Research On Load Forecasting Of Central Air Conditioning And Optimization Control Of Chilled Water System

Central air-conditioning wastes huge energy while bringing people a comfortable indoor environment.Studying its energy-saving measures will generate considerable social and economic value.In the actual operation of the central air conditioner,due to the constant changes in weather,human flow and other factors,the cooling capacity of the cold machine changes at any time.In this case,the chilled water system that relies only on the negative feedback adjustment will deteriorate the control effect..In order to solve the energy waste and improve the control effect,this paper proposes a feedforward-based optimistic control method for chilled water system.The method obtains the load forecast value at the next moment through the load forecasting model,and brings the predicted value into the energy-optimization model for energy-saving optimization,and then optimizes the optimized chilled water supply and return water temperature difference with the current temperature difference.The difference is reduced as the feedforward compensation temperature difference,and the improved ant colony algorithm is used to set the temperature difference control of the chilled water system by the PID controller.Through simulation experiments,it is found that using the method proposed in this paper,the temperature difference output value is basically advanced to meet the energy consumption optimization requirements temperature difference,thus verifying the feasibility of energy saving and improving control effect.The specific research contents are as follows:(1)For the load forecasting,there are few training samples,the input variables are short,and there are strong nonlinear problems between the variables.The least squares support vector machine model is used for load forecasting.In order to improve the accuracy of prediction,this paper uses particle swarm optimization and improved particle swarm optimization to optimize ? and C values.To verify the generalization ability of prediction models,this paper uses these three prediction models for October 17 and October.The simulation was carried out on the 23 rd and the 2nd of November.Through simulation comparison,the least squares support vector machine model based on improved particle swarm optimization algorithm has the highest prediction accuracy and the strongest generalization ability in these three days.Therefore,this model is used as the feedforward-based optimization control of chilled water system.Load forecasting model.(2)The data provided for the competition lacks the parameters required for the model in the process of modeling.The system identification method is used to model the chilled water system transfer function and the energy consumption of the chilled water pump based on the existing data.The energy consumption model of the chiller is established,and the accuracy of the model is verified by the evaluation index or test data.The results show that the model established in this paper has high accuracy.(3)In order to provide feedforward compensation temperature difference and temperature difference setting for optimal control of chilled water system,geneticalgorithm is used to optimize the total energy consumption model of chilled water pump and chiller,and the chilled water temperature difference,flow rate and total energy obtained after optimization are optimized.The consumption is summarized into a table.(4)In order to achieve rapid,accurate and stable control effect on the chilled water system,the improved ant colony algorithm is used to clarify the PID controller parameters and the basic ant colony algorithm is added as a comparison.The results show that the improved ant colony algorithm control effect Better to meet the chilled water system control requirements.(5)In order to verify the feedforward-based chilled water system optimization control method proposed in this paper,energy saving and control effect can be realized,and information such as PID controller parameters and feedforward compensation temperature difference set by the improved ant colony algorithm is brought into the freezing.Temperature difference control is performed in the water system control model.The simulation results show that the chilled water system can control the temperature difference output value to reach the target value according to the energy consumption optimization required temperature difference.