The Research On Adaptive Optimal Control Model For Building Cooling And Heating Sources

Energy consumption of building cooling and heating sources, which directly has an impact on characteristic of energy consumption for HVAC systems, initial investment, operating cost and environment, accounts for a big percentage in building energy consumption for HVAC systems.In researches on optimal control for building cooling and heating sources, most researches focuse mainly at the component or sub-system level. A little information about the the global optimation for building cooling and heating sources system has been reported in literature, but it is inconveniently applied to practical engineerings because of existing drawbacks. In practical engineerings, some simple control stratages are adopted in building cooling and heating sources system, for instance the order for bringing chillers online and offline, water temperature contol and water flow contol, et al, which hardly achieve the global optimation for building cooling and heating sources system. So it is necessary to find a global optimal control for building cooling and heating sources, which not only achieves the global optimization, but also meet with requirements in practical engineerings.The researches on the optimal control for building cooling and heating sources include three main fields, namely, preparing mathematic model for HVAC systems, model parameter identification and optimal algorithm.The characteristic analysis and related mathematic models for devices exert significant impact on optimal control for building cooling and heating sources. In the research on the optimal control for building cooling and heating sources in this paper, a building cooling and heating sources in a publishing house in Changsha city is studied to prepare related models for chillers, pumps, cooling towers and boilers, respectively.To resolve the contradictory between the calculation accuracy and the calculation time, fuzzy self-tuning forgetting factor method for HVAC systems is developed to apply to optimal control of building cooling and heating sources in this study based on recursive least-squares estimation theorem. In variable forgetting factor method, different equipment models need different criterion functions for forgetting factor to keep up with time-varying parameters, but different equipment models can use a universal criterion function after using fuzzy self-tuning forgetting factor method, which not only reduce the complexity, but also keep enough accuracy. After the fuzzy control theory is applied, the calculation for parameter estimation hardly increases and the universal criterion function can be applied to different system and equipment.To validate the universality and accuracy of the fuzzy self-tuning forgetting factor method, a small-scale steam LiBr absorption chiller, an air-cooling heat pump and a screw chiller are investigated to compare the fuzzy self-tuning forgetting factor method with the forgetting factor method and other variable forgetting factor methods in relative references. The results show that the fuzzy self-tuning forgetting factor method not only satisfies the requirement for accuracy, but also keeps up with time-varying parameters rapidly.Based on system identification with self-tuning forgetting factor, the adaptive optimal control model for building cooling and heating sources is developed, which includes three parts, namely, parameter estimation model, equipment model and optimal control model. In the parameter estimation model, the recursive least-squares algorithm with self-tuning forgetting factor is applied to keep up with time-varying parameters, which can save time and reduce the calculation. In equipment model, the identification model is adopted to predict the energy consumption for equipments in HVAC system. In optimal calculation model, the penalty function is constructed to transform constrained optimization problem into unconstrained optimization problem according to the objective function and constraints. In the optimization for building cooling and heating sources, the genetic algorithm (GA) is used to deal with discrete variables and constrains, which belongs to a class of probabilistic search methods that strike a remarkable balance between exploration and exploitation of the search space and combines elements of directed and stochastic search methods.To verify the validity of the adaptive optimal control model for building cooling and heating sources, A HVAC system is studied in a publishing house to test the optimal control model for building cooling and heating sources. After optimized, the energy consumption of the HVAC system is saved by about 7%.In order to simplify the on-line optimal control for building cooling and heating sources and save calculation time, the near-optimal control strategies are summarized to simplify the adaptive optimal control model for building cooling and heating sources. Based on the simplified model, a building cooling and heating source is optimized. In this paper, the simplified control model is compared with non-simplified model. The results show that the saving energy ratio of these two models is very approximate and indicate that the simplified model not only reduce the calculation, but also improve energy efficiency.In the optimal control of building cooling and heating sources, the field control is one of the important parts. Besides ON/Off controllers, conventional PID controllers are still the dominant controllers in the field control due to their stable behavior. Howerver, conventional PID controllers are appropriate for the linear process with the certain mathematic model, and fail to reach the satified control results for the nonlinear and time-varying system.The adaptive fuzzy PID controller is designed to adjust the output of the controller according to erro and chang rate of error. A tempreture contol system in a air-conditioned room is investigated to compare the adaptive fuzzy PID controller and convemtional PID controller. The results demonstrate the fine performance and strong robustness of the proposed adaptive fuzzy PID controller.The innovations and originalities of the paper are mainly included as follows:(1) To reduce calculation time, the identification model for temperature difference between inlet water and outlet water of cooling tower is developed.(2) To resolve the contradictory between the calculation accuracy and the calculation time, fuzzy self-tuning forgetting factor method for HVAC systems is developed. After using fuzzy self-tuning forgetting factor method, different equipment models can use a universal criterion function, which not only reduce the complexity, but also keep enough accuracy. After the fuzzy control theory is applied, the calculation for parameter estimation hardly increases(3) Based on system identification with self-tuning forgetting factor, the adaptive optimal control model for building cooling and heating sources is developed.(4) In order to simplify the on-line optimal control for building cooling and heating sources and save calculation time, the simplified adaptive optimal control model for building cooling and heating sources based on previous researches.(5) The adaptive fuzzy PID controller is designed to to adjust the output of the controller on line, which improves the the fine performance and strong robustness of the control system.