Carbon cycle time sequence dynamic regulation and control method for edible mushroom and hydroponic vegetable planting bin

Abstract

The invention relates to a carbon cycle time sequence dynamic regulation and control method for an edible mushroom and hydroponic vegetable planting bin. Compared with the prior art, the defects that an edible mushroom and hydroponic vegetable composite planting system is low in carbon resource utilization rate and extensive in regulation and control and lacks a systematic solution are overcome. The method comprises the following steps: determining a regulation and control period of the edible mushrooms and the hydroponic vegetables; collecting and preprocessing carbon dioxide time sequence data of edible fungus and hydroponic vegetable groups; constructing a time sequence carbon assimilation rate model of the hydroponic vegetables based on an improved raccoon optimization algorithm; constructing an edible fungus segmented time sequence respiration rate model; constructing an edible mushroom and hydroponic vegetable planting bin carbon cycle time sequence dynamic regulation and control model; and dynamically regulating and controlling the carbon cycle time sequence of the planting bin. The invention innovatively provides a group carbon cycle time sequence dynamic regulation and control method and system by fusing growth characteristics of respiration of edible mushrooms and photosynthesis of hydroponic vegetables, and realizes accurate supply and demand matching in a mushroom and vegetable symbiotic system.

Description

Technical Field

[0001] The present invention relates to the technical field of planting warehouses, and specifically to a method for dynamically regulating the carbon cycle sequence of an edible mushroom and hydroponic vegetable planting warehouse. Background Art

[0002] Crops such as edible mushrooms and hydroponic vegetables are usually planted in a separated mode. In this mode, the produced during the growth process of edible mushrooms cannot be effectively utilized and is directly discharged into the atmosphere, which not only causes waste of resources, but also exacerbates the climate crisis due to the accumulation of greenhouse gases. Moreover, the locally high concentration of will also affect the growth and development of the edible mushrooms themselves, seriously affecting the yield and quality.

[0003] As a common green plant, hydroponic vegetables highly depend on photosynthesis during their growth process. During photosynthesis, they need to absorb carbon dioxide to synthesize organic substances to achieve growth and development. The supply of emitted by edible mushrooms and the demand of hydroponic vegetables for happen to form a potential supply-demand relationship, which is expected to achieve near-zero carbon. Through a series of effective carbon emission reduction measures and carbon absorption means, the emissions and absorption amounts of are basically balanced within a certain period of time, which provides a new idea for solving the resource utilization and environmental problems in the current agricultural field. Although the concept of symbiotic cultivation of vegetables and edible mushrooms has been mentioned, most of these studies only stay at the conceptual level, simply expounding the possibility and potential advantages of the symbiosis of mushrooms and vegetables, and have not deeply studied how to give specific optimal planting ratios and dynamic allocation methods for the population carbon cycle between edible mushroom sticks and hydroponic vegetables in an actual symbiotic system.

[0004] The population photosynthetic characteristics of hydroponic vegetables change non-linearly dynamically with multiple variables such as light intensity and growth stage. Existing traditional methods mostly use static parameter fitting and cannot capture the instantaneous photosynthetic response and temporal carbon demand fluctuations under different photoperiods. However, accurately adapting to the changes in the photosynthetic characteristics of vegetables throughout the growth cycle and providing an efficient and stable algorithm support for the intelligent environmental control of vegetable hydroponics are important contents of the dynamic regulation of the carbon cycle in the planting warehouse. At the same time, it is found in actual applications that there is still a lack of accurate measurement and analysis of key data such as the amount of carbon dioxide produced by mushrooms at different growth stages and the amount of carbon dioxide required by the whole plant of vegetables, and it is even more impossible to provide a scientific basis for determining a reasonable cultivation ratio, resulting in difficult to achieve precise symbiotic cultivation management of mushrooms and vegetables in actual applications.

[0005] Therefore, how to research a temporal dynamic allocation method for low-carbon symbiotic cultivation of mushrooms and vegetables has become an urgent technical problem to be solved. Summary of the Invention

[0006] The object of the present invention is to solve the defects of low carbon resource utilization rate, extensive regulation, and lack of systematic solutions in the existing composite planting system of edible fungi and hydroponic vegetables, and to provide a method for dynamically regulating the carbon cycle time sequence of the planting bin of edible fungi and hydroponic vegetables to solve the above problems.

[0007] To achieve the above object, the technical solution of the present invention is as follows:

[0008] A method for dynamically regulating the carbon cycle time sequence of the planting bin of edible fungi and hydroponic vegetables, comprising the following steps:

[0009] Determination of the regulation cycle of edible fungi and hydroponic vegetables: Coupling and calibrating the time axis of the fruiting body period of edible fungi with the growth time sequence of hydroponic vegetables to determine the complete regulation cycle of the planting bin of edible fungi and hydroponic vegetables ;

[0010] Collection and preprocessing of time sequence data of carbon dioxide in groups of edible fungi and hydroponic vegetables;

[0011] Constructing a time sequence carbon assimilation rate model of hydroponic vegetables based on an improved raccoon optimization algorithm;

[0012] Constructing a segmented time sequence respiration rate model of edible fungi;

[0013] Constructing a dynamic regulation model of the carbon cycle time sequence of the planting bin of edible fungi and hydroponic vegetables;

[0014] Dynamic regulation of the carbon cycle time sequence of the planting bin: Incorporating the dynamic regulation model of the carbon cycle time sequence of the planting bin of edible fungi and hydroponic vegetables into the planting bin system to achieve dynamic regulation of the carbon cycle time sequence.

[0015] The determination of the regulation cycle of the edible fungi and hydroponic vegetables includes the following steps:

[0016] Respectively measure the growth cycles of edible fungi and hydroponic vegetables. Let the single crop growth cycle of the edible fungi stick be days, and it grows crops under suitable conditions. Then the growth cycle of the edible fungi , and the growth cycle of the hydroponic vegetables is days;

[0017] Coupling and calculating the complete regulation cycle of the planting bin of edible fungi and hydroponic vegetables ,

[0018] is the total cycle of co-cultivation of edible fungi and hydroponic vegetables, is to calculate the least common multiple,

[0019] Then the number of batches of fungus sticks planted during the entire growth cycle , and the number of batches of hydroponic vegetables planted , ensure they are harvested together.

[0020] The collection and preprocessing of the time-series data of carbon dioxide in the edible mushroom and hydroponic vegetable populations are as follows: Obtain the time-series release dataset of the respiratory action during the fruiting body growth stage of the edible mushroom and the population carbon assimilation time-series demand dataset under different light intensities during the entire growth period of the hydroponic vegetable. For abnormal data, use the method for denoising processing; it includes the following steps:

[0021] Using the edible mushroom population carbon flux monitoring device, under the condition of a suitable and stable microenvironment, through high-precision sensors, continuously monitor the respiratory action of a single edible mushroom stick during the fruiting body growth stage to obtain the time-series release dataset ;

[0022] Using the plant population carbon assimilation rate measurement device, measure the population carbon assimilation time-series demand dataset of the whole hydroponic vegetable plant under different light intensities during the entire growth cycle ;

[0023] Adopt the method to process the time-series release dataset and the population carbon assimilation time-series demand dataset for outliers.

[0024] Suppose the time-series release dataset contains values, denoted as , , …, , where represents the th data in this dataset;

[0025] Suppose the population carbon assimilation time-series demand dataset contains values, denoted as , , …, , where represents the th data in this dataset. Calculate the mean of the time-series release dataset and the population carbon assimilation time-series demand dataset and the standard deviation and ,

[0026] ,

[0027] ,

[0028] Among them, 、 respectively represent the mean and standard deviation of the time-series release data set , 、 respectively represent the mean and standard deviation of the population carbon assimilation time-series demand data set , 、 are the summation indices, indicating the a-th data in the time-series release data set ; indicates the b-th data in the population carbon assimilation time-series demand data set ;

[0029] Then calculate the Z value of each data point:

[0030] , where represents the Z value of the a-th data point in the time-series release data set , represents the Z value of the b-th data point in the population carbon assimilation time-series demand data set ;

[0031] Set the threshold > 3 and > 3, and remove the outliers beyond the threshold;

[0032] Time series change rate calculation:

[0033] Divide the time-series release data set , the population carbon assimilation time-series demand data set by = 60 s intervals to calculate the concentration change rate:

[0034] ,

[0035] ,

[0036] where: and are both in ppm / min, is the time in seconds, represents the concentration measured by the carbon flux monitoring device for the edible mushroom population at time, represents the concentration measured by the carbon flux monitoring device for the edible mushroom population at time, represents the concentration measured by the carbon assimilation rate measurement device for the plant population at time, represents the concentration measured by the carbon assimilation rate measurement device for the plant population at time, is the release rate of edible mushrooms per minute per minute, is the absorption rate of hydroponic vegetables per minute per minute;

[0037] By calculating the rate of change of the concentration of edible mushrooms and hydroponic vegetables at multiple time periods per day during the planting cycle, calculate the average release rate of edible mushrooms per minute per day during the planting cycle and the average absorption rate of hydroponic vegetables under different light intensities and the average absorption rate of hydroponic vegetables under different light intensities per minute, which are respectively expressed as:

[0038] ,

[0039] ,

[0040] where N is the number of measurements at multiple effective time periods in a day, and i is the summation index, is the number of days of edible mushroom cultivation, is the number of days of hydroponic vegetable cultivation, is the release rate of edible mushrooms per minute on the day, is the absorption rate of hydroponic vegetables per minute on the day under different light intensities per minute, is the growth cycle of edible mushrooms, is the growth cycle of hydroponic vegetables;

[0041] Combined with the effective volume of the carbon flux monitoring device for the edible mushroom population calculate the release rate in days and generate a time series dataset of the release rate of the respiratory action of edible mushrooms , and the specific calculation method is:

[0042] ​​ ,

[0043] Among them, represents the respiration rate of edible fungi on the th day, with the unit of m / day, 3 and the unit of is m 3 ;

[0044] Let the daylight duration of the photoperiod for hydroponic vegetable cultivation be hours, and the night duration be hours. Combining with the effective volume of the plant population carbon assimilation rate measurement device, it is converted into the absorption rate in days, generating a time series dataset of the carbon assimilation rate of hydroponic vegetables , , and the calculation method is:

[0045] ,

[0046] Among them, represents the carbon assimilation rate of hydroponic vegetables on the th day, with the unit of m 3 / day, and the unit of 3 is m

[0047] The construction of the time series carbon assimilation rate model of hydroponic vegetables based on the improved raccoon optimization algorithm includes the following steps:

[0048] Divide the time series dataset of the carbon assimilation rate of hydroponic vegetables into a training set and a test set , where the number of days of hydroponic vegetable cultivation and light intensity are used as features, and the carbon assimilation rate is used as the label;

[0049] Use the method to standardize the data:

[0050] ,

[0051] ,

[0052] ,

[0053] Among them, , , are the number of days of hydroponic vegetable cultivation, light intensity, and carbon assimilation rate respectively, , , are the standardized hydroponic vegetable planting days, standardized light intensity, and standardized carbon assimilation rate, and are respectively the mean and standard deviation of and are respectively the mean and standard deviation of and is the mean and standard deviation of The unit of ;

[0054] Set up the relevant vector machine model,

[0055] Select the kernel function of the relevant vector machine model as the Gaussian kernel function ,

[0056] where and are the standardized feature vectors. The improved raccoon optimization algorithm is used to optimize the kernel width parameter γ and noise precision parameter α of the relevant vector machine model;

[0057] Set the range of the kernel width parameter γ of the relevant vector machine model as γ ∈ [10 -3 , 1], and the range of the noise precision parameter α as α ∈ [10 -3 , 1]. Set the population size N of the improved raccoon optimization algorithm to 20, and the maximum number of iterations is set to 50 times. The fitness function F is defined as the mean squared error of the test set of the relevant vector machine model, and the calculation formula is:

[0058] ,

[0059] where is the number of samples in the test set, m represents the m-th sample in the test set, is the actual carbon assimilation rate value of the m-th sample in the test set, is the predicted value of the model for the m-th sample;

[0060] Initialize the population and encode the position of each individual. The encoding method is as follows:

[0061] ,

[0062] where and respectively represent the encoded values of the kernel width parameter and noise precision parameter of the relevant vector machine model corresponding to the -th individual in the population, and are respectively the lower bound and upper bound of the kernel width parameter γ, , are respectively the lower and upper bounds of the noise precision parameter α, is a random number uniformly distributed in the interval [0, 1], is the index of an individual in the population, and N is the population size;

[0063] The position vectors of N = 20 raccoon individuals are generated randomly , , and the fitness value of each individual is calculated using the fitness function F as the initial value;

[0064] Design a population size dynamic adjustment module: Calculate the average Euclidean distance of each individual in the solution space, and use the standardized distance variance as the population diversity evaluation index; When the population is initialized or updated each time, the diversity of the current population is evaluated by calculating the average Euclidean distance between the individual positions in the population and the average of all individual positions. When the population diversity is lower than the threshold 0.3 and the population size is lower than the maximum population size 100, update the population: Randomly generate new sample individuals to expand the population size; When the population diversity is higher than the threshold 0.3 and the population size is higher than the minimum population size 30, update the population: Calculate the fitness value of each individual using the fitness function F, eliminate the worst individual, and reduce the population size;

[0065] On the time series dataset of the carbon assimilation rate of hydroponic vegetables, use the encoded γ and α in the training set to train the relevant vector machine model, and use the trained relevant vector machine model to make predictions on the test set Calculate the fitness value through the difference between the prediction result and the true value, traverse the fitness values of all individuals, and select the individual with the minimum fitness value as the current optimal solution. The γ and α parameter combinations corresponding to the optimal solution are respectively denoted as and , and its position vector is denoted as ;

[0066] In the search space, select individuals in the population for position update, and the update formula is as follows:

[0067] ,

[0068] where N is the population size, p is the current iteration number, represents the component of the i-th individual in the j-th dimension at the p-th iteration, represents the component of the i-th individual in the j-th dimension at the (p + 1)-th iteration, represents the component of the optimal individual in the j-th dimension at the p-th iteration, is a random number uniformly distributed in the interval [0, 1], I represents a random integer from the integer set {1, 2}, and i represents the number of the individual selected for position update;

[0069] The fitness value of each individual is calculated using the fitness function F, and a greedy selection is performed once. If the fitness of the updated individual is better than the fitness value before the update, the updated position is retained; otherwise, no position update is performed. The formula for the greedy selection strategy is as follows:

[0070] ,

[0071] where, is the position of the i-th individual at the (p + 1)-th iteration, is the position of the i-th individual at the p-th iteration, and are the fitness values corresponding to the positions of the i-th individual at the (p + 1)-th iteration and the p-th iteration, respectively;

[0072] Design a learning rate adjustment strategy for position adaptive update:

[0073] Construct a composite function based on the iteration number decay factor and the diversity feedback coefficient. The learning rate update formula is:

[0074] ,

[0075] where, is the current iteration number, is the decay factor, is the maximum iteration number, is the number of diversity solution sets, i is the summation variable, is the diversity feedback coefficient, is the population diversity, represents the learning rate at the p-th iteration, represents the learning rate at the (p + 1)-th iteration, is the exponential operation of the natural constant;

[0076] Adopt a dual regulation mechanism to achieve the collaborative optimization of global exploration and local exploitation: adopt a global search mode at the initial stage of iteration. As the number of iterations increases, start the local exploitation mode to accelerate the speed of locating the global optimal neighborhood;

[0077] Select another individual in the population and update the position in the search space,

[0078] First, randomly generate a temporary position vector in the search space, and its calculation formula is:

[0079] ,

[0080] The position update is as follows:

[0081] ,

[0082] ;

[0083] where i represents the number of the individual selected for position update, N is the population size, and respectively represent the lower and upper bounds of the j-th dimension in the search space, is a random number uniformly distributed in the interval [0,1], I represents a random integer from the set of integers {1, 2}, represents the component of the i-th individual in the j-th dimension at the p-th iteration, represents the component of the i-th individual in the j-th dimension at the (p + 1)-th iteration, represents the temporary position vector corresponding fitness value of the solution, is the fitness value corresponding to the position of the i-th individual at the p-th iteration;

[0084] In the search space, simulate the random perturbation mechanism when an individual faces dynamic perturbations,

[0085] When the random perturbation mechanism is triggered, a random position is generated near the current position of each individual, and the formula is as follows:

[0086] ,

[0087] ,

[0088] ,

[0089] where i represents the number of the individual selected for position update, represents the number of iterations, and are the upper and lower bounds of the j-th dimension variable updated with the number of iterations, is the maximum number of iterations;

[0090] Calculate the fitness value of each individual and execute the greedy selection strategy again;

[0091] If the fitness of the updated individual is better than the fitness value before the update, accept the updated position, otherwise do not perform position update;

[0092] Continue the iteration until the preset maximum number of iterations = 50, after the iteration ends, output the optimal solution and the corresponding optimal values , the optimal solution is the optimal combination of the kernel width parameter γ and the noise precision parameter α of the relevance vector machine model. Substitute the optimal parameters and into the relevance vector machine model, and input the training set . The relevance vector machine model fits the non-linear relationship between the input features and the carbon assimilation rate, and outputs the time-series carbon assimilation rate model ;

[0093] Train the time-series carbon assimilation rate model multiple times, and use 5-fold cross-validation to select the model with the smallest average root mean square error and the largest coefficient of determination as the final time-series carbon assimilation rate model .

[0094] The construction of the segmented time-series respiration rate model of edible fungi includes the following steps:

[0095] Based on the respiration rate time-series data set of edible fungi , according to the time ranges corresponding to the fruiting body differentiation stage and the development stage of edible fungi, extract the data corresponding to the stages within the time range. Use the number of days of edible fungi cultivation as the input, and use the respiration rate of edible fungi during the fruiting body differentiation stage and the development stage as the output. Adopt the polynomial regression algorithm to build segmented models for these two stages: release rate , where

[0096] ,

[0097] ,

[0098] where is the time-series respiration rate model for the fruiting body differentiation stage of edible fungi, is the time-series respiration rate model for the fruiting body development stage of edible fungi, is the cycle of the fruiting body differentiation stage of edible fungi, is the growth cycle of edible fungi, is the number of days of edible fungi cultivation, n and m are the orders of the polynomials, , , …, , , , …, are the coefficients to be determined;

[0099] Fit the coefficients to be determined by the least squares method to form the segmented time-series respiration rate model of edible fungi, and its formula is as follows:

[0100] 。

[0101] The construction of the carbon cycle time-sequence dynamic regulation model for edible fungi and hydroponic vegetables planting bins includes the following steps:

[0102] Obtaining the optimal planting ratio of edible fungi and hydroponic vegetables: discretize the complete regulation period T of the edible fungi and hydroponic vegetables planting bin into k time nodes to form a time series , where k is the number of time nodes, represents the kth time node in the time series, , , is the time interval between adjacent time nodes, represents the (c + 1)th time node in the time series, represents the cth time node in the time series,

[0103] Use light intensity , the number of hydroponic vegetables planted , the segmented time-sequence respiration rate model of edible fungi and the time-sequence carbon assimilation rate model of hydroponic vegetables as the input of the carbon cycle time-sequence dynamic regulation model for edible fungi and hydroponic vegetables planting bins. On the basis of satisfying carbon metabolism balance, construct an optimization objective function to determine the optimal number of edible fungi planted , and the optimization objective function is:

[0104] ,

[0105] Among them, and are the numbers of edible fungi and hydroponic vegetables planted respectively, is the number of batches of edible fungi stick planting, is the number of batches of hydroponic vegetables planting, is the time node, is the respiration release rate of edible fungi at the cth time node, is the carbon assimilation rate of hydroponic vegetables at the cth time node, represents the ceiling operation on the number of edible fungi planted , is the regularization coefficient, is about penalty function, controls the intensity of the penalty,

[0106] After obtaining the optimal through the optimization objective function, the optimal planting ratio of edible fungi and hydroponic vegetables is obtained

[0107] The planting bin adopts the carbon cycle time-sequence dynamic regulation model of edible fungi and hydroponic vegetables to carry out environmental dynamic regulation under the optimal planting ratio of edible fungi and hydroponic vegetables. The formula of the carbon cycle time-sequence dynamic regulation model of edible fungi and hydroponic vegetables planting bin is as follows:

[0108] ,

[0109] Among them, is the target value of the carbon cycle time-sequence dynamic regulation of the planting bin, is the concentration controlled by the adaptive cycle control algorithm in the planting bin, concentration, is the concentration gain generated by the storage and release dynamic control mechanism of the planting bin in the planting bin, concentration gain, is the real-time time;

[0110] The adaptive cycle control algorithm controls the power of the air exchange cycle fan between the edible fungi planting bin and the hydroponic vegetable planting bin through the PID algorithm, that is, the realization of the PWM duty cycle. According to the carbon cycle balance relationship of edible fungi and hydroponic vegetables in time sequence, the change factor of the time-sequence Kp proportionality coefficient is calculated by integrating the segmented time-sequence respiration rate model of edible fungi and the time-sequence carbon assimilation rate model of hydroponic vegetables, and then an adaptive PID regulation algorithm is established. Its formula is as follows:

[0111] ,

[0112] Among them, is the target PWM duty cycle of the air exchange cycle fan of the planting bin, is the current time of the system, is the daytime of the photoperiod, is the co-planting days of edible fungi and hydroponic vegetables in the planting bin, is the day's carbon assimilation rate of hydroponic vegetables and the respiration of edible fungi release rate ratio, , is the differential coefficient of the PID control algorithm, is the real-time concentration difference between the edible fungi and hydroponic vegetables planting bin;

[0113] The constraint conditions of the adaptive cycle control algorithm are: ,

[0114] ​ , are the real-time concentrations of edible fungi and hydroponic vegetable growing chambers, , are the upper limits of the appropriate concentrations of edible fungi and hydroponic vegetable growing chambers;

[0115] Concentration gain is obtained through the storage and release dynamic control mechanism of the growing chamber , and the control logic is as follows: obtained, The control logic is as follows:

[0116] ,

[0117] Among them, is the storage dynamic control mechanism of the growing chamber , is the release dynamic control mechanism of the growing chamber ;

[0118] The control logic of the release dynamic control mechanism of the growing chamber is as follows: The control logic of the release dynamic control mechanism of the growing chamber is as follows:

[0119] ,

[0120] Among them, is the lower limit of the appropriate concentration in the hydroponic vegetable growing chamber, is the night duration;

[0121] The control logic constraint condition of the release dynamic control mechanism of the growing chamber is: The control logic constraint condition of the release dynamic control mechanism of the growing chamber is:

[0122] .

[0123] The dynamic regulation of the carbon cycle time sequence of the growing chamber includes the following steps:

[0124] Substitute the parameter values into the adaptive cycle control algorithm. The parameter values include the current system time , the daytime of the light cycle , the carbon assimilation rate of hydroponic vegetables on the th day and the release rate of the respiratory action of edible fungi as well as the real-time concentration difference The constraint conditions of the adaptive cycle control algorithm are used to calculate the target PWM duty cycle of the air exchange circulation fan in real time, control the power of the air exchange circulation fan, and achieve timing adaptive regulation, so that the concentrations in the edible mushroom and hydroponic vegetable growing chambers

[0125] growing chamber Application of the storage and release dynamic control mechanism:

[0126] growing chamber Storage dynamic control mechanism: The concentration in the edible mushroom growing chamber is monitored in real time. When > the storage dynamic control mechanism of the growing chamber is triggered . The system starts the compressor for air compression operation. When < < the compressor stops working. Among them, is the upper limit of the suitable concentration in the edible mushroom growing chamber, is the lower limit of the suitable concentration in the edible mushroom growing chamber;

[0127] growing chamber Release dynamic control mechanism: Monitor the concentration , the current system time and value in the hydroponic vegetable growing chamber;

[0128] When and the constraint conditions are met, the release dynamic control mechanism of the growing chamber is triggered . The system starts the compressor for compressed air release operation. When it is monitored that < the compressor stops working. Among them, is the upper limit of the suitable concentration in the hydroponic vegetable growing chamber, is the lower limit of the suitable concentration in the hydroponic vegetable growing chamber, is the total number of days of co - planting edible mushrooms and hydroponic vegetables in the growing chamber, is the carbon assimilation rate of hydroponic vegetables on the th day and the release rate of the respiratory action of edible mushrooms .

[0129] The planting chamber includes the following: Data acquisition subsystem: Sensors are set to monitor in real time the growth environment parameters of edible fungi and hydroponic vegetables, as well as the key data of concentration, and transmit the collected data to the data fusion subsystem. The sensors include temperature and humidity sensors, sensors, sensors, light intensity sensors, solution value and value measurement sensors;

[0130] Data fusion subsystem: By coupling and calibrating the time axes of the fruiting body stage of edible fungi and the growth time sequence of hydroponic vegetables, obtain the total regulation cycle of the planting chamber and the number of batches of edible fungi sticks planted during the entire growth cycle and the number of batches of hydroponic vegetables planted ; Through the embedded segmented time sequence respiration rate model of edible fungi and the time sequence carbon assimilation rate model of hydroponic vegetables, automatically match and generate the optimal planting ratio of edible fungi and hydroponic vegetables and the day carbon assimilation rate of hydroponic vegetables and the respiration of edible fungi release rate ratio to obtain a dynamic regulation method for the carbon cycle time sequence of the planting chamber for the current planting varieties; The data fusion subsystem combines the real-time environmental parameter values transmitted by the data acquisition subsystem and sends regulation instructions to the dynamic feedback control subsystem;

[0131] Dynamic feedback control subsystem: Receive the regulation instructions transmitted by the data fusion subsystem, drive the actuators to cooperate, and regulate the environmental parameters in the planting chamber;

[0132] Cultivation box, air compression storage device and control device. An airtight partition is slidably arranged in the cultivation box, and the airtight partition divides the cultivation box into an edible fungi planting chamber and a hydroponic vegetable planting chamber. Ventilation openings are respectively arranged on the cultivation box at the edible fungi planting chamber and the hydroponic vegetable planting chamber. An air exchange circulation fan is arranged above the airtight partition. The air compression storage device includes a compressor and an intake pipe and an outlet pipe arranged on both sides of the compressor. The intake pipe is connected to the inside of the edible fungi planting chamber, and the outlet pipe is connected to the inside of the hydroponic vegetable planting chamber. The ventilation openings, intake pipe and outlet pipe are all controlled to open and close by electromagnetic valves. The air exchange circulation fan is controlled to open and close and adjust the wind speed by a motor. The control device is used to receive sensor signals and control the compressor, electromagnetic valves and air exchange circulation fan to work according to the preset dynamic regulation method of the carbon cycle time sequence of the planting chamber.

[0133] A computer-readable storage medium stores a computer program thereon. When the computer program is executed by a processor, a method for dynamically regulating the carbon cycle timing of an edible mushroom and a hydroponic vegetable planting chamber is implemented. A computer device includes a memory, a processor, and a computer program stored on the memory and running on the processor. When the processor executes, a method for dynamically regulating the carbon cycle timing of an edible mushroom and a hydroponic vegetable planting chamber is implemented.

[0134] Beneficial effects

[0135] Compared with the prior art, the method for dynamically regulating the carbon cycle timing of an edible mushroom and a hydroponic vegetable planting chamber of the present invention innovatively proposes a method and system for dynamically regulating the group carbon cycle timing by integrating the growth characteristics of the respiration of edible mushrooms and the photosynthesis of hydroponic vegetables, and realizes precise supply-demand matching in the mushroom-vegetable symbiotic system, significantly improving the utilization efficiency of carbon resources and effectively reducing carbon emissions.

[0136] The present invention uses a hybrid intelligent optimization algorithm of a relevant vector machine model optimized by an improved raccoon optimization algorithm to establish a hydroponic vegetable temporal carbon assimilation rate model. At the same time, combined with the temporal respiration rate model of edible mushrooms constructed by polynomial regression analysis, a multi-model fusion edible mushroom-hydroponic vegetable adaptive dynamic regulation model is constructed, which has strong scalability and can adapt to different combinations of mushroom species and hydroponic vegetable categories. By designing a complete model integration system solution, full-cycle precise matching of cumulative release and absorption amounts is achieved, filling the technical gap in systematic carbon cycle regulation solutions.

[0137] The present invention uses a dynamic regulation system. Based on the embedded multi-model and the constraint conditions of the inhibition concentration of edible mushroom growth and the suitable concentration of hydroponic vegetables to regulate the air exchange circulation fan speed and the opening degree of the air path valve, ensuring that the concentration in the system is always in the optimal range. Compared with the traditional static threshold control method, the present invention significantly improves the regulation accuracy and response speed through the fusion of multi-source environmental data and the closed-loop feedback mechanism. Description of the drawings

[0138] Figure 1 is the method sequence diagram of the present invention;

[0139] Figure 2 is the temporal carbon assimilation rate model diagram of a single hydroponic lettuce during a growth cycle in an embodiment of the present invention;

[0140] Figure 3 is the temporal respiration rate model diagram of a single Pleurotus ostreatus mushroom stick during the fruiting body period in an embodiment of the present invention;

[0141] Figure 4 is the total planting cycle of the planting bin system in the embodiment of the present invention Determine the light intensity for hydroponic lettuce planting within is 150 Coupling model curve graph of sequential absorption and release Specific implementation manner

[0142] To have a further understanding and recognition of the structural features and achieved effects of the present invention, the following is a detailed description with preferred embodiments and accompanying drawings:

[0143] As Figure 1 shown, a method for sequential dynamic regulation of the carbon cycle in a planting bin for edible fungi and hydroponic vegetables according to the present invention includes the following steps:

[0144] The first step is to determine the regulation cycle of edible fungi and hydroponic vegetables: couple and calibrate the time axes of the fruiting body period of edible fungi and the growth sequence of hydroponic vegetables to determine the complete regulation cycle of the planting bin for edible fungi and hydroponic vegetables . Here, the problem of carbon cycle supply-demand mismatch caused by the asynchronous growth cycles of edible fungi and vegetables in the traditional planting mode is solved. Through sequential phase calibration, precise coordination of the growth stages of edible fungi and vegetables throughout the cycle is achieved, significantly improving the dynamic balance ability within the system

[0145] (1) Measure the growth cycles of edible fungi and hydroponic vegetables respectively. Let the single-crop growth cycle of the edible fungi stick be days, and it grows crops under suitable conditions. Then the growth cycle of edible fungi , and the growth cycle of hydroponic vegetables is days

[0146] (2) Coupling calculation of the complete regulation cycle of the planting bin for edible fungi and hydroponic vegetables ,

[0147] is the total cycle for co-cultivation of edible fungi and hydroponic vegetables is to calculate the least common multiple

[0148] Then the number of batches of edible fungi sticks planted throughout the growth cycle , and the number of batches of hydroponic vegetables planted , ensuring simultaneous harvesting

[0149] Step 2: Collection and preprocessing of time-series carbon dioxide data for edible fungi and hydroponic vegetables. The Z-score method is used to denoise the collected time-series data, solving the problem of outliers caused by environmental interference or equipment errors in the traditional data collection process and providing a high-quality data basis for subsequent model construction.

[0150] Obtain the time-series release dataset of the respiration during the fruiting body growth stage of edible fungi and the time-series demand dataset of the population carbon assimilation of hydroponic vegetables under different light intensities throughout the growth period. Denoise the abnormal data using method; it includes the following steps:

[0151] (1) Using the monitoring device for the carbon flux of the edible fungi population, under the condition of a suitable and stable microenvironment, the respiration of a single edible fungi stick during the fruiting body growth stage is monitored in real time through a high-precision sensor to obtain the time-series release dataset .

[0152] (2) Using the measurement equipment for the population carbon assimilation rate of plants, measure the time-series demand dataset of the population carbon assimilation of the whole hydroponic vegetable plant under different light intensities throughout the growth cycle .

[0153] (3) Use method to process the outliers in the time-series release dataset and the time-series demand dataset of the population carbon assimilation . Let the

[0154] time-series release dataset contain values, denoted as , , …, , , where represents the th data in this dataset;

[0155] Let the time-series demand dataset of the population carbon assimilation contain values, denoted as , , …, , where represents the th data in this dataset. Calculate the time-series release dataset and the Time series demand data set mean value and standard deviation ,

[0156] ,

[0157] ,

[0158] wherein, 、 respectively represent time series release data set mean value, standard deviation, 、 respectively represent the mean value and standard deviation of the population carbon assimilation time series demand data set ; 、 are summation indices, indicating the a-th data in the time series release data set ; indicating the b-th data in the population carbon assimilation time series demand data set ;

[0159] Then calculate the Z value of each data point:

[0160] , wherein, represents the Z value of the a-th data point in the time series release data set , represents the Z value of the b-th data point in the population carbon assimilation time series demand data set ;

[0161] Set the threshold > 3 and > 3, and remove the outliers exceeding the threshold.

[0162] (4)Time series change rate calculation:

[0163] A1) Calculate the concentration change rate of the time series release data set and the population carbon assimilation time series demand data set respectively at intervals of = 60 s:

[0164] ,

[0165] ,

[0166] Wherein: and The units of both are ppm / min, is the time in seconds, represents The concentration measured by the carbon flux monitoring device for the edible mushroom population at represents The concentration measured by the carbon flux monitoring device for the edible mushroom population at represents The concentration measured by the carbon assimilation rate measurement device for the plant population at represents The concentration measured by the carbon assimilation rate measurement device for the plant population at is the release rate of per minute for the edible mushroom, is the absorption rate of per minute for the hydroponic vegetables;

[0167] By calculating the concentration change rate of the edible mushroom and the hydroponic vegetables at multiple time periods per day during the planting cycle, calculate the average release rate of per minute for the edible mushroom per day during the planting cycle and the average absorption rate of per minute for the hydroponic vegetables under different light intensities , which are respectively expressed as:

[0168] ,

[0169] ,

[0170] Wherein, N is the number of measurements at multiple effective time periods in a day, i is the summation index, is the number of days for edible mushroom planting, is the number of days for hydroponic vegetable planting, is the th day's release rate of per minute for the edible mushroom, is the absorption rate of th day's per minute for the hydroponic vegetables under different light intensities, is the growth cycle of the edible mushroom, is the growth cycle of the hydroponic vegetables;

[0171] A2) Combine the effective volume of the carbon flux monitoring device for the edible mushroom population to calculate the Release rate, generating the respiration of edible fungi Release rate time series dataset , and the specific calculation method is as follows:

[0172] ,

[0173] Among them, represents the respiration release rate of edible fungi on the th day, with the unit of m / day, 3 and the unit of is m 3 ;

[0174] A3) Suppose the daytime duration of the light cycle for hydroponic vegetable cultivation is hours, and the nighttime duration is hours. Combining with the effective volume of the plant population carbon assimilation rate measurement device, it is converted into the absorption rate in units of days, generating the time series dataset of the carbon assimilation rate of hydroponic vegetables , and the calculation method is:

[0175] ,

[0176] Among them, represents the carbon assimilation rate of hydroponic vegetables on the th day, with the unit of m 3 / day, and the unit of 3 is m

[0177] The third step is to construct a time series carbon assimilation rate model for hydroponic vegetables based on the improved raccoon optimization algorithm.

[0178] Modeling the carbon assimilation rate of hydroponic vegetables faces significant technical challenges. Firstly, existing technologies mainly focus on single-leaf measurement of vegetables, lacking accurate determination and analysis of key data such as the carbon dioxide demand of the whole plant. Secondly, the time series carbon assimilation rate of hydroponic vegetables is affected by the coupling of multiple variables, showing significant non-linear dynamic change characteristics. Traditional fixed-parameter models are difficult to capture the changing rules of this multi-factor dynamic coupling. In the face of environmental mutations, the model prediction error is large, unable to meet the requirements of precise carbon cycle regulation. Therefore, the present invention proposes a method for modeling the time series carbon assimilation rate of hydroponic vegetables based on the improved raccoon optimization algorithm.

[0179] StandardScaler method is used for standardization to make the contribution weights of different features to the relevant vector machine model consistent, improving the parameter optimization efficiency and the model convergence speed. To further optimize the kernel width parameter γ and the noise precision parameter α of the relevant vector machine model, an improved raccoon optimization algorithm is used to optimize the hyperparameters. Compared with the traditional raccoon optimization algorithm, 1) a dynamic population size adjustment module is designed: the standardized variance of the average Euclidean distance of individuals is calculated to evaluate the population diversity. When the diversity is lower than the threshold of 0.3 and the size is less than 100, new individuals are randomly generated to expand the population to enhance global exploration; when the diversity is higher than the threshold of 0.3 and the size exceeds 30, poor solutions are eliminated to reduce the population to improve the optimization efficiency, thus dynamically balancing the population size and diversity, avoiding premature convergence and taking into account both the search accuracy and the computational efficiency; 2) for the position update of population individuals, a learning rate adjustment strategy for position adaptive update is designed, and a composite function based on the iteration number decay factor and the diversity feedback coefficient is constructed. In the initial stage of iteration, global exploration is carried out with a larger step size to quickly cover different regions of the solution space, avoiding the exploration blind area caused by the random distribution of initial parameters. As the iteration progresses, the learning rate gradually decreases and turns to local fine development, focusing on the neighborhood of the current optimal solution to improve the parameter fine-tuning accuracy. Based on this, a relevant vector machine model is constructed and the final model is selected through 5-fold cross-validation. This algorithm dynamically adjusts the population size, balances global exploration and local development, and avoids premature convergence; adopts a dual position update strategy to improve the convergence speed and parameter optimization accuracy; the optimized relevant vector machine model has better kernel width parameter and noise precision parameter, enhances the non-linear fitting ability for time series data, improves sparsity and generalization ability, reduces the risk of overfitting, and is not sensitive to initial parameters, adapting to the characteristics of different datasets, providing a reliable dynamic model support for carbon cycle regulation.

[0180] (1) Divide the time series dataset of the carbon assimilation rate of hydroponic vegetables into a training set and a test set , where the planting days and light intensity of hydroponic vegetables are used as features, and the carbon assimilation rate is used as the label.

[0181] (2) Use the method to standardize the data:

[0182] ,

[0183] ,

[0184] ,

[0185] Among them, , , are the planting days, light intensity, and carbon assimilation rate of hydroponic vegetables respectively, , , are the standardized hydroponic vegetable planting days, the standardized light intensity, and the standardized carbon assimilation rate respectively. , are respectively the mean and standard deviation of , are respectively the mean and standard deviation of , is the mean and standard deviation of The unit of .

[0186] (3) Set the relevance vector machine model, and select the Gaussian kernel function as the kernel function of the relevance vector machine model ,

[0187] where , are the standardized feature vectors, and the improved raccoon optimization algorithm is used to optimize the kernel width parameter γ and the noise precision parameter α of the relevance vector machine model;

[0188] Set the range of the kernel width parameter γ of the relevance vector machine model as γ ∈ [10 -3 , 1], and the range of the noise precision parameter α as α ∈ [10 -3 , 1]. Set the population size N of the improved raccoon optimization algorithm to 20, and the maximum number of iterations is set to 50 times. The fitness function F is defined as the mean square error of the test set of the relevance vector machine model, and the calculation formula is:

[0189] ,

[0190] where is the number of samples in the test set, m represents the m-th sample in the test set, is the actual carbon assimilation rate value of the m-th sample in the test set, is the predicted value of the model for the m-th sample.

[0191] (4) Initialize the population, and encode the position of each individual. The encoding method is as follows:

[0192] ,

[0193] where , respectively represent the encoding values of the kernel width parameter and the noise precision parameter of the relevance vector machine model corresponding to the -th individual in the population, , are the lower and upper bounds of the nuclear width parameter γ, respectively, and are the lower and upper bounds of the noise precision parameter α, respectively, is a random number uniformly distributed in the interval [0, 1], is the index of an individual in the population, and N is the population size;

[0194] Generate the position vectors of N = 20 raccoon individuals in a random manner , , and calculate the fitness value of each individual using the fitness function F as the initial value.

[0195] (5) Design a population size dynamic adjustment module: Calculate the average Euclidean distance of each individual in the solution space, and use the distance variance after standardization as the population diversity evaluation index; When initializing the population or updating it each time, evaluate the diversity of the current population by calculating the average Euclidean distance between the individual positions in the population and the average of all individual positions. When the population diversity is lower than the threshold 0.3 and the population size is lower than the maximum population size of 100, update the population: Randomly generate new sample individuals to expand the population size; When the population diversity is higher than the threshold 0.3 and the population size is higher than the minimum population size of 30, update the population: Calculate the fitness value of each individual using the fitness function F, eliminate the worst individuals, and reduce the population size.

[0196] (6) On the time series dataset of the carbon assimilation rate of hydroponic vegetables, use the encoded γ and α to train the relevant vector machine model on the training set , and use the trained relevant vector machine model to make predictions on the test set . Calculate the fitness value through the difference between the prediction result and the true value, traverse the fitness values of all individuals, and select the individual with the minimum fitness value as the current optimal solution. The parameter combinations of γ and α corresponding to the optimal solution are denoted as and respectively, and its position vector is denoted as .

[0197] (7) In the search space, select individuals in the population for position update, and the update formula is:

[0198] ,

[0199] where N is the population size, p is the current iteration number, represents the component of the i-th individual in the j-th dimension at the p-th iteration, represents the component of the i-th individual in the j-th dimension at the (p + 1)-th iteration, represents the component of the optimal individual in the j-th dimension at the p-th iteration, is a random number uniformly distributed in the interval [0, 1], I represents a random integer from the integer set {1, 2}, and i represents the number of the individual selected for position update.

[0200] (8) Calculate the fitness value of each individual using the fitness function F, and perform a greedy selection. If the fitness of the updated individual is better than that before the update, the updated position is retained; otherwise, the position is not updated. The greedy selection strategy formula is as follows:

[0201] ,

[0202] where, is the position of the i-th individual at the (p + 1)-th iteration, is the position of the i-th individual at the p-th iteration, and are the fitness values corresponding to the positions of the i-th individual at the (p + 1)-th iteration and the p-th iteration, respectively.

[0203] (9) Design a learning rate adjustment strategy for position adaptive update:

[0204] Construct a composite function based on the iteration number decay factor and the diversity feedback coefficient. The learning rate update formula is:

[0205] ,

[0206] where, is the current iteration number, is the decay factor, is the maximum iteration number, is the number of diversity solution sets, i is the summation variable, is the diversity feedback coefficient, is the population diversity, represents the learning rate at the p-th iteration, represents the learning rate at the (p + 1)-th iteration; a dual regulation mechanism is adopted to achieve the collaborative optimization of global exploration and local exploitation: a global search mode is adopted at the initial stage of iteration, and as the iteration number increases, the local exploitation mode is started to accelerate the speed of locating the global optimal neighborhood.

[0207] (10) Select another individual in the population and update the position in the search space.

[0208] First, randomly generate a temporary position vector in the search space, and its calculation formula is:

[0209] ,

[0210] The position is updated as follows:

[0211] ,

[0212] ;

[0213] where \(i\) represents the number of the individual selected for position update, \(N\) is the population size, and represent the lower and upper bounds of the \(j\)-th dimension in the search space respectively, is a random number uniformly distributed in the interval \([0, 1]\), \(I\) represents a random integer from the set of integers \(\{1, 2\}\), represents the component of the \(i\)-th individual in the \(j\)-th dimension at the \(p\)-th iteration, represents the component of the \(i\)-th individual in the \(j\)-th dimension at the \((p + 1)\)-th iteration, represents the temporary position vector corresponding to the fitness value of the solution, is the fitness value corresponding to the position of the \(i\)-th individual at the \(p\)-th iteration.

[0214] (11) In the search space, simulate the random perturbation mechanism of individuals when facing dynamic perturbations,

[0215] When the random perturbation mechanism is triggered, a random position is generated near the current position of each individual, and the formula is as follows:

[0216] ,

[0217] ,

[0218] ,

[0219] where \(i\) represents the number of the individual selected for position update, represents the number of iterations, and are the upper and lower bounds of the \(j\)-th variable updated with the number of iterations, is the maximum number of iterations.

[0220] (12) Calculate the fitness value of each individual and execute the greedy selection strategy again;

[0221] If the fitness of the updated individual is better than the fitness value before the update, accept the updated position, otherwise do not perform position update.

[0222] (13) Keep iterating until the preset maximum number of iterations = 50. After the iteration ends, output the optimal solution and the corresponding optimal value , the optimal solution is the optimal combination of the kernel width parameter γ and the noise precision parameter α of the relevance vector machine model. Substitute the optimal parameters and into the relevance vector machine model, and input the training set . The relevance vector machine model fits the non-linear relationship between the input features and the carbon assimilation rate, and outputs the time-series carbon assimilation rate model .

[0223] (14) Train the time-series carbon assimilation rate model multiple times, and use 5-fold cross-validation to select the model with the smallest average root mean square error and the largest determination coefficient as the final time-series carbon assimilation rate model .

[0224] Step 4: Construct a segmented time-series respiration rate model for edible fungi. The polynomial regression analysis method with time as the independent variable and respiration rate as the dependent variable can accurately describe the dynamic characteristics of the fruiting body growth stage of edible fungi. By combining segmented modeling and continuous fitting, the prediction accuracy and applicability of the model are significantly improved. It can provide a quantitative basis for optimizing the carbon cycle of the mushroom-vegetable symbiotic system; in terms of scalability, the model framework can be adapted to different mushroom species, providing a methodological reference for carbon cycle research in agricultural edible mushroom cultivation.

[0225] (1) Based on the respiration rate time-series dataset of the release rate , according to the time range corresponding to the differentiation stage and development stage of the fruiting body of edible fungi, extract the data corresponding to the stages within the time range. Use the number of days of edible mushroom cultivation as the input, and use the respiration rate of edible fungi during the differentiation stage and development stage of the fruiting body release rate as the output, and use the polynomial regression algorithm to perform segmented modeling for these two stages:

[0226] ,

[0227] ,

[0228] where is the time-series respiration rate model for the differentiation stage of the fruiting body of edible fungi, is the time-series respiration rate model for the development stage of the fruiting body of edible fungi, is the period of the differentiation stage of the fruiting body of edible fungi, is the growth period of edible fungi, is the number of days of edible mushroom cultivation, n and m are the orders of the polynomials, , , …, 、 , , …, are coefficients to be determined.

[0229] (2)Obtain the coefficients to be determined by least squares fitting to form a piecewise time-series respiration rate model for edible fungi , and its formula is as follows:

[0230] .

[0231] Step 5: Construct a piecewise time-series dynamic regulation model for the carbon cycle in the cultivation chambers of edible fungi and hydroponic vegetables.

[0232] (1)Obtain the optimal planting ratio of edible fungi and hydroponic vegetables: Discretize the complete regulation period T of the cultivation chambers of edible fungi and hydroponic vegetables into k time nodes to form a time series , where k is the number of time nodes, represents the kth time node in the time series, , , is the time interval between adjacent time nodes, represents the (c + 1)th time node in the time series, represents the cth time node in the time series,

[0233] Use light intensity , the number of hydroponic vegetables planted , the piecewise time-series respiration rate model of edible fungi and the time-series carbon assimilation rate model of hydroponic vegetables as the inputs of the piecewise time-series dynamic regulation model for the carbon cycle in the cultivation chambers of edible fungi and hydroponic vegetables. On the basis of meeting the carbon metabolic balance, construct an optimization objective function to determine the optimal number of edible fungi planted , and the optimization objective function is:

[0234] ,

[0235] where, and are the numbers of edible fungi and hydroponic vegetables planted respectively, is the number of batches of edible fungus sticks planted, is the number of batches of hydroponic vegetables planted, is the time node, is the respiration release rate of edible fungi at the cth time node, is the carbon assimilation rate of hydroponic vegetables at the cth time node, represents the ceiling operation on the number of edible fungi planted , is the regularization coefficient, is about the penalty function to control the intensity of the penalty

[0236] After obtaining the optimal solution by optimizing the objective function, the optimal planting ratio of edible fungi and hydroponic vegetables is obtained .

[0237] (2)Under the optimal planting ratio of edible fungi and hydroponic vegetables The environmental dynamic regulation is carried out by using the carbon cycle time-series dynamic regulation model of the edible fungi and hydroponic vegetables planting warehouse. The formula of the carbon cycle time-series dynamic regulation model of the edible fungi and hydroponic vegetables planting warehouse is as follows:

[0238] ,

[0239] where is the target value of the carbon cycle time-series dynamic regulation of the planting warehouse is the concentration controlled by the adaptive cycle control algorithm in the planting warehouse , is the concentration gain generated by the storage and release dynamic control mechanism of the planting warehouse in the planting warehouse , is the real-time time

[0240] (3) The adaptive cycle control algorithm controls the power of the air exchange cycle fan between the edible fungi planting warehouse and the hydroponic vegetables planting warehouse through the PID algorithm, that is, the implementation of the PWM duty cycle. According to the carbon cycle balance relationship of edible fungi and hydroponic vegetables in time series, the change factor of the time series Kp proportionality coefficient is calculated by integrating the segmented time series respiration rate model of edible fungi and the time series carbon assimilation rate model of hydroponic vegetables, and then an adaptive PID control algorithm is established. Its formula is as follows:

[0241] ,

[0242] where is the target PWM duty cycle of the air exchange cycle fan of the planting warehouse is the current time of the system is the daytime of the light cycle is the common planting days of edible fungi and hydroponic vegetables in the planting warehouse is the carbon assimilation rate of hydroponic vegetables on the th day and the respiration of edible fungi release rate of is the differential coefficient of the PID control algorithm, Real-time monitoring of edible fungi and hydroponic vegetable planting warehouses Concentration difference;

[0243] The constraints of the adaptive loop control algorithm are: ,

[0244] , They are real-time data of edible fungi and hydroponic vegetable planting warehouses. concentration, , They are suitable for edible fungi and hydroponic vegetable planting warehouses. Upper concentration limit.

[0245] (4) Concentration gain Through the planting warehouse Storage and release dynamic control mechanism get, The control logic is as follows:

[0246]

[0247] in, For planting warehouse Storage dynamic control mechanism, For planting warehouse Release dynamic control mechanisms;

[0248] Planting Barn The control logic of releasing the dynamic control mechanism is as follows:

[0249]

[0250] in, This is the lower limit of the suitable concentration in the hydroponic vegetable cultivation room. The length of the night;

[0251] Planting Barn The control logic constraints for releasing the dynamic control mechanism are:

[0252] .

[0253] The sixth step is to dynamically control the carbon cycle timing of the planting warehouse: Integrate the dynamic control model of the carbon cycle timing of the edible fungus and hydroponic vegetable planting warehouse into the planting warehouse system to achieve dynamic control of the carbon cycle timing.

[0254] (1) Substitute the parameter values ​​into In the adaptive loop control algorithm, the parameter values ​​include the current system time , daytime hours of the photoperiod , the carbon assimilation rate of hydroponic vegetables in Tian shui and the respiration rate of edible fungi ratio of and the real-time concentration difference inside the planting chamber of edible fungi and hydroponic vegetables, and combined with the constraint conditions of the adaptive cycle control algorithm, the target PWM duty cycle of the air exchange circulation fan is calculated in real time, and the power of the air exchange circulation fan is controlled to achieve adaptive regulation in time sequence, so that the concentrations inside the planting chambers of edible fungi and hydroponic vegetables are all within their suitable concentration ranges for growth.

[0255] (2) Application of the storage and release dynamic control mechanism in the planting chamber:

[0256] B1) Storage dynamic control mechanism in the planting chamber: The concentration inside the edible fungi planting chamber is monitored in real time. When > , the storage dynamic control mechanism of the planting chamber is triggered . The system starts the compressor to perform air compression operation; when < , the compressor stops working; where is the upper limit of the suitable concentration of the edible fungi planting chamber, is the lower limit of the suitable concentration of the edible fungi planting chamber;

[0257] B2) Release dynamic control mechanism in the planting chamber: The concentration inside the hydroponic vegetable planting chamber, the current system time and and value are monitored;

[0258] When and the constraint conditions are met, the release dynamic control mechanism of the planting chamber is triggered . The system starts the compressor to perform compressed air release operation; when is monitored, the compressor stops working; where is the upper limit of the suitable concentration of the hydroponic vegetable planting chamber, the lower limit of the suitable concentration inside the hydroponic vegetable planting chamber, The number of days that edible fungi and hydroponic vegetables are grown together in the growing chamber. For the Carbon assimilation rate of hydroponic vegetables per day Respiration of edible fungi Release rate The ratio of .

[0259] The following is further described by taking Oyster mushroom and hydroponic lettuce as examples.

[0260] Step 1: Determine the regulation cycle of Oyster mushroom and hydroponic lettuce. The growth cycle of hydroponic lettuce is Days; Single-crop growth cycle of Pleurotus ostreatus Days, long under suitable conditions The growth cycle of Pleurotus ostreatus days; therefore, the complete regulation cycle of the oyster mushroom and hydroponic lettuce cultivation warehouse is obtained by coupling calculation Days, number of mushroom stick batches planted during the entire growth cycle Batch, number of batches of hydroponic lettuce planted Batch, so after planting a batch of oyster mushrooms and hydroponic lettuce separately, they can be harvested together.

[0261] Step 2: Collecting Oyster Mushrooms and Hydroponic Lettuce Groups Time series data is preprocessed.

[0262] (1) The average size of the selected Pleurotus ostreatus sticks was 10*10*16 cm. When the fruiting body period came, the single Pleurotus ostreatus sticks were placed in the edible mushroom colony carbon flux monitoring device for cultivation. All parameters were set within the range suitable for growth, that is, the ambient temperature was controlled at 22°C, the humidity was maintained at 85%, and the light was normal scattered light. During the entire growth stage, five groups of samples were collected in the morning, afternoon, and night time periods respectively. Concentration data, select the time series within the ventilation interval of two devices The concentration data is valid data. The outliers in the data set are removed by Z-score method, and then the time series Calculate the change rate and obtain the respiration of Pleurotus ostreatus Release rate time series dataset.

[0263] (2) The selected hydroponic lettuce variety was glass lettuce, which was cultured using Japanese garden-style universal nutrient solution. The hydroponic lettuce seedlings were placed in a hydroponic tank with a size of 29.6*19.7 cm. Eight lettuces were planted in each hydroponic tank and divided into five groups. Each group of lettuces was cultured under different light intensities, with the light intensities set to 50, 100, 150, 200, and 300, respectively. During the planting period, the growth environment temperature and relative humidity are maintained at 22 / 18 ± 1 °C and 65 / 60%, respectively, during the day and night, and the photoperiod is set to 14 / 10 hours, that is hours, hours, and the nutrient solution concentration is configured to be approximately 2.80 dS·m -1 . Every five days, the hydroponic lettuce is placed in a plant population carbon flux measurement device to measure the time-series carbon assimilation data for 5 minutes. The outliers in the dataset are removed by the Z-score method, and then through the time-series change rate calculation, a time-series dataset of the carbon assimilation rate of hydroponic lettuce is obtained.

[0264] Step 3: Based on the time-series dataset of the carbon assimilation rate of hydroponic lettuce, with the planting days of hydroponic lettuce , light intensity as the two-dimensional input, and the carbon assimilation rate as the output, construct a time-series carbon assimilation rate model for hydroponic lettuce: Select the kernel function of the relevant vector machine model as the Gaussian kernel function, and use the improved raccoon optimization algorithm to optimize the kernel width parameter γ and the noise precision parameter α of the relevant vector machine model; set the range of the kernel width parameter γ of the relevant vector machine model to γ ∈ [10-3, 1], and the range of the noise precision parameter α to α ∈ [10-3, 1]. Set the population size N of the improved raccoon optimization algorithm to 20, and the maximum number of iterations is set to 50 times; the optimal parameter combination and obtained by using the improved raccoon optimization algorithm are set as the parameters of the relevant vector machine RVM model. Input the standardized training data, and the relevant vector machine model fits the nonlinear relationship between the input features and the carbon assimilation rate, and outputs the time-series carbon assimilation rate model. Train the time-series carbon assimilation rate model multiple times, and use 5-fold cross-validation to select the model with the smallest average root mean square error and the largest determination coefficient as the final time-series carbon assimilation rate model for hydroponic lettuce , and the time-series carbon assimilation rate model diagram of a single hydroponic lettuce during a growth cycle is as Figure 2 shown.

[0265] Step 4: Based on the time-series dataset of the respiration rate of Pleurotus ostreatus, according to the time ranges corresponding to the fruiting body differentiation stage and the development stage of Pleurotus ostreatus, extract the data corresponding to the stages within the time range. Using the planting days of Pleurotus ostreatus as the input, and the respiration rate of Pleurotus ostreatus during the fruiting body differentiation stage and the development stage of Pleurotus ostreatus as the output, use the polynomial regression algorithm to build a segmented model for these two stages; it is known that the cycle of the fruiting body differentiation stage of Pleurotus ostreatus is days, and the growth cycle of Pleurotus ostreatus is days, and the growth cycle of Pleurotus ostreatus is days, and the growth cycle of Pleurotus ostreatus is On the day, by comparing the root mean square error (RMSE) of each order polynomial, it was found that the second-order model could balance accuracy and complexity in both stages. The RMSE in the fruiting body differentiation stage and the development stage was better than that of other orders. Therefore, the model order for both stages was determined to be the second order. Using the least squares method to fit the coefficients, the formula for the piecewise time-series respiration rate model of Pleurotus ostreatus was determined as follows:

[0266] ,

[0267] The piecewise time-series respiration rate model diagram of a single Pleurotus ostreatus mushroom stick during the fruiting body period is as Figure 3 shown.

[0268] Step 5: Construct a time-series dynamic regulation model for the carbon cycle in the cultivation chambers of Pleurotus ostreatus and hydroponic lettuce.

[0269] (1) Obtaining the optimal planting ratio of Pleurotus ostreatus and hydroponic lettuce: Discretize the complete regulation cycle of the cultivation chambers of Pleurotus ostreatus and hydroponic lettuce days into 720 time nodes to form a time series , , , is the time interval between adjacent time nodes, represents the (c + 1)-th time node in the time series, represents the c-th time node in the time series,

[0270] Take the light intensity , the number of hydroponic lettuce plants , the piecewise time-series respiration rate model of Pleurotus ostreatus and the time-series carbon assimilation rate model of hydroponic lettuce as the inputs of the time-series dynamic regulation model for the carbon cycle in the cultivation chambers of Pleurotus ostreatus and hydroponic lettuce. Obtain the optimal number of Pleurotus ostreatus plants by optimizing the objective function, and obtain the optimal planting ratio of Pleurotus ostreatus and hydroponic lettuce ;

[0271] (2) Under the planting mode with the optimal planting ratio of Pleurotus ostreatus and hydroponic lettuce , construct a time-series dynamic regulation model for the carbon cycle in the cultivation chambers of Pleurotus ostreatus and hydroponic lettuce. Step 6: Incorporate the constructed time-series dynamic regulation model for the carbon cycle in the cultivation chambers of Pleurotus ostreatus and hydroponic lettuce into the cultivation chamber system to achieve dynamic regulation of the carbon cycle in time series.

[0272] After calculation, under the condition of single cultivation, the total release amount of a single Pleurotus ostreatus mushroom stick within 30 days is 44.1234×10- 3 m 3 , and when hydroponic lettuce is planted under the light intensity at The total absorption is 2.8001×10- 3 m 3 , and it can be obtained that the net emission is 41.3224×10 -3 m 3 . When Pleurotus ostreatus and hydroponic lettuce are co-cultivated at a planting ratio of 1:15, the net emission is 2.1084×10 -3 m 3 . Therefore, compared with single cultivation, the co-cultivation method can reduce the emission into the atmosphere by 94.90%, achieving nearly zero carbon emissions. The total planting cycle of the planting warehouse system is set within the light intensity for hydroponic lettuce planting of 150, and the coupling model curve graph of sequential absorption and release is as shown. The area of the orange region is when the respiration rate of Pleurotus ostreatus in the early stage is much greater than the carbon assimilation rate of hydroponic lettuce, and the storage dynamic control mechanism of the planting warehouse is executed, and the total Figure 4 amount stored by the compressor compression; the area of the purple region is when the respiration rate of Pleurotus ostreatus in the later stage is much less than the carbon assimilation rate of hydroponic lettuce, and the release dynamic control mechanism of the planting warehouse is executed, and the total amount stored by the compressor release; the blue region is when the adaptive cycle control algorithm is adopted, and the total amount exchanged between the Pleurotus ostreatus planting warehouse and the hydroponic lettuce planting warehouse; in addition, the phenomenon that the carbon assimilation rate decreases with the increase of planting days in the early stage of hydroponic lettuce growth is, on the one hand, due to certain mechanical errors of the measuring equipment, and on the other hand, due to the relatively small seedlings of hydroponic lettuce in the early stage of growth, the carbon assimilation rate is relatively weak. The fluctuations in the early stage of growth are normal physiological phenomena, which can be ignored compared to the improvement of photosynthetic capacity brought by the increase of leaf area index in the middle and later stages.

[0273] When the cultivation cycle ends, the air with a high concentration of stored in the compressor is completely released. During the entire cultivation cycle, when the compressor pumps air from the Pleurotus ostreatus planting area and compresses and stores it in the early stage, a certain volume of air will be inhaled from the outside, and when the compressed gas is supplemented to the hydroponic lettuce planting area in the later stage, an equal volume of gas will be discharged to the outside. However, when discharging, due to the relatively low concentration in the hydroponic lettuce planting area, which is close to the average atmospheric concentration, the additional emission caused by ventilation is negligible and almost will not affect the in the atmosphere.​​​​​The concentration has an impact, thus ensuring that nearly zero carbon emissions can be achieved throughout the cultivation cycle, meeting the development requirements of green agriculture.

[0274] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for dynamically controlling the carbon cycle timing of edible fungi and hydroponic vegetable planting warehouses, characterized in that: The following steps are involved: 11) Determination of the regulation cycle of edible fungi and hydroponic vegetables: Coupling and calibrating the time axis of the edible fungi fruiting body period and the hydroponic vegetable growth sequence to determine the complete regulation cycle of the edible fungi and hydroponic vegetable planting warehouse ; 12) Collection and preprocessing of carbon dioxide time series data of edible fungi and hydroponic vegetable populations; 13) Construct a temporal carbon assimilation rate model for hydroponic vegetables based on the improved raccoon optimization algorithm; 14) Construct a segmented time series respiration rate model for edible fungi; 15) Construct a dynamic regulation model of carbon cycle in edible fungi and hydroponic vegetable cultivation warehouses; 16) Dynamic regulation of carbon cycle timing in planting warehouses: Integrate the dynamic regulation model of carbon cycle timing in edible fungi and hydroponic vegetable planting warehouses into the planting warehouse system to achieve dynamic regulation of carbon cycle timing.

2. The method for dynamically controlling the carbon cycle timing of an edible fungus and hydroponic vegetable planting warehouse according to claim 1 is characterized in that: The determination of the regulation cycle of the edible fungi and the hydroponic vegetables comprises the following steps: 21) Determine the growth cycle of edible fungi and hydroponic vegetables respectively. Assume that the growth cycle of a single crop of edible fungi is Days, long under suitable conditions The growth cycle of edible fungi The growth cycle of hydroponic vegetables is sky; 22) Coupling calculation of the complete control cycle of edible fungi and hydroponic vegetable cultivation warehouse , The total cycle of co-cultivation of edible fungi and hydroponic vegetables. To calculate the least common multiple, The number of mushroom stick batches planted during the entire growth cycle is , number of batches of hydroponic vegetables planted , make sure to harvest together.

3. The method for dynamically controlling the carbon cycle timing of an edible fungus and hydroponic vegetable planting warehouse according to claim 1 is characterized in that: The carbon dioxide time series data collection and preprocessing of edible fungi and hydroponic vegetable groups are as follows: Time-series release dataset and carbon assimilation of hydroponic vegetables under different light intensities throughout the growth period Time series demand data set, for abnormal data Method for denoising; It includes the following steps: 31) Using the edible fungus colony carbon flux monitoring device, under the condition of suitable and stable microenvironment, high-precision The sensor monitors the respiration of a single mushroom stick in the fruiting body growth stage in real time, and obtains Time series release data set ; 32) Use the plant group carbon assimilation rate measurement device to measure the group carbon assimilation of the whole hydroponic vegetable under different light intensities during the entire growth cycle Timing Requirements Dataset ; 33) Adoption Law Time series release data set and group carbon assimilation Timing Requirements Dataset To handle the outliers in set up Time series release data set Include value, denoted as , , …, ,in Indicates the data set individual data; Assume that the group carbon assimilation Timing Requirements Dataset Include value, denoted as , , …, ,in Indicates the data set data, respectively, calculated Time series release data set and group carbon assimilation Timing Requirements Dataset The mean and standard deviation , ， ， in, , Respectively represent Time series release data set The mean and standard deviation of , Represents group carbon assimilation Timing Requirements Dataset The mean and standard deviation of , To sum the index, express Time series release data set The ath data in; Group carbon assimilation Timing Requirements Dataset The bth data in; Then calculate the Z value of each data point: ,in, represent Time series release data set The Z value of the a-th data point in Representative group carbon assimilation Timing Requirements Dataset The Z value of the b-th data point in Setting Thresholds >3 and >3, outliers exceeding the threshold are eliminated; 34) Timing Rate of change calculation: 341) Time series release data set , Group Carbon Assimilation Timing Requirements Dataset Press = Calculate the concentration change rate at 60s intervals: ， ， in: and The unit is ppm / min. is the time in seconds, represent The carbon flux monitoring device of edible fungi was used to measure the concentration, represent The carbon flux monitoring device of edible fungi was used to measure the concentration, represent The carbon assimilation rate of plant populations was measured by the device concentration, represent The carbon assimilation rate of plant populations was measured by the device concentration, Edible mushrooms per minute Release rate, Hydroponic vegetables every minute Absorption rate; By calculating the concentration change rate of edible fungi and hydroponic vegetables in multiple time periods every day during the planting cycle, the concentration of edible fungi per minute per day during the planting cycle is calculated. Average release rate and hydroponic vegetables under different light intensities Average absorption rate , respectively expressed as: ， ， Where N is the number of measurements in multiple valid time periods in a day, i is the summation index, The number of days for growing edible fungi, Number of days for hydroponic vegetable planting, For the Edible fungi every minute Release rate, For different light intensities Hydroponic vegetables per minute Absorption rate, The growth cycle of edible fungi. For the growth cycle of hydroponic vegetables; 342) Combined with the effective volume of the edible fungus group carbon flux monitoring device Calculate the number of days Release rate, generating edible fungus respiration Release rate time series dataset , the specific calculation method is: ， in, Representative Respiration of edible mushrooms Release rate, in m 3 / sky, The unit is m 3 ; 343) Suppose the daytime duration of the photoperiod for hydroponic vegetable cultivation is hours, night duration is hours, combined with the effective volume of the plant community carbon assimilation rate measurement equipment Convert to days Absorption rate, generating a time series dataset of carbon assimilation rate of hydroponic vegetables , The calculation method is: ， in, Representative Carbon assimilation rate of hydroponic vegetables per day, unit: m 3 / sky, The unit is m 3 .

4. The method for dynamically controlling the carbon cycle timing of an edible fungus and hydroponic vegetable planting warehouse according to claim 1 is characterized in that: The construction of a temporal carbon assimilation rate model for hydroponic vegetables based on the improved raccoon optimization algorithm comprises the following steps: 41) Hydroponic vegetable carbon assimilation rate time series dataset Divide into training set With the test set , where the number of days of hydroponic vegetable cultivation and light intensity are the features, and the carbon assimilation rate is the label; 42) Adoption Method to standardize the data: ， ， ， in, , , They are the number of days for hydroponic vegetable cultivation, light intensity, and carbon assimilation rate. , , They are the standardized hydroponic vegetable planting days, the standardized light intensity, and the standardized carbon assimilation rate. , They are The mean and standard deviation of , They are The mean and standard deviation of , for The mean and standard deviation of The unit is ; 43) Set the relevance vector machine model, The kernel function of the RVM model is selected as the Gaussian kernel function , in, , The standardized feature vector is used to optimize the kernel width parameter γ and noise precision parameter α of the correlation vector machine model using the improved raccoon optimization algorithm. The range of the kernel width parameter γ of the RVM model is set to γ∈[10 -3 ,1], the range of noise accuracy parameter α is α∈[10 -3 ,1], set the population size N of the improved raccoon optimization algorithm to 20, and the maximum number of iterations The number of times is set to 50, and the fitness function F is defined as the mean square error of the RVM model test set, and the calculation formula is: ， in, is the number of samples in the test set, m represents the mth sample in the test set, is the actual carbon assimilation rate value of the mth sample in the test set, is the model's predicted value for the mth sample; 44) Initialize the population and encode the position of each individual in the following way: ， in, , Representing the population The encoding values ​​of the kernel width parameter and noise precision parameter of the correlation vector machine model corresponding to each individual, , are the lower and upper bounds of the kernel width parameter γ, respectively. , are the lower and upper bounds of the noise accuracy parameter α, respectively. is a random number uniformly distributed in the interval [0,1]. is the index of the individual in the population, and N is the population size; The position vectors of N=20 raccoons are generated randomly. , , and use the fitness function F to calculate the fitness value of each individual as the initial value; 45) Design a dynamic population size adjustment module: calculate the average Euclidean distance of each individual in the solution space, and use the standardized distance variance as the population diversity evaluation indicator; when the population is initialized or updated each time, evaluate the diversity of the current population by calculating the average Euclidean distance between the individual position in the population and the average of all individual positions. When the population diversity is lower than the threshold of 0.3 and the population size is lower than the maximum population size of 100, update the population: randomly generate new sample individuals and expand the population size; when the population diversity is higher than the threshold of 0.3 and the population size is higher than the minimum population size of 30, update the population: use the fitness function F to calculate the fitness value of each individual, eliminate the worst individuals, and reduce the population size; 46) In the hydroponic vegetable carbon assimilation rate time series data set, the encoded γ and α were used in the training set Train the relevant vector machine model and use the trained relevant vector machine model in the test set The prediction is made on the , and the fitness value is calculated by the difference between the predicted result and the true value. The fitness values ​​of all individuals are traversed, and the individual with the smallest fitness is selected as the current optimal solution. The γ and α parameter combinations corresponding to the optimal solution are recorded as and , whose position vector is recorded as ; 47) In the search space, select Each individual updates its position, and the update formula is as follows: ， Among them, N is the population size, p is the current number of iterations, represents the component of the i-th individual in the j-th dimension at the p-th iteration, represents the component of the i-th individual in the j-th dimension at the p+1-th iteration, represents the component of the optimal individual in the jth dimension at the pth iteration, is a random number uniformly distributed in the interval [0,1], I represents a random integer from the integer set {1, 2}, and i represents the number of the individual selected for location update; 48) The fitness function F is used to calculate the fitness value of each individual, and a greedy selection is performed. If the fitness of the individual after the update is better than the fitness value before the update, the updated position is retained, otherwise the position is not updated. The greedy selection strategy formula is as follows: ， in, is the position of the ith individual at the p+1th iteration, is the position of the ith individual at the pth iteration, and are the fitness values ​​corresponding to the positions of the i-th individual at the p+1-th iteration and the p-th iteration respectively; 49) Design a learning rate adjustment strategy for position adaptive updates: Construct a composite function based on the iteration attenuation factor and the diversity feedback coefficient. The learning rate update formula is: ， in, is the current iteration number, is the attenuation factor, is the maximum number of iterations, is the number of diverse solution sets, i is the summation variable, is the diversity feedback coefficient, For population diversity, represents the learning rate at the pth iteration, represents the learning rate at the p+1th iteration, is the exponential operation of natural constants; A dual control mechanism is used to achieve the coordinated optimization of global exploration and local development: the global search mode is used in the early stage of iteration, and the local development mode is started as the number of iterations increases to speed up the positioning of the global optimal neighborhood; 410) Select another individuals, update their positions in the search space, First, randomly generate a temporary position vector in the search space , and its calculation formula is: ， The location is updated as follows: ， ； Among them, i represents the number of the individual selected for location update, N is the population size, and Represent the lower and upper bounds of the j-th dimension in the search space, respectively. is a random number uniformly distributed in the interval [0,1], I represents a random integer from the integer set {1, 2}, represents the component of the i-th individual in the j-th dimension at the p-th iteration, represents the component of the i-th individual in the j-th dimension at the p+1-th iteration, Represents the temporary position vector The corresponding solution fitness value, is the fitness value corresponding to the position of the i-th individual at the p-th iteration; 411) In the search space, simulate the random perturbation mechanism of individuals when facing dynamic perturbations, When the random perturbation mechanism is triggered, a random position is generated near the current position of each individual, with the following formula: ， ， ， Among them, i represents the number of the individual selected for location update, represents the number of iterations, and are the upper and lower bounds of the j-th dimension variable updated with the number of iterations, is the maximum number of iterations; 412) Calculate the fitness value of each individual and execute the greedy selection strategy again; If the fitness of the individual after the update is better than the fitness value before the update, the updated position is accepted, otherwise the position is not updated; 413) Continue iterating until the preset maximum number of iterations is reached =50, after the iteration, the optimal solution is output And the corresponding optimal value The optimal solution is the optimal combination of the kernel width parameter γ and the noise precision parameter α of the correlation vector machine model. and Substitute the relevant vector machine model and input the training set , the RVM model fits the nonlinear relationship between input features and carbon assimilation rate, and outputs a temporal carbon assimilation rate model ; 414) Train the time series carbon assimilation rate model multiple times, and use 5-fold cross validation to select the model with the smallest mean root mean square error and the largest coefficient of determination as the final time series carbon assimilation rate model .

5. The method for dynamically controlling the carbon cycle timing of an edible fungus and hydroponic vegetable planting warehouse according to claim 1 is characterized in that: The construction of the edible fungus segmented time series respiration rate model comprises the following steps: 51) Based on the respiration of edible fungi Release rate time series dataset According to the time range corresponding to the differentiation stage and development stage of edible fungi fruiting bodies, the data of the corresponding stage of the time range are extracted, and the number of days of edible fungi cultivation is calculated. As input, the respiration of edible fungi during the fruiting body differentiation and development stages Release rate For the output, the polynomial regression algorithm is used for segmented modeling of these two stages: ， ， in, It is a temporal respiration rate model for the fruiting body differentiation stage of edible fungi. It is a temporal respiration rate model for the fruiting body development stages of edible fungi. It is the period of differentiation of fruiting bodies of edible fungi. It is the growth cycle of edible fungi. is the number of days for growing edible fungi, n and m are the orders of the polynomial, , , …, , , , …, is the coefficient to be determined; 52) The coefficients to be determined are fitted by the least squares method to form a segmented time series respiration rate model of edible fungi , the formula is as follows: 。 6. The method for dynamically controlling the carbon cycle timing of an edible fungus and hydroponic vegetable planting warehouse according to claim 1 is characterized in that: The construction of the carbon cycle time series dynamic regulation model of the edible fungus and hydroponic vegetable planting warehouse includes the following steps: 61) Obtaining the optimal planting ratio of edible fungi and hydroponic vegetables: Discretize the complete control cycle T of the edible fungi and hydroponic vegetable planting warehouse into k time nodes to form a time series , where k is the number of time nodes, represents the kth time node in the time series, , , is the time interval between adjacent time nodes, represents the c+1th time node in the time series, represents the cth time node in the time series, Light intensity , Number of hydroponic vegetables planted , Edible Fungi Segmented Time Series Respiration Rate Model A temporal carbon assimilation rate model for hydroponic vegetables As the input of the dynamic control model of carbon cycle timing of edible fungi and hydroponic vegetable planting warehouse, on the basis of satisfying carbon metabolism balance, an optimization objective function is constructed to determine the optimal number of edible fungi to be planted. , the optimization objective function is: ， in, and are the planting quantities of edible fungi and hydroponic vegetables, is the number of edible mushroom sticks planted, The number of batches of hydroponic vegetables planted, is the time node, is the respiration of edible fungi at the cth time point Release rate, is the carbon assimilation rate of hydroponic vegetables at the cth time node, Expressed as the number of edible fungi planted Perform rounding operation upwards. is the regularization coefficient, About The penalty function is Control the intensity of punishment. By optimizing the objective function, the optimal After that, the optimal planting ratio of edible fungi and hydroponic vegetables is obtained. ; 62) Optimal planting ratio of edible fungi and hydroponic vegetables in the planting warehouse In the planting mode, the carbon cycle timing dynamic control model of edible fungi and hydroponic vegetable planting warehouse is used to dynamically control the environment. The formula of the carbon cycle timing dynamic control model of edible fungi and hydroponic vegetable planting warehouse is as follows: ， in, Dynamically adjust the target value of the carbon cycle timing in the planting warehouse. For planting barns The adaptive loop control algorithm controls the concentration, Adopting a grow barn for the grow barn Storage and release dynamic control mechanism Concentration gain, is the real time; 63) The adaptive cycle control algorithm controls the air exchange circulation fan power between the edible fungus cultivation warehouse and the hydroponic vegetable cultivation warehouse through the PID algorithm, that is, the realization of the PWM duty cycle. According to the time-series carbon cycle balance relationship between edible fungi and hydroponic vegetables, the edible fungus segmented time-series respiration rate model and the hydroponic vegetable time-series carbon assimilation rate model are integrated to calculate the change factor of the time-series Kp proportional coefficient, and then establish an adaptive PID control algorithm, the formula of which is as follows: ， in, The target PWM duty cycle of the air exchange circulation fan in the growing room, is the current system time, is the daytime time of the photoperiod, The number of days that edible fungi and hydroponic vegetables are grown together in the growing chamber. For the Carbon assimilation rate of hydroponic vegetables per day Respiration of edible fungi Release rate The ratio of , is the differential coefficient of the PID control algorithm, Real-time monitoring of edible fungi and hydroponic vegetable planting warehouses Concentration difference; The constraints of the adaptive loop control algorithm are: , , They are real-time data of edible fungi and hydroponic vegetable planting warehouses. concentration, , They are suitable for edible fungi and hydroponic vegetable planting warehouses. Upper concentration limit; 64) Concentration gain Through the planting warehouse Storage and release dynamic control mechanism get, The control logic is as follows: ， in, For planting warehouse Storage dynamic control mechanism, For planting warehouse Release dynamic control mechanisms; Planting Barn The control logic of releasing the dynamic control mechanism is as follows: ， in, This is the lower limit of the suitable concentration in the hydroponic vegetable cultivation room. The length of the night; Planting Barn The control logic constraints for releasing the dynamic control mechanism are: 。 7. The method for dynamically controlling the carbon cycle timing of an edible fungus and hydroponic vegetable planting warehouse according to claim 1 is characterized in that: The dynamic regulation of the carbon cycle timing of the planting bin includes the following steps: 71) Substitute parameter values ​​into In the adaptive loop control algorithm, the parameter values ​​include the current system time , daytime hours of the photoperiod , No. Carbon assimilation rate of hydroponic vegetables Respiration of edible fungi Release rate Ratio And real-time monitoring of edible fungi and hydroponic vegetable cultivation warehouses Concentration difference , and combined with The constraints of the adaptive cycle control algorithm calculate the target PWM duty cycle of the air exchange circulation fan in real time, control the power of the air exchange circulation fan, and achieve timing Adaptive regulation makes the edible fungi and hydroponic vegetables grow in the warehouse The concentrations are all within the suitable concentration range for their growth; 72) Planting Barn Application of dynamic control mechanism of storage and release: 721) Planting Barn Storage dynamic control mechanism: real-time monitoring of the edible fungus cultivation warehouse concentration ,when > When the planting chamber is triggered Storage Dynamic Control Mechanism , the system starts the compressor to perform air compression operation; when < , the compressor stops working; among them, Suitable for edible fungus cultivation warehouse The upper limit of concentration, Suitable for edible fungus cultivation warehouse Lower concentration limit; 722) Planting Barn Unleashing dynamic control mechanisms: Monitoring the hydroponic vegetable growing chamber concentration , Current system time as well as The value of when When the constraints are met, the planting warehouse is triggered Unleashing dynamic control mechanisms , the system starts the compressor to release compressed air; when it detects , the compressor stops working; among them, Suitable for hydroponic vegetable planting warehouse The upper limit of concentration, It is the lower limit of suitable concentration in hydroponic vegetable cultivation room. The number of days that edible fungi and hydroponic vegetables are grown together in the growing chamber. For the Carbon assimilation rate of hydroponic vegetables per day Respiration of edible fungi Release rate The ratio of .

8. The method for dynamically controlling the carbon cycle timing of an edible fungus and hydroponic vegetable planting warehouse according to claim 7 is characterized in that: The grow barn includes the following: Data acquisition subsystem: Set up sensors to monitor the growth environment parameters of edible fungi and hydroponic vegetables in real time. The sensors include temperature and humidity sensors, sensor, Sensors, light intensity sensors, solutions Value and Value measurement sensor; Data fusion subsystem: by coupling and calibrating the time axis of the edible fungus fruiting body period and the hydroponic vegetable growth sequence, the total control cycle of the planting warehouse is obtained And the number of edible mushroom sticks planted during the entire growth cycle and number of hydroponic vegetable planting batches ; The optimal planting ratio of edible fungi and hydroponic vegetables is automatically generated by matching the embedded edible fungi segmented time series respiration rate model and the hydroponic vegetable time series carbon assimilation rate model. and Carbon assimilation rate of hydroponic vegetables Respiration of edible fungi Release rate Ratio , thereby obtaining a dynamic control method for the carbon cycle timing of the planting warehouse for the current planting category; The data fusion subsystem combines the real-time environmental parameter values ​​transmitted by the data acquisition subsystem and sends control instructions to the dynamic feedback control subsystem; Dynamic feedback control subsystem: receives control instructions transmitted by the data fusion subsystem, drives the actuators to cooperate and control the environmental parameters in the planting chamber; A cultivation box, an air compression storage device and a control device, wherein an airtight partition is slidably arranged in the cultivation box, the airtight partition divides the cultivation box into an edible fungus cultivation bin and a hydroponic vegetable cultivation bin, the cultivation box is provided with ventilation holes in the edible fungus cultivation bin and the hydroponic vegetable cultivation bin, respectively, an air exchange circulation fan is arranged on the upper part of the airtight partition, the air compression storage device comprises a compressor and an air inlet pipe and an air outlet pipe arranged on both sides of the compressor, the air inlet pipe is connected to the edible fungus cultivation bin, the air outlet pipe is connected to the hydroponic vegetable cultivation bin, the ventilation hole, the air inlet pipe and the air outlet pipe are all controlled to open and close by an electromagnetic valve, the air exchange circulation fan is controlled to open and close and adjust the wind speed by a motor, and the control device is used to receive The sensor signal controls the operation of the compressor, solenoid valve and air exchange circulation fan according to the preset dynamic regulation method of the carbon cycle timing of the planting bin.

9. A computer-readable storage medium, characterized in that: A computer program is stored on the storage medium. When the computer program is executed by the processor, the method for dynamically controlling the timing of carbon cycle in an edible fungus and hydroponic vegetable planting bin as described in any one of claims 1 to 7 is implemented.

10. A computer device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and running on the processor. When the processor executes, the method for dynamically controlling the timing of carbon cycle in an edible fungus and hydroponic vegetable planting bin as described in any one of claims 1 to 7 is implemented.

Images