Ammonia Gas Analysis And Prediction And Numerical Simulation Of Yellow-Feather Broiler Chamber In Winter

To investigate the variation of ammonia in yellow-feathered broiler chamber in winter and to control the key environmental variables in the chamber reasonably,this paper analyzes the environmental parameters of temperature,relative humidity and ammonia concentration in the house by obtaining the data of the monitoring system in different ventilation modes in the winter of two consecutive years.The three machine learning models are used to predict the ammonia concentration in the chamber.Finally,CFD technology was applied to analyze the temperature and ammonia concentration distribution,and the structure of the chamber was optimized and improved.The main contents and conclusions of this research are as follows:(1)The changes of environmental parameters in broiler chamber in winter were analyzed.After hourly processing of the collected minute data,the daily variation law of key environmental parameters is obtained,and the ammonia concentration of different heights is analyzed;the data under different frequency opening conditions of fan and fan shutdown are compared to analyze the advantages and disadvantages of different ventilation modes.The main conclusions are as follows: in the first year of winter test,the average values of temperature,relative humidity and ammonia concentration are 13.2 ℃,79.7% and 10.0mg/m3 respectively;in the second year of the winter test,the average values of temperature,relative humidity and ammonia concentration were 10.6 ℃,84.0%,3.2 mg/m3 respectively;the concentration of ammonia in the middle layer is higher than that of the upper layer,and less than that of the lower layer;in the chicken house with the fan opening frequency of 10 Hz,the average values of temperature,relative humidity and ammonia concentration are 10.4 ℃,85.5% and 3.35 mg/m3 respectively.In the chicken houses with fan opening frequency of20 Hz,the average values of temperature,relative humidity and ammonia concentration are10.2 ℃,75.4% and 1.6mg/m3 respectively,that is,under winter conditions condition,the frequency of 10 Hz fan can meet the needs of the control in the chamber.(2)Pearson correlation coefficient was used to analyze the correlation between temperature and humidity at different monitoring points and ammonia concentration.It was found that the temperature and humidity at the outlet of internal circulation had a very significant correlation with ammonia concentration in the chamber(P< 0.01).The temperature,relative humidity,ammonia concentration,age and feeding time of the circulating air outlet in the chicken house were taken as the input variables of the model,and the ammonia concentration was predicted using the DNN model PSO-DNN model and RF algorithm.The results showed that the mean absolute error of PSO-DNN model was 1.89mg/m3,and the prediction accuracy was the highest.When the PSO-DNN model was applied to the experiment under different ventilation modes in the later stage,it was found that the prediction results of the model were highly adaptable and could meet the needs of ammonia gas early warning.(3)CFD technology and ammonia diffusion were used to carry out steady state numerical simulation,the simulation found that the chicken living layer temperature uniformity is good.The circulation channels inside and outside the broiler chamber were added into the model to simulate the air flow.Ammonia concentration changes were simulated under the influence of temperature at three times of the day,the relationship between the two correlation,and found that ammonia since floor layer formation.The correlation between the two was verified.And it was found that after the ammonia was generated from the floor layer,most of the gas rose to the top floor and then was extracted by the negative pressure fan after descending.After optimizing the side window fan,the top layer of ammonia could be directly extracted by the side window fan,which improved the ammonia environment in the chamber.