growth, trigger a high-level warning operation; If the recalculated edible mushroom growth compliance value does not reach the preset compliance threshold value, but the compliance value is within the critical interval of the preset compliance threshold value, trigger a low-level warning operation; The warning operation includes a warning report containing the abnormal cultivation unit position number, image data, and growth compliance calculation results; Send the warning report to a remote monitoring terminal.

Images