Research On MFAC Of Variable Air Volume System Based On Swarm Intelligence Algorithm

At this stage,China is on the road of industrial development and urbanization,the demand for electricity resources is gradually increasing,and social energy problems are becoming more and more frequent,while the continuous increase in construction energy consumption and energy consumption brought about by social resources,environmental protection problems,has become a major problem in the field of urban construction.At present,most buildings and office buildings generally use fixed-air conditioning systems,whose energy consumption can usually reach more than 50% of the total energy consumption of the building.The variable air conditioning system,because of its power saving,easy to control,good comfort and other advantages,has been widely used in many large and medium-sized public,civil and industrial buildings,and gradually replace the fixed air volume system,from the world’s view of its share also showed a gradual increase in the trend,the relevant researchers and experts on the optimal control of variable air conditioning system research results are increasingly.At present,the control method of VAV system used in domestic buildings is more common in the PID control method.Under the condition that the controlled equipment is simple and the working condition is stable,the traditional conventional PID control method can obtain the more satisfactory control performance requirements.But for the VAV system,VAV system has a great lag,loop highly coupled characteristics,and due to the large and complex condition of its own equipment and more interference factors,it is difficult to establish an accurate VAV system mechanism model,so the control performance of the PID control method is limited,the design of the controller is easy to control the VAV system when the work of lagging conditions,not satisfactory The control effect is not satisfactory,and the comfort in the room is not satisfactory.For the above problems,this paper proposes to use a new model-free adaptive control(MFAC)method to control the stable operation of VAV system.Since the four key parameters of the MFAC control method have a significant impact on the good or bad control performance,the key parameters of MFAC are rectified using a hybrid krill algorithm.The main research work accomplished in this paper is divided into the following points:(1)For the variable air volume system,which is a complex multi-input and multi-output system with great lag and highly coupled loops,it is difficult to obtain an accurate variable air volume model by analyzing the principle of the system to construct the mathematical relationship between variables,so this paper proposes a method based on singular spectrum analysis and temporal convolution network to establish its data model for the existing variable air volume system,and the data model is used for the later The data model lays an important foundation for the data-driven control of the variable air volume system.(2)To address the problem that the PID control method is difficult to effectively control the stable operation of the VAV system,this paper adopts the more advanced Model-free Adaptive Control(MFAC)which does not require the system mechanism model.Since MFAC has four key parameters that significantly affect the control performance,but these four parameters are difficult to be adjusted,this paper proposes a new intelligent optimization algorithm,Particle Swarm Optimization-Krill Herd(PSOKH)algorithm,to automatically adjust these four parameters.This paper proposes a new intelligent optimization algorithm,Particle Swarm Optimization-Krill Herd(PSOKH)algorithm,to automatically adjust these four parameters.The MFAC method based on the hybrid krill algorithm(PSOKH-MFAC)reduces the negative impact of artificial adjustment of the set parameters on the control performance and makes the control effect more stable and robust.(3)The complete VAV system has several loops,and it is impractical to consider the loops of the whole system and control them directly.Therefore,the research object of this paper is two very important loops in the VAV system: the inverter-fan-duct static pressure loop and the fresh air valve-carbon dioxide concentration loop.To complete the effective control of these two coupled loops,a data model of the VAV system based on singular spectrum analysis and time convolution network is firstly used,and then the model-free adaptive control with rectified parameters by hybrid krill swarm algorithm is used to control the smooth operation between the loops of end devices.