A Multi-Objective Optimization Method for Lateral Ventilation in Chicken Houses Based on MOPSO and NS-MFO Algorithms

Abstract

This invention relates to a multi-objective optimization method for lateral ventilation in chicken houses based on MOPSO and NS-MFO algorithms, belonging to the field of scientific animal husbandry technology. This invention studies the relationship between lateral ventilation methods in chicken houses during winter and the internal environment of caged, enclosed chicken houses, as well as the corresponding flock survival rate. Considering both economic efficiency and flock comfort, a multi-objective programming model with multi-factor constraints for lateral ventilation in caged, enclosed chicken houses during winter is qualitatively constructed. This model is then solved using the multi-objective particle swarm optimization (MOPSO) algorithm and the non-dominated moth-flame algorithm (NS-MFO), yielding a relatively optimal method for managing lateral ventilation in chicken houses during winter. This invention can be widely applied in scientific animal husbandry settings.

Description

Technical Field

[0001] This invention relates to a multi-objective optimization method for transverse ventilation in chicken houses based on MOPSO and NS-MFO algorithms, belonging to the field of scientific breeding technology. Background Technology

[0002] In winter raising caged chickens, the organic control of the internal environment of the chicken house is crucial. If the temperature inside the chicken house is low and the humidity is too high, the chickens will increase their feed intake to maintain their perceived temperature, thereby reducing the feed conversion ratio. Low temperature and high humidity will cause water vapor condensation, increasing the incidence of arthritis and negatively impacting the survival rate and egg production. However, if ventilation is neglected in order to prevent heat loss from the chicken house, sufficient fresh air and oxygen cannot be provided to the chickens, leading to excessive levels of harmful gases such as carbon dioxide and nitrogen, increasing the incidence of respiratory diseases, and also affecting the survival rate. Currently, horizontal ventilation is a commonly used environmental control method. Specifically, it involves sealing all possible air intakes in the caged chicken house, such as the longitudinal air inlets, evaporative cooling pads, and gaps in doors and windows, and evenly distributing several small rectangular air intake windows on both side walls. Adjusting the airflow from these windows and the exhaust fans regulates the temperature and humidity inside the chicken house.

[0003] However, there is currently no systematic quantitative research on the specific impact of factors such as the opening degree and angle of small windows, the number of fans turned on, the fan duty cycle, and the fan air volume on the indoor environment in transverse ventilation. Therefore, in the process of regulation, it is impossible to guarantee that the best ventilation effect can be achieved with the minimum economic cost. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing management methods and provide a multi-objective optimization method for cross ventilation in chicken houses based on particle swarm optimization and non-dominated moth flame algorithms, using MOPSO and NS-MFO algorithms.

[0005] The multi-objective optimization transverse ventilation method for chicken houses based on MOPSO and NS-MFO algorithms described in this invention includes the following steps:

[0006] S1. To study the relationship between the horizontal ventilation method of chicken houses in winter and the internal environment of caged, enclosed chicken houses and the survival rate of chickens.

[0007] S2. Taking into account both economic efficiency and flock comfort, a multi-objective programming model with multi-factor constraints for cross ventilation in chicken houses during winter is qualitatively constructed.

[0008] S3. Based on the MOPSO algorithm and NS-MFO algorithm, the multi-objective programming model is solved to obtain a relatively better winter chicken house horizontal ventilation management mode;

[0009] S4. Adjust the current cross ventilation management mode based on the obtained optimal value.

[0010] Preferably, in the multi-objective programming model with multi-factor constraints involved in step S2, the multi-factor constraint characteristic refers to the fact that the established programming model has n multi-decision variables x = (x1, x2, x3, ..., x...). n ), where x1, x2, x3, ..., x n The variable factors include the opening degree of the small window, the angle of the small window, the number of fans turned on, the duty cycle of the fans, the air volume of the fans, and the outdoor temperature and humidity.

[0011] Preferably, in the multi-objective programming model with multi-factor constraints involved in step S2, the multi-objective programming problem refers to the m multi-objective vectors y = (y1, y2, y3, ..., y4) of the established programming model. m ), where y1, y2, y3, ..., y m The objectives of regulation include economic costs, chicken house environment, flock survival rate, and egg production.

[0012] Preferably, the multi-objective programming model established in step S2 is expressed as follows:

[0013]

[0014] In the formula, x = (x1, x2, x3, ..., x n Let y = (y1, y2, y3, ..., y4) be the decision variable. m ) is the target vector;

[0015] Multiple transverse ventilation experiments were conducted to establish x = (x1, x2, x3, ..., x n ) and y = (y1, y2, y3, ..., y m The constraints between these factors include the mathematical relationships between flock survival rate and window opening, window angle, number of fans, fan duty cycle, fan air volume, and outdoor temperature and humidity; the mathematical relationships between chicken house maintenance costs and window opening, window angle, number of fans, fan duty cycle, fan air volume, and outdoor temperature and humidity; and the mathematical relationships between egg production and window opening, window angle, number of fans, fan duty cycle, fan air volume, and outdoor temperature and humidity.

[0016] This is used to establish g k (x), h j (x), g k (x), h j (x) represents the P and Q inequality constraints and equality constraints that need to be satisfied;

[0017] Among them, g k(x) includes ensuring that the flock survival rate is not less than the set value, the egg production is not less than the set value, and the economic cost is not greater than the budgeted amount, h j (x) includes ensuring that the air volume at the fan outlet is consistent with the air volume at the small window, and that changes in indoor temperature and humidity are consistent with changes in air freshness.

[0018] Preferably, the MOPSO algorithm in step S3 is based on the principle of massless particle simulation of bird flocks. It finds the optimal solution through cooperation and information sharing among individual birds in the flock. Each particle is considered as a candidate solution of the model described in step S2. The particle has two attributes: velocity and position. Velocity reflects the speed of movement, and position reflects the direction of movement. During the iteration process, the particle continuously searches for the position p with the best individual fitness. best and the position g with the best global fitness best And update the particle properties; the particle updates its velocity and position in the following ways:

[0019]

[0020]

[0021] In the formula, a1 and a2 are acceleration factors, c1 and c2 are learning factors, ω is the inertia weight, and t is the current iteration number.

[0022] Preferably, the MOPSO algorithm in step S3 includes the following steps:

[0023] S31. Initialize the velocity and position of the particle swarm in the multi-objective programming model for cross ventilation in the chicken house, and set the parameters;

[0024] S32. Set the current position of the particles in the initial particle population of the multi-objective programming model for horizontal ventilation in the chicken coop as p. best ;

[0025] S33. Calculate the objective function value of each particle in the particle swarm of the multi-objective programming model for cross ventilation in the chicken house, and update the non-dominated solutions in the particle swarm to the external storage set.

[0026] S34. Update the optimal position p of individual particles in the multi-objective programming model for cross ventilation in the chicken coop. best Select the global optimal position g of the particle swarm best ;

[0027] S35. Update the position and velocity of particles in the multi-objective programming model for cross ventilation in chicken coops;

[0028] S36. Calculate the updated particle objective function value and update the non-dominated solutions in the particle swarm to the external storage set.

[0029] S37. Determine whether the maximum number of iterations has been reached. If the result is yes, randomly select a solution from the external storage set as a feasible ventilation management scheme for the chicken coop horizontal ventilation multi-objective programming model. If the result is no, return to step S34.

[0030] Preferably, the MOPSO algorithm described in step S3 is used to solve the local optimization problem, in which the three decision variables are the window opening, the number of fans turned on, and the outdoor temperature and humidity, and a preliminary optimization solution set for the window opening, the number of fans turned on, and the outdoor temperature and humidity is obtained.

[0031] This preliminary optimized solution set is used as a new equality constraint for the global optimization problem solved by the NS-MFO algorithm described in step S3, and the global optimal solution is calculated using the NS-MFO algorithm.

[0032] Preferably, the NS-MFO algorithm described in step S3 is based on the principle that moths, which are active at night, rely on light for navigation. When a light source is placed around the moth, it will eventually fly towards the light source. The steps for solving the multi-objective programming model for lateral ventilation in a chicken coop are as follows:

[0033] S38. The system randomly generates a population of moths under the original multi-factor ventilation conditions. Based on the original multi-factor ventilation conditions population of moths, the fitness ranking and the number of optimal ventilation conditions are calculated, and the optimal ventilation condition subarray is constructed.

[0034] S39. Multi-factor ventilation conditions: Moths spirally approach the optimal ventilation conditions, updating the second-generation moth population.

[0035] S310. After the update is completed, merge the unsorted multi-factor ventilation condition moth populations according to the number of optimal ventilation conditions. After sorting, divide the first half of the sorted population into subarrays based on optimal ventilation conditions.

[0036] S311. Under multi-factor ventilation conditions, moths spirally approach the optimal ventilation conditions, update the population position, and calculate the fitness values ​​of the current multi-factor ventilation condition moth population to prepare for the next iteration.

[0037] S312. Determine whether to end. If the threshold of near-optimal ventilation conditions has not been reached, return to step S310. Otherwise, terminate the iteration and record the population location information.

[0038] Preferably, step S3 involves a relatively better winter chicken house transverse ventilation management mode, which regulates variable factors such as window opening, window angle, number of fans, fan duty cycle, fan air volume, and outdoor temperature and humidity through a determined feasible scheme.

[0039] The beneficial effects of this invention are as follows: The multi-objective optimization method for lateral ventilation in chicken houses based on MOPSO and NS-MFO algorithms described in this invention studies the relationship between the lateral ventilation mode of chicken houses in winter and the internal environment of caged closed chicken houses, as well as the corresponding survival rate of the chicken flock. Taking into account the objectives of economy and chicken flock comfort, a multi-objective programming model with multi-factor constraints for lateral ventilation in caged closed chicken houses in winter is qualitatively constructed. The multi-objective programming model is solved based on the multi-objective particle swarm optimization algorithm (MOPSO) and the non-dominated moth-flame algorithm (NS-MFO) to obtain a relatively better method for lateral ventilation management in chicken houses in winter. Attached Figure Description

[0040] Figure 1 This is a flowchart illustrating the overall process principle of the present invention.

[0041] Figure 2 This is a flowchart of the MOPSO algorithm of this invention.

[0042] Figure 3 This is a flowchart of the NS-MFO algorithm calculation of the present invention. Detailed Implementation

[0043] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0044] The purpose of this invention is to overcome the shortcomings of the existing management methods and provide a multi-objective optimization method for cross ventilation in chicken houses based on particle swarm optimization and non-dominated moth flame algorithms, using MOPSO and NS-MFO algorithms.

[0045] The purpose of the present invention can be achieved by the following technical solutions:

[0046] A multi-objective optimization method for transverse ventilation in chicken coops based on particle swarm optimization and non-dominated moth-flame algorithms, using MOPSO and NS-MFO algorithms, is described below. The specific implementation steps are as follows: Figure 1 As shown, that is:

[0047] S1. To study the relationship between the horizontal ventilation method of chicken houses in winter and the internal environment of caged, enclosed chicken houses and the survival rate of chickens.

[0048] S2. Taking into account both economic efficiency and flock comfort, a multi-objective programming model with multi-factor constraints for lateral ventilation in winter in caged, enclosed chicken houses is qualitatively constructed.

[0049] S3. Based on the MOPSO algorithm and NS-MFO algorithm, the multi-objective programming model is solved to obtain a relatively better management method for horizontal ventilation in chicken houses during winter.

[0050] S4. Adjust the current transverse ventilation mode according to the obtained optimal value.

[0051] According to one embodiment of the present invention, in the multi-objective programming model with multi-factor constraint characteristics involved in steps S1 and S2, the multi-factor constraint characteristics mainly refer to the n multi-decision variables x = (x1, x2, x3, ..., xn) of the established programming model. n ), where x1, x2, x3, ..., x n Variable factors include, but are not limited to, the opening degree of the small window, the angle of the small window, the number of fans turned on, the duty cycle of the fans, the air volume of the fans, and the outdoor temperature and humidity.

[0052] In the multi-objective programming model with multi-factor constraints involved in steps S1 and S2, the multi-objective programming problem mainly refers to the m multi-objective vectors y = (y1, y2, y3, ..., y4) of the established programming model. m ), where y1, y2, y3, ..., y m This includes, but is not limited to, controlling economic costs, chicken house environment, flock survival rate, egg production, and other objectives.

[0053] According to one embodiment of the present invention, the multi-objective programming model established in step S2 is specifically represented as follows:

[0054]

[0055] Where x = (x1, x2, x3, ..., x n Let y = (y1, y2, y3, ..., y4) be the decision variable. m Let x be the target vector. Multiple sets of transverse ventilation experiments were conducted, establishing x = (x1, x2, x3, ..., x...). n ) and y = (y1, y2, y3, ..., y m The constraints between these parameters include, but are not limited to, the mathematical relationships between flock survival rate and window opening, window angle, number of fans operating, fan duty cycle, fan air volume, and outdoor temperature and humidity; the mathematical relationships between chicken house maintenance costs and window opening, window angle, number of fans operating, fan duty cycle, fan air volume, and outdoor temperature and humidity; and the mathematical relationships between egg production and window opening, window angle, number of fans operating, fan duty cycle, fan air volume, and outdoor temperature and humidity. Based on these relationships, g is established. k (x), h j (x). g k (x), h j (x) represents the P and Q inequality and equality constraints that need to be satisfied. Where g k (x) Including but not limited to ensuring that the flock survival rate is not less than the set value, the egg production is not less than the set value, and the economic cost is not greater than the budgeted amount, h j(x) Including but not limited to ensuring that the air volume at the fan outlet is consistent with the air volume at the small window, and that changes in indoor temperature and humidity are consistent with changes in air freshness.

[0056] Example 1:

[0057] According to one embodiment of the present invention, the MOPSO solution algorithm described in step S3 is based on the idea of ​​designing massless particles to simulate a flock of birds, and finding the optimal solution through cooperation and information sharing among the individual birds. Each particle can be considered as a candidate solution of the model described in step S2. The particles have two attributes: velocity and position. Velocity reflects the speed of movement, and position reflects the direction of movement. During the iteration process, the particles continuously search for the position p with the best individual fitness. best and the position g with the best global fitness best And update the particle properties. The way a particle updates its velocity and position is as follows:

[0058]

[0059]

[0060] Where a1 and a2 are acceleration factors, c1 and c2 are learning factors, ω is the inertia weight, and t is the current iteration number.

[0061] Example 2:

[0062] According to one embodiment of the present invention, the NS-MFO solution algorithm described in step S3 has the following specific steps: Figure 2 As shown, that is:

[0063] S31: Initialize the velocity and position of the particle swarm in the multi-objective programming model for cross ventilation in the chicken house, and set the parameters.

[0064] S32: Set the current position of the particles in the initial particle population of the multi-objective programming model for cross ventilation in the chicken coop to p. best .

[0065] S33: Calculate the objective function value of each particle in the particle swarm of the multi-objective programming model for cross ventilation in chicken coop, and update the non-dominated solutions in the particle swarm to the external storage set.

[0066] S34: Update the optimal position p of individual particles in the multi-objective programming model for cross ventilation in chicken coops. best Select the global optimal position g of the particle swarm best .

[0067] S35: Update the position and velocity of particles in the multi-objective programming model for cross ventilation in chicken coops.

[0068] S36: Calculate the updated particle objective function value and update the non-dominated solutions in the particle swarm to the external storage set.

[0069] S37: Determine whether the maximum number of iterations has been reached. If the result is yes, randomly select a solution from the external storage set as a feasible ventilation management scheme for the chicken coop horizontal ventilation multi-objective programming model. If the result is no, return to step S3-4.

[0070] According to one embodiment of the present invention, the MOPSO algorithm described in step S3 is used to solve a local optimization problem. The three decision variables of this local optimization problem are the window opening, the number of fans turned on, and the outdoor temperature and humidity. A preliminary optimized solution set for the window opening, the number of fans turned on, and the outdoor temperature and humidity is obtained. This preliminary optimized solution set is used as a new equality constraint for the global optimization problem solved by the NS-MFO algorithm described in step S3, and the NS-MFO algorithm is used to calculate the global optimal solution.

[0071] Example 3:

[0072] According to one embodiment of the present invention, the MOPSO solution algorithm described in step S3 is based on the idea that nocturnal moths navigate by light. By placing a light source around the moth, it will eventually fly towards the light source. The specific steps for solving the multi-objective programming model for lateral ventilation in a chicken coop are as follows: Figure 3 As shown, it can be briefly summarized into the following 5 steps:

[0073] S38: The system randomly generates a population of moths under the original multi-factor ventilation conditions, calculates the fitness ranking and the number of optimal ventilation conditions based on the original multi-factor ventilation conditions population, and constructs an optimal ventilation condition subarray.

[0074] S39: Multi-factor ventilation conditions: Moths spirally approach the optimal ventilation conditions, updating the second-generation moth population.

[0075] S310: After the update is completed, merge the unsorted multi-factor ventilation condition moth populations according to the number of optimal ventilation conditions. After sorting, divide the first half of the sorted population into subarrays based on optimal ventilation conditions.

[0076] S311: Moths under multi-factor ventilation conditions spiral towards the optimal ventilation conditions, update the population position, and calculate the fitness values ​​of the current multi-factor ventilation condition moth population to prepare for the next iteration.

[0077] S312: Determine whether to end. If the threshold of near-optimal ventilation conditions has not been reached, return to step 3; otherwise, terminate the iteration and record the population location information.

[0078] Example 4:

[0079] According to one embodiment of the present invention, in step S4, the multi-objective planning model for the transverse ventilation of the chicken house is jointly solved according to the two multi-objective optimization problem solving methods described in step S3. The determined feasible scheme is used to regulate variable factors such as the opening degree of the small window, the angle of the small window, the number of fans turned on, the duty cycle of the fans, the air volume of the fans, and the outdoor temperature and humidity.

[0080] This invention proposes a method for lateral ventilation of chicken houses in winter based on multi-objective optimization particle swarm optimization algorithm. Taking into account both economic efficiency and the real-time comfort of the flock, a multi-objective programming model for lateral ventilation of caged chicken houses in winter is constructed. The optimal solution is obtained by jointly solving the model using the MOPSO algorithm and the NS-MFO algorithm, so as to minimize the operating cost of caged chicken houses, maximize the survival rate of chickens, and maximize the egg production of chickens in winter.

[0081] This invention can be widely used in scientific aquaculture.

Claims

1. A multi-objective optimization method for transverse ventilation in chicken houses based on MOPSO and NS-MFO algorithms, characterized in that, Includes the following steps: S1. To study the relationship between the horizontal ventilation method of chicken houses in winter and the internal environment of caged, enclosed chicken houses and the survival rate of chickens. S2. Taking into account both economic efficiency and flock comfort, a multi-objective programming model with multi-factor constraints for cross ventilation in chicken houses during winter is qualitatively constructed. S3. Based on the MOPSO algorithm and NS-MFO algorithm, the multi-objective programming model is solved to obtain a relatively better winter chicken house horizontal ventilation management mode; S4. Adjust the current cross ventilation management mode based on the obtained optimal value; In the multi-objective programming model with multi-factor constraints involved in step S2, the multi-factor constraint characteristic refers to the fact that the established programming model has n multi-decision variables x = (x1, x2, x3, …, x…). n ), where x1, x2, x3, …, x n The variable factors include the opening degree of the small window, the angle of the small window, the number of fans in operation, the duty cycle of the fans, the air volume of the fans, and the outdoor temperature and humidity. In the multi-objective programming model with multi-factor constraints involved in step S2, the multi-objective programming problem refers to the problem of the established programming model with m multi-objective vectors y = (y1, y2, y3, …, y m ), where y1, y2, y3, …, y m The objectives of regulation include economic costs, chicken house environment, flock survival rate, and egg production; The multi-objective programming model established in step S2 is expressed as follows: (1) In the formula, x = ( x 1, x 2, x 3, …, x n ) is the decision variable. y = ( y 1, y 2, y 3, …, y m ) is the target vector; Multiple sets of transverse ventilation experiments were conducted to establish x = ( x 1, x 2, x 3, …, x n )and y = ( y 1, y 2, y 3, …, y m The constraints between these factors include the mathematical relationships between flock survival rate and window opening, window angle, number of fans, fan duty cycle, fan air volume, and outdoor temperature and humidity; the mathematical relationships between chicken house maintenance costs and window opening, window angle, number of fans, fan duty cycle, fan air volume, and outdoor temperature and humidity; and the mathematical relationships between egg production and window opening, window angle, number of fans, fan duty cycle, fan air volume, and outdoor temperature and humidity. Established in this way g k ( x ), h j ( x ), g k ( x ), h j ( x ) is what needs to be satisfied P , Q One inequality constraint and one equality constraint; in, g k ( x This includes ensuring that the survival rate of the flock is not less than the set value, the egg production is not less than the set value, and the economic cost does not exceed the budgeted amount. h j ( x This includes ensuring that the air volume output from the fan is consistent with the air volume intake through the small window, and that changes in indoor temperature and humidity are consistent with changes in air freshness. The MOPSO algorithm in step S3 is based on the principle of massless particle simulation of bird flocks. It finds the optimal solution through cooperation and information sharing among individual birds in the flock. Each particle is considered a candidate solution of the model described in step S2. The particle has two attributes: velocity and position. Velocity reflects the speed of movement, and position reflects the direction of movement. During the iteration process, the particle continuously searches for the position with the best individual fitness. p best and the position with the best global fitness g best And update the particle properties; the particle updates its velocity and position in the following ways: (2) (3) In the formula, a 1. a 2 is the acceleration factor. c 1. c 2 is the learning factor. ω The inertial weight is t, where t is the current iteration number. The MOPSO algorithm in step S3 includes the following steps: S31. Initialize the velocity and position of the particle swarm in the multi-objective programming model for cross ventilation in the chicken house, and set the parameters; S32. Set the current position of the particles in the initial particle population of the multi-objective programming model for cross ventilation in the chicken coop as... p best ; S33. Calculate the objective function value of each particle in the particle swarm of the multi-objective programming model for cross ventilation in the chicken house, and update the non-dominated solutions in the particle swarm to the external storage set. S34. Update the optimal position of individual particles in the multi-objective programming model for cross ventilation in the chicken coop. p best Select the global optimal position of the particle swarm. g best ; S35. Update the position and velocity of particles in the multi-objective programming model for cross ventilation in chicken coops; S36. Calculate the updated particle objective function value and update the non-dominated solutions in the particle swarm to the external storage set. S37. Determine whether the maximum number of iterations has been reached. If the result is yes, randomly select a solution from the external storage set as a feasible ventilation management scheme for the chicken coop horizontal ventilation multi-objective programming model. If the result is no, return to step S34.

2. The multi-objective optimization method for transverse ventilation in chicken houses based on MOPSO and NS-MFO algorithms according to claim 1, characterized in that, The MOPSO algorithm described in step S3 is used to solve the local optimization problem. The three decision variables of the local optimization problem are the opening of the small window, the number of fans turned on, and the outdoor temperature and humidity. The preliminary optimization solution set of the opening of the small window, the number of fans turned on, and the outdoor temperature and humidity is obtained. This preliminary optimized solution set is used as a new equality constraint for the global optimization problem solved by the NS-MFO algorithm described in step S3, and the global optimal solution is calculated using the NS-MFO algorithm.

3. The multi-objective optimization transverse ventilation method for chicken houses based on MOPSO and NS-MFO algorithms according to claim 2, characterized in that, The NS-MFO algorithm described in step S3 is based on the principle that moths, which are active at night, rely on light for navigation. By placing a light source around the moth, it guides the moth to fly towards the light source. The steps for solving the multi-objective programming model for lateral ventilation in a chicken coop are as follows: S38. The system randomly generates a population of moths under the original multi-factor ventilation conditions. Based on the original multi-factor ventilation conditions population of moths, the fitness ranking and the number of optimal ventilation conditions are calculated, and an optimal ventilation condition subarray is constructed. S39. Multi-factor ventilation conditions: Moths spirally approach the optimal ventilation conditions, updating the second-generation moth population. S310. After the update is completed, merge the unsorted multi-factor ventilation condition moth populations according to the number of optimal ventilation conditions. After sorting, divide the first half of the sorted population into subarrays based on optimal ventilation conditions. S311. Under multi-factor ventilation conditions, moths spirally approach the optimal ventilation conditions, update the population position, and calculate the fitness values ​​of the current multi-factor ventilation condition moth population to prepare for the next iteration. S312. Determine whether to end. If the threshold of near-optimal ventilation conditions has not been reached, return to step S310. Otherwise, terminate the iteration and record the population location information.

4. The multi-objective optimization transverse ventilation method for chicken houses based on MOPSO and NS-MFO algorithms according to claim 1 or 3, characterized in that, In step S3, a relatively optimal winter chicken house transverse ventilation management mode is involved. This mode regulates variable factors such as window opening, window angle, number of fans, fan duty cycle, fan air volume, and outdoor temperature and humidity through a determined feasible scheme.

Images