Research On The Optimal Configuration Of Combined Cooling,Heating And Power System Considering Dynamic Tracking Performance Of The Device

The configuration optimization of the integrated energy system(IES)is the basis of the scheduling and control.However,in previous papers,most researchers underestimated the role of considering device dynamics at the configuration stage.When the configuration plan is completed,the dynamic characteristics of the device are basically determined,which also leads to the determination of the optimal boundary of the system scheduling and control problems,and the advantages of the IES cannot be fully exploited.Therefore,this paper proposes a configuration optimization method that takes into account the frequency domain index used to describe device dynamic tracking performance,in order to improve the dynamic characteristics of the IES from the root.The main research contents of this paper include the following aspects:1.An evaluation method that takes into account the dynamic tracking performance of device in frequency domain is proposed.In this method,in order to describe the difference of dynamic performance between different devices,from the perspective of output correlation and responding speed,this paper innovatively proposes a frequency domain dynamic index composed of the resonant frequency and the resonance peak value of the device.Simultaneously,this paper uses the subjective and objective weighting method(analytic hierarchy processentropy weighting method)to determine the weights of each index,and establishes the objective function for the configuration optimization of the IES in the form of weighting.2.The steady-state model and closed-loop control system model of the device suitable for configuration optimization are established,which lays a model foundation for configuration optimization research.Based on the mechanism model of the main device,the closed-loop control system model is obtained through data identification and PI controller parameter optimization,which can provide a model basis for the calculation of the frequency domain index used to describe device dynamic tracking performance.3.A deterministic configuration optimization method that takes into account the frequency domain indicator used to describe dynamic tracking performance is proposed.The method takes the comprehensive index embedded in the frequency domain index as the objective function,constructs equipment characteristic constraints and energy balance constraints,and obtains the configuration optimization model.Through case analysis,the impact of dynamic tracking performance indicator on configuration results is studied,and the differences in the configuration results under different index priorities are compared and analyzed.The research results show that the energy supply response of the configuration scheme obtained by considering the dynamic performance of the device is faster,and the imbalance of supply and demand in the dynamic process of this scheme is significantly reduced,but the economic cost of the scheme is also greater.4.A configuration optimization method that comprehensively considers the dynamic tracking performance of device and the uncertainty of renewable energy output is proposed.Based on the t location-scale distribution description of the error,this paper generates scenes with time correlation through multivariate normal standard sampling,and combines the kmeans clustering method and Calinski-Harabasz criterion to perform scene clustering.Based on the above results,this paper establishes a stochastic optimization model and a multi-discrete scene distribution robust model that take into account the dynamic tracking performance and the uncertainty of renewable energy output,respectively.Through case analysis,the result shows that the dynamic performance index and economic index of the scheme considering the uncertainty increase,and the economy and dynamic performance become worse,while the scheme considering dynamic tracking performance and uncertainty can ensure that the energy supply can quickly meet the demand.Considering the uncertainty of renewable energy and the uncertainty of probability distribution,the configuration result of distributionally robust optimization is more conservative than that of stochastic optimization.