A henhouse environment control method, system, device and medium

Abstract

The application discloses a chicken house environment control method, system, device and medium, and relates to the field of chicken house environment control.The method comprises the following steps: determining a target average thermal sensation temperature required by broilers in a chicken house; determining a ventilation control strategy when the broilers in the chicken house maintain the target average thermal sensation temperature by using a thermal sensation temperature prediction model and a genetic algorithm; the ventilation control strategy comprises the number of fans to be turned on in a chicken house ventilation system and the angle of ventilation windows to be opened; the thermal sensation temperature prediction model is determined by using training data and a machine learning algorithm based on a proxy model; the training data comprises different chicken house parameters and corresponding actual average thermal sensation temperatures of broilers; the chicken house parameters comprise an air temperature outside the chicken house, the number of fans to be turned on in the chicken house ventilation system and the angle of ventilation windows to be opened in the chicken house ventilation system.The application can improve the accuracy of chicken house environment control.

Description

A method, system, equipment and medium for controlling the environment of a chicken house. Technical Field

[0001] This invention relates to the field of chicken house environmental control, and in particular to a method, system, equipment and medium for chicken house environmental control. Background Technology

[0002] In chicken house environmental control systems, temperature and airflow velocity are typically controlled through ventilation. Most existing large, enclosed chicken houses employ mechanical negative pressure ventilation based on longitudinal ventilation. This is achieved by adjusting negative pressure fans, ventilation windows, and evaporative cooling systems to control target temperature and airflow velocity within the house. Broiler chickens have varying temperature and airflow requirements at different ages. A well-controlled chicken house should be able to meet the environmental control needs of chickens of different ages and under different environmental conditions, adjusting the ventilation strategy according to the actual needs of the broilers and real-time environmental conditions.

[0003] Currently, most chicken houses in China rely on manual experience to determine the number of fans to operate, the size of small windows, and the size and angle of the wind deflectors at the evaporative cooling pads to control environmental parameters. This experience-based control strategy struggles to guarantee the accuracy and timeliness of environmental control, often resulting in delayed adjustments or excessively wide control ranges. Excessive control can cause significant temperature fluctuations in a short period, stressing the broilers. Furthermore, experience-based control measures require substantial manpower and are highly dependent on experienced poultry farmers, hindering the expansion of broiler farming and the promotion of advanced technologies. While modern poultry farming has introduced automated control equipment, current automated control systems largely rely on empirical data and focus on controlling single chicken house parameters, lacking effective algorithmic support for managing the multi-factor control of actual chicken houses.

[0004] Therefore, improving the accuracy of chicken house environmental control remains an urgent problem to be solved. Summary of the Invention

[0005] Based on this, embodiments of the present invention provide a method, system, equipment, and medium for controlling the environment of a chicken house, so as to improve the accuracy of controlling the environment of a chicken house.

[0006] To achieve the above objectives, embodiments of the present invention provide the following solutions:

[0007] A method for controlling the environment of a chicken coop includes:

[0008] Determine the target average perceived temperature that needs to be maintained for broilers in the chicken house;

[0009] A somatosensory temperature prediction model and a genetic algorithm are used to determine the ventilation control strategy for maintaining broilers in the chicken house at the target average somatosensory temperature. The ventilation control strategy includes: the number of fans to be turned on in the chicken house ventilation system and the angle at which the ventilation windows need to be opened.

[0010] The perceived temperature prediction model is determined using training data and a machine learning algorithm based on a surrogate model. The training data includes different chicken house parameters and the corresponding actual average perceived temperature of the broilers. The chicken house parameters include the outside air temperature of the chicken house, the number of fans in the chicken house ventilation system that are turned on, and the opening angle of the ventilation windows in the chicken house ventilation system.

[0011] Optionally, a flexibly controlled temperature prediction model and a genetic algorithm are used to determine the ventilation control strategy for maintaining broilers in the chicken house at the target average flexibly controlled temperature, specifically including:

[0012] Using the perceived temperature prediction model as the solution function of the genetic algorithm, under the constraints of the constraints, the algorithm iterates multiple times with the goal of minimizing the difference between the average perceived temperature output by the perceived temperature prediction model and the target average perceived temperature, and obtains the ventilation control strategy for maintaining the broilers in the chicken house at the target average perceived temperature.

[0013] The constraints are determined based on the outside air temperature, the range of the actual number of fans in the chicken house ventilation system, and the range of the actual opening angle of the ventilation windows in the chicken house ventilation system.

[0014] Optionally, the method for determining the perceived temperature prediction model specifically includes:

[0015] Obtain training data;

[0016] The training data is input into the artificial neural network. The goal is to minimize the difference between the average perceived temperature predicted by the artificial neural network and the actual average perceived temperature of the broilers. The backpropagation algorithm is used for training, and the trained artificial neural network is determined as the perceived temperature prediction model.

[0017] The present invention also provides a chicken coop environment control system, comprising:

[0018] The target temperature determination module is used to determine the target average perceived temperature that broilers need to maintain in the chicken house;

[0019] A ventilation control strategy determination module is used to determine the ventilation control strategy for maintaining broilers in the chicken house at a target average perceived temperature using a body temperature prediction model and a genetic algorithm. The ventilation control strategy includes: the number of fans that need to be turned on in the chicken house ventilation system and the angle at which the ventilation windows need to be opened.

[0020] The perceived temperature prediction model is determined using training data and a machine learning algorithm based on a surrogate model. The training data includes different chicken house parameters and the corresponding actual average perceived temperature of the broilers. The chicken house parameters include the outside air temperature of the chicken house, the number of fans in the chicken house ventilation system that are turned on, and the opening angle of the ventilation windows in the chicken house ventilation system.

[0021] Optionally, the ventilation control strategy determination module specifically includes:

[0022] The solution unit is used to use the perceived temperature prediction model as the solution function of the genetic algorithm. Under the constraints of the constraints, it performs multiple iterations to solve the problem with the objective of minimizing the difference between the average perceived temperature output by the perceived temperature prediction model and the target average perceived temperature, so as to obtain the ventilation control strategy for maintaining the broilers in the chicken house at the target average perceived temperature.

[0023] The constraints are determined based on the outside air temperature, the range of the actual number of fans in the chicken house ventilation system, and the range of the actual opening angle of the ventilation windows in the chicken house ventilation system.

[0024] Optionally, the chicken coop environment control system further includes: a model determination module, used to determine the perceived temperature prediction model;

[0025] The model determination module specifically includes:

[0026] The data acquisition unit is used to acquire training data.

[0027] The training unit is used to input the training data into the artificial neural network, with the goal of minimizing the difference between the average perceived temperature predicted by the artificial neural network and the actual average perceived temperature of the broilers. The backpropagation algorithm is used for training, and the trained artificial neural network is determined as the perceived temperature prediction model.

[0028] The present invention also provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the above-described chicken coop environment control method.

[0029] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described chicken coop environment control method.

[0030] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0031] This invention employs training data and a machine learning algorithm based on a surrogate model to determine a perceived temperature prediction model. This perceived temperature prediction model, combined with a genetic algorithm, determines the ventilation control strategy for maintaining broilers in the chicken house at a target average perceived temperature. Compared to existing control strategies based on human experience, this approach ensures the accuracy of environmental control in the chicken house. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 is a flowchart of the chicken coop environment control method provided in an embodiment of the present invention;

[0034] Figure 2 is a hierarchical structure diagram of the artificial neural network provided in an embodiment of the present invention;

[0035] Figure 3 is a neural network training regression diagram with R = 0.99906 provided in an embodiment of the present invention;

[0036] Figure 4 is a neural network training regression diagram with R = 0.99193 provided in the embodiment of the present invention;

[0037] Figure 5 is a neural network training regression diagram with R = 0.99375 provided in an embodiment of the present invention;

[0038] Figure 6 is a neural network training regression diagram when R = 0.99774 provided in an embodiment of the present invention;

[0039] Figure 7 is a comparison chart of the predicted values ​​and actual sample values ​​of the body temperature prediction model provided in the embodiment of the present invention;

[0040] Figure 8 is a schematic diagram of the iterative calculation process of the genetic algorithm provided in an embodiment of the present invention;

[0041] Figure 9 is a structural diagram of the chicken house environment control system provided in an embodiment of the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] The purpose of this invention is to provide a method, system, device and medium for controlling the environment of a chicken house. By using machine learning algorithms and genetic algorithms based on surrogate models, the ventilation control strategy for maintaining broilers in the chicken house at a target average perceived temperature is determined, thereby ensuring the accuracy of the control of the chicken house environment.

[0044] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0045] Example 1

[0046] Referring to Figure 1, the chicken coop environment control method of this embodiment includes:

[0047] Step 101: Determine the target average perceived temperature that the broilers need to maintain in the chicken house.

[0048] Step 102: Use a perceived temperature prediction model and a genetic algorithm to determine the ventilation control strategy for maintaining the target average perceived temperature of broilers in the chicken house; the ventilation control strategy includes: the number of fans that need to be turned on in the chicken house ventilation system and the angle at which the ventilation windows need to be opened.

[0049] The perceived temperature prediction model is determined using training data and a machine learning algorithm based on a surrogate model. The training data includes different chicken house parameters and the corresponding actual average perceived temperature of the broilers. The chicken house parameters include the outside air temperature of the chicken house, the number of fans in the chicken house ventilation system that are turned on, and the opening angle of the ventilation windows in the chicken house ventilation system.

[0050] In one example, step 102 specifically includes:

[0051] Using the perceived temperature prediction model as the solution function of the genetic algorithm, under the constraints, the algorithm iterates multiple times with the objective of minimizing the difference between the average perceived temperature output by the perceived temperature prediction model and the target average perceived temperature, thereby obtaining the ventilation control strategy for maintaining broilers in the chicken house at the target average perceived temperature.

[0052] The constraints are determined based on the outside air temperature, the range of the actual number of fans in the chicken house ventilation system, and the range of the actual opening angle of the ventilation windows in the chicken house ventilation system.

[0053] In one example, the method for determining the perceived temperature prediction model in step 102 specifically includes: acquiring training data; inputting the training data into an artificial neural network, aiming to minimize the difference between the average perceived temperature predicted by the artificial neural network output and the actual average perceived temperature of the broilers, training the artificial neural network using the backpropagation algorithm, and determining the trained artificial neural network as the perceived temperature prediction model.

[0054] The following section introduces the various components determined by the perceived temperature prediction model.

[0055] 1. Determine the input and output variables

[0056] Ambient air temperature, the number of fans in operation, and the opening angle of the small window were selected as input variables, and the average perceived temperature of the broilers was selected as the output variable. An algorithmic model relating these input and output variables was established using a surrogate model approach.

[0057] The formula for calculating the perceived temperature of broiler chickens is as follows:

[0058] t s =tv×k+(φ-φ tar )×λ;

[0059] In the formula, t s t is the perceived temperature, v is the dry-bulb temperature, k is the wind speed, φ is the wind cooling coefficient, and φ is the relative humidity. tar λ represents the target humidity, and λ is the humidity coefficient. When calculating the perceived temperature for broilers, the relative humidity, humidity coefficient, and air cooling coefficient are determined based on actual measurements in the chicken house.

[0060] 2. Experimental Design

[0061] To construct a surrogate model, it is necessary to select sample points within the design range of the input variables to determine the sample points used to build the surrogate model. A well-designed experimental plan can lead to a more scientific experimental arrangement, ensuring that the experimental points are evenly distributed across the experimental space. This can significantly reduce the number of experiments, shorten the overall research time, and guarantee the accuracy and representativeness of the surrogate model.

[0062] Uniform design, created by Chinese mathematicians Academician Wang Yuan and Professor Fang Kaitai, is an optimized experimental design method that can uniformly distribute experimental points within the experimental range. It can effectively select representative experimental points to reflect the uniform distribution of the concentration changes of the mixture components with as few experiments as possible, and has been widely used in many research and application fields.

[0063] 3. Sample Data Collection

[0064] This example uses a uniform design method to design the test points for constructing the surrogate model. Under the designed test conditions, CFD computer numerical simulation technology is used to simulate the airflow organization in the chicken house and calculate the average perceived temperature of the chicken house under each condition.

[0065] For chicken houses with available testing conditions, the sample data required to build the surrogate model can also be obtained through actual testing.

[0066] 4. Model selection and training

[0067] Surrogate models are a commonly used optimization method in engineering. In actual engineering optimization design and control strategy research, the relationship between influencing parameters and desired control objectives is often not directly determined. Surrogate models constructed using surrogate model methods can well represent the mapping relationship between the influencing parameters and the actual objective function.

[0068] This embodiment applies a surrogate model to ventilation control in intensive chicken farms. Using a machine learning algorithm based on the surrogate model, an airflow organization regression model is established to determine the relationship between environmental factors, control variables, and chicken farm control objectives—essentially a perceived temperature prediction model. Based on this model, a multi-objective genetic algorithm is used for optimization to determine the optimal ventilation control strategy under different environmental conditions and control objectives. A method for calculating the control strategy of the chicken farm ventilation system is designed, providing a basis for the design and regulation of the chicken farm environmental control system.

[0069] Artificial neural networks are the most widely used type of surrogate model. They mimic the structure and function of neurons in the human brain, possessing strong learning and adaptability, and are highly effective in handling complex nonlinear problems and noisy data. Artificial neural networks do not require the intrinsic relationships between data; they only need the corresponding inputs and outputs to perform internal training, gradually adjusting the connection weights and thresholds of neurons to achieve the final training effect. In neural network surrogate models, backpropagation (BP) neural networks train through error feedback, exhibiting strong generalization and fault tolerance, and have been widely applied in various fields. Therefore, this embodiment selects a BP neural network as the construction method for the chicken coop airflow organization control model.

[0070] A backpropagation (BP) neural network consists of three parts: an input layer, hidden layers, and an output layer. Each layer is composed of neurons. The dataset is first submitted from the input layer to the hidden layers, then passed and processed through the weights between neurons, and finally output to the output layer. The transmission principle is shown in Figure 2. In Figure 2, x i W represents the i-th input parameter. ij W represents the weight of each input parameter i in the j-th hidden layer. jk f1 represents the weights of the output parameters of the j-th hidden layer in the k-th output layer, f2 represents the activation function of the hidden layer, b1 represents the bias of the neurons in the hidden layer, f2 represents the activation function of the output layer, b2 represents the bias of the neurons in the output layer, and y represents the output.

[0071] A backpropagation (BP) neural network is an artificial neural network based on the backpropagation algorithm. It determines the model's accuracy by calculating an error function E, and adjusts the network weights layer by layer through backpropagation of the error signal until the difference between the output value and the actual value is minimized, thus achieving model convergence. In this process, the transfer function of the output layer typically uses a linear activation function. The error function E is calculated by summing the squares of the differences between the output value and the actual value, as shown in the following formula:

[0072]

[0073] y k λ represents the predicted average perceived temperature for broiler chickens. k This represents the actual value of the average perceived temperature of the broiler chickens, and m is the number of training data points. This error signal propagates layer by layer from the output layer to the input layer and adjusts the network weights in reverse.

[0074] The performance of surrogate models is typically evaluated based on their accuracy, efficiency, and simplicity. Accuracy, or precision, is the most important evaluation criterion; higher precision indicates a better understanding and prediction of the data points within the experimental space. Accuracy metrics for artificial neural networks generally include the correlation coefficient R and the mean squared error (MSE), calculated using the following formulas:

[0075]

[0076]

[0077] Where n represents the number of samples, y i λ represents the output value predicted by the algorithm. i The actual input values ​​of the sample represent the correlation coefficient R. Generally speaking, the closer the correlation coefficient R is to 1 and the smaller the mean square error MSE, the higher the accuracy of the neural network model.

[0078] This example uses MATLAB to implement the training process of a surrogate model, selecting 9 hidden layer neurons and employing the Levenberg-Marquardt algorithm. The model for predicting perceived temperature is constructed using MATLAB output. This model is output as a MATLAB function file, which is then converted into the corresponding programming language during the algorithm design of the chicken coop control system. The code for the function file is as follows:

[0079]

[0080]

[0081]

[0082]

[0083]

[0084]

[0085]

[0086] The regression results of the model output using MATLAB are shown in Figures 3-6.

[0087] After training, validation, and testing of the BP neural network, the model showed a high degree of fit overall, with an overall correlation coefficient of 0.99774 and R values ​​for each dataset all above 0.95, indicating that the neural network model has high accuracy.

[0088] The predicted values ​​of the constructed perceived temperature prediction model were compared with the actual sample values, and the comparison chart is shown in Figure 7.

[0089] As shown in Figure 7, the predicted perceived temperature results of the perceived temperature prediction model are close to the actual values, with relative errors all within 10% and a maximum error of no more than 1.5℃. The prediction accuracy of the perceived temperature prediction model can meet the environmental control requirements of the chicken house.

[0090] The solution to the ventilation control strategy is explained below.

[0091] After constructing the perceived temperature prediction model, it is necessary to determine the optimal control strategy for the chicken house ventilation system based on real-time environmental conditions and control requirements. Specifically, this involves determining how many fans need to be turned on and the opening angle of ventilation windows to achieve a specific average perceived temperature for broilers at different outdoor temperatures. When solving for the control strategy, the function of the perceived temperature prediction model is used as the solution function. Given the output (target average perceived temperature) and the outside air temperature, iterative calculations are performed to find the optimal combination of the number of fans and the ventilation window angle. The ventilation control combination solution process is implemented using the genetic algorithm toolbox in MATLAB. In the actual control of the chicken house, the outside ambient temperature is an uncontrollable factor, and the perceived temperature requirements of broilers of different ages vary. Therefore, when solving for the ventilation control strategy, the outside air temperature is set as a fixed constraint, and the optimization parameter range is set according to the actual range of fans and ventilation window angles. The constructed perceived temperature prediction model is used as the response function of the genetic algorithm. Under a specific temperature solution, iterative searches for extreme values ​​are performed within the range of control factors to determine the optimal control combination that satisfies the objective function.

[0092] The genetic algorithm defaults to finding the optimal solution that minimizes the function space. When solving for the optimal solution, the absolute value of the average perceived temperature output by the perceived temperature prediction model and the target average perceived temperature are used as the fitness value of the genetic function for iterative solution to obtain the optimal solution, which is the optimal ventilation control strategy (the ventilation control strategy at the target average perceived temperature).

[0093] For example, if the target average perceived temperature is 25℃, the response function is defined as the absolute error between the average perceived temperature output by the perceived temperature prediction model and 25℃. A genetic algorithm is used to find the solution corresponding to the minimum value of the response function fun(x) in the computational space. The closer the response function is to 0, the closer the perceived temperature corresponding to the solution is to 25℃.

[0094] Find the optimal solution by using boundary constraints, where

[000] represents the lower limit of the three variables and [309024] represents the upper limit of the three variables. If a range constraint is imposed on a certain variable, the range can be limited. For example,

[2000] ~[209024] will fix the outside air temperature to 20℃. Similarly, range constraints can be added to the ventilation window angle and the number of fans.

[0095] Figure 8 shows the iterative calculation process of the genetic algorithm. As can be seen from Figure 8, after 61 iterations, the optimal control scheme required to maintain the average perceived temperature of 25°C inside the building when the outside air temperature is 20°C is 5.189 fans and the ventilation window opening angle is 41.051°.

[0096] Example 2

[0097] In order to implement the method corresponding to Embodiment 1 above and achieve the corresponding functions and technical effects, a chicken coop environment control system is provided below.

[0098] Referring to Figure 9, the system includes:

[0099] The target temperature determination module 201 is used to determine the target average perceived temperature that broilers need to maintain in the chicken house.

[0100] The ventilation control strategy determination module 202 is used to determine the ventilation control strategy for maintaining the target average perceived temperature of broilers in the chicken house by using a body temperature prediction model and a genetic algorithm; the ventilation control strategy includes: the number of fans that need to be turned on in the chicken house ventilation system and the angle at which the ventilation windows need to be opened.

[0101] The perceived temperature prediction model is determined using training data and a machine learning algorithm based on a surrogate model. The training data includes different chicken house parameters and the corresponding actual average perceived temperature of the broilers. The chicken house parameters include the outside air temperature of the chicken house, the number of fans in the chicken house ventilation system that are turned on, and the opening angle of the ventilation windows in the chicken house ventilation system.

[0102] In one example, the ventilation control strategy determination module 202 specifically includes:

[0103] The solution unit is used to use the perceived temperature prediction model as the solution function of the genetic algorithm. Under the constraints of the constraints, it performs multiple iterations to obtain the ventilation control strategy for maintaining the broilers in the chicken house at the target average perceived temperature, with the objective of minimizing the difference between the average perceived temperature output by the perceived temperature prediction model and the target average perceived temperature.

[0104] The constraints are determined based on the outside air temperature, the range of the actual number of fans in the chicken house ventilation system, and the range of the actual opening angle of the ventilation windows in the chicken house ventilation system.

[0105] In one example, the chicken coop environmental control system further includes a model determination module for determining the perceived temperature prediction model.

[0106] The model determination module specifically includes:

[0107] The data acquisition unit is used to acquire training data.

[0108] The training unit is used to input the training data into the artificial neural network, with the goal of minimizing the difference between the average perceived temperature predicted by the artificial neural network and the actual average perceived temperature of the broilers. The backpropagation algorithm is used for training, and the trained artificial neural network is determined as the perceived temperature prediction model.

[0109] Example 3

[0110] This embodiment provides an electronic device, including a memory and a processor. The memory is used to store computer programs, and the processor runs the computer programs to enable the electronic device to perform the chicken coop environment control method of Embodiment 1.

[0111] Alternatively, the aforementioned electronic device may be a server.

[0112] In addition, this embodiment of the invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the chicken coop environment control method of Embodiment 1.

[0113] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0114] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for controlling the environment of a chicken coop, characterized in that, include: Determine the target average perceived temperature that needs to be maintained for broilers in the chicken house; A somatosensory temperature prediction model and a genetic algorithm were used to determine the ventilation control strategy for maintaining broilers in the chicken house at the target average somatosensory temperature. The ventilation control strategy includes: the number of fans to be turned on in the chicken house ventilation system and the angle at which the ventilation windows need to be opened; wherein, the perceived temperature prediction model is determined using training data and a machine learning algorithm based on a proxy model; the training data includes: different chicken house parameters and the corresponding actual average perceived temperature of the broilers; the chicken house parameters include: the outside air temperature, the number of fans to be turned on in the chicken house ventilation system, and the angle at which the ventilation windows in the chicken house ventilation system are opened; the ventilation control strategy for maintaining the broilers in the chicken house at the target average perceived temperature is determined using the perceived temperature prediction model and a genetic algorithm, specifically including: using the perceived temperature prediction model as the solution function of the genetic algorithm, and under constraints, performing multiple iterations to minimize the difference between the average perceived temperature predicted by the perceived temperature prediction model and the target average perceived temperature, thereby obtaining the ventilation control strategy for maintaining the broilers in the chicken house at the target average perceived temperature; wherein, the constraints are based on the outside air temperature, the range of the actual number of fans in the chicken house ventilation system, and the chicken house ventilation system... The actual range of ventilation window opening angles is determined. After constructing the perceived temperature prediction model, the optimal control strategy for the chicken house ventilation system needs to be determined based on real-time environmental conditions and control requirements. When solving the control strategy, the function of the perceived temperature prediction model is used as the solution function. Given the target average perceived temperature and the outside air temperature, iterative calculations are performed to find the optimal combination of the number of fans and the ventilation window angle. When solving the ventilation control strategy, the outside air temperature is set as a fixed constraint, and the range of optimization parameters is set according to the actual range of fan and ventilation window angles. The constructed perceived temperature prediction model is used as the response function of the genetic algorithm. Under a specific temperature solution, iterative searches for extreme values ​​are performed within the range of control factors to determine the optimal control combination that satisfies the objective function. When determining the input and output variables of the perceived temperature prediction model, the outside air temperature, the number of fans in operation, and the small window opening angle are selected as input variables, and the average perceived temperature of the broilers is selected as the output variable. The surrogate model method is used to establish the algorithm model between the above input and output variables. The formula for calculating the perceived temperature of broilers is as follows: In the formula, As perceived temperature Dry bulb temperature, For wind speed, The coefficient of performance is the air-cooling factor. Relative humidity, For target humidity, The relative humidity, humidity coefficient, and air cooling coefficient are determined based on actual chicken house measurements when calculating the perceived temperature of broilers.

2. The chicken coop environment control method according to claim 1, characterized in that, The method for determining the perceived temperature prediction model specifically includes: acquiring training data; inputting the training data into an artificial neural network, aiming to minimize the difference between the average perceived temperature predicted by the artificial neural network output and the actual average perceived temperature of the broilers, training the artificial neural network using a backpropagation algorithm, and determining the trained artificial neural network as the perceived temperature prediction model.

3. A chicken coop environment control system, characterized in that, include: The target temperature determination module is used to determine the target average perceived temperature that broilers need to maintain in the chicken house; The ventilation control strategy determination module is used to determine the ventilation control strategy for maintaining the target average perceived temperature of broilers in the chicken house by using a somatosensory temperature prediction model and a genetic algorithm. The ventilation control strategy includes: the number of fans to be turned on in the chicken house ventilation system and the angle at which the ventilation windows need to be opened; wherein, the perceived temperature prediction model is determined using training data and a machine learning algorithm based on a surrogate model; the training data includes: different chicken house parameters and the corresponding actual average perceived temperature of the broilers; the chicken house parameters include: the outside air temperature, the number of fans to be turned on in the chicken house ventilation system, and the angle at which the ventilation windows in the chicken house ventilation system are opened; the ventilation control strategy determination module specifically includes: a solution unit, used to use the perceived temperature prediction model as the solution function of a genetic algorithm, and under constraints, to perform multiple iterations to obtain the ventilation control strategy when the broilers in the chicken house maintain the target average perceived temperature, with the objective of minimizing the difference between the average perceived temperature output by the perceived temperature prediction model and the target average perceived temperature; wherein, the constraints are based on the outside air temperature, the range of the actual number of fans in the chicken house ventilation system, and the actual opening angle of the ventilation windows in the chicken house ventilation system. The range is defined; after constructing the perceived temperature prediction model, the optimal control strategy for the chicken house ventilation system needs to be determined based on real-time environmental conditions and control requirements; when solving the control strategy, the function of the perceived temperature prediction model is used as the solution function, and the optimal combination of the number of fans and the angle of the ventilation windows is iteratively calculated when the target average perceived temperature and the outside air temperature are known; when solving the ventilation control strategy, the outside air temperature is set as a fixed constraint condition, and the range of optimization parameters is set according to the actual range of fans and ventilation window angles; the constructed perceived temperature prediction model is used as the response function of the genetic algorithm, and the extreme value is iteratively searched within the range of control factors under a specific temperature solution to determine the optimal control combination that satisfies the objective function; when determining the input and output variables of the perceived temperature prediction model, the outside air temperature, the number of fans turned on, and the opening angle of the small windows are selected as input variables, and the average perceived temperature of the broilers is selected as the output variable. The algorithm model between the above input variables and output variables is established using the surrogate model method; the formula for calculating the perceived temperature of broilers is as follows: In the formula, As perceived temperature Dry bulb temperature, For wind speed, The coefficient of performance is the air-cooling factor. Relative humidity, For target humidity, The relative humidity, humidity coefficient, and air cooling coefficient are determined based on actual chicken house measurements when calculating the perceived temperature of broilers.

4. The chicken coop environment control system according to claim 3, characterized in that, Also includes: The model determination module is used to determine the perceived temperature prediction model; The model determination module specifically includes: a data acquisition unit, used to acquire training data; The training unit is used to input the training data into the artificial neural network, with the goal of minimizing the difference between the average perceived temperature predicted by the artificial neural network and the actual average perceived temperature of the broilers. The backpropagation algorithm is used for training, and the trained artificial neural network is determined as the perceived temperature prediction model.

5. An electronic device, characterized in that, The device includes a memory and a processor, the memory being used to store a computer program, and the processor running the computer program to cause the electronic device to perform the chicken coop environment control method according to any one of claims 1 to 2.

6. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the chicken house environment control method as described in any one of claims 1 to 2.

Images