| Carbon peak and carbon neutrality is the national important layout during construction.Wind power,solar power and hydropower are renewable and clean energy,which are complementarty to each other in the time scale of year and day.The constrcution of windsolar-hydro(WSH)multi-energy hybrid system plays a decisive role in realizing the maximum comprehensive benefit of the basin,protecting the natural environment and promoting the energy transformation and reform in China.The combined operation of the WSH hybrid system in the basin is affected by various factors such as meteoroloial processes,hydrological cycles,sub-power generation systems,and water and electricity demand.As more and more wind power and photovoltaics are integrated into the power grid,the pressure of peak and frequency modulation of the power grid has been intensified,and the safe and stable operation of the power system has been affected.The combined power transmission of hydropower,wind power,and photovoltaics is an effective way to solve intermittent energy power transmission,and it also brings new connotations and challenges to the research work of reservoirs.Therefore,how to suppress the frequent flucatuations of wind and solar power output and ensure the safe and stable operation of the power system is a keyscientific issue for the development of wind and solar energy.On the other hand,the complementary operation of WSH hybrid system is affected by multiple uncertain factors such as wind speed,solar radiation intensity,runoff and power load.How to accurately quantify the uncertain components of the hybrid system and consider these uncertainties in the optimal operation model is another key scientific problem.This research takes the WSH complementary base in the Yalong River Basin as the research object,and conducts research from the aspects of probabilistic forecasting of uncertain components,efficient solution algorithm for optimal operation,construction and solution of probabilistic optimal operation model,and multi-attribute decision-making.Relevant research results can provide theoretical basis and technocal support for optimal operation and multi-attribute decisionmaking of WSH multi-energy complemantary system under multple uncertainties.The main research content and innovation results of this paper are as follows:(1)Accurately quantifying the uncertainty in WSH hydrid system is an important foundation for ensuring the safe and stable operation of the hybdrid system.A time series variable probabilsitic forecasting framework is proposed to quantify uncertainty.The research work digs out historical features,periodic features and physical features from time series data;and uses correlation coefficients to make primary selection of the generated features,reducing the search space for feature combination optimization.Furthermore,a probabilistic forecasting framework for time series variables based on shared weighted long and short-term memory networks and Gaussian process regression is proposed.At the same time,the feature combination optimization and hyperparameter optimization are performed through the 0-1 planning idea and the Bayesian optimization algorithm,respectively,to ensure that the results are under the optimal feature input and optimal hyperparameters.Then,the forecast model was comprehensively evaluated from four aspects: deterministic forecasting,probabilistic forecasting,reliability and training time.Finally,a comprehensive comparison was made on multiple datasets with multiple state-of-the-art time series variable forecasting models.The experimental results verify the performance of the model proposed in this study in terms of forecast accuracy,probability forecast reliability and training time,and it has the ability to accurately quantify the multiple uncertain components in the WSH multi-energy hybrid system.(2)Research on deterministic optimal operation of WSH multi-energy hybrid systems is conducive to the analysis of complementary operation modes,and is also the basis for probabilistic optimal operation.The research work first analyzed the complementary characteristics of wind power,photovoltaic and hydropower on different time scales within the year and day.Then,considering the power generation and output stability objectives,a deterministic single-objective and multi-objective optimal operation model was constructed.In order to efficiently solve the deterministic optimal operation model,an improved mothflame single-objective optimization algorithm and improved multi-objective moth-flame optimization algorithm based on R-domination are proposed.In order to verify the convergence and distribution of the proposed algorithm,it is compared with the current state-of-the-art evolutionary algorithm on the benchmark functions.Finally,selecting multiple typical days,the proposed algorithms are used to solve the deterministic optimal operation model in the case study.Comparative analysis of the operation process and output complementary characteristics of different operation models on different typical days are performed.(3)Research on probabilistic optimal operation of WSH multi-energy hybrid system based on probabilistic forecasting and deterministic optimal operation.The research work first clarified the uncertain components in the WSH multi-energy hybrid system.Then,the general formula for the deterministic optimal operation model of the hybrid system is summarized,and the probabilistic optimal operation model of the hybrid system is constructed on this basis.Next,a framework for solving probabilistic optimal operation model based on "prediction-simulation-optimization-estimation-extraction" is proposed.Finally,taking the WSH complementary base in the Yalong River Basin as the research object,24-step ahead probabilistic forecasting of uncertainty components is performed.1000 sets of future scenarios are simulated,the deterministic operation models for these future scenarios are optimized,the probability density function of future model state variables is estimated.The joint operation rules of the WSH hybrid system in the Yalong River Basin are extracted and the risk of the extraction rules is evaluated.(4)Aiming at the problems of incomplete decision-making standards and low singletime decision-making error tolerance in the WSH hybrid system,a multi-stage integrated decision-making model is proposed in this research work.First,the whole decision is divided into multiple stages to complete the championship.Then,multiple decision-making methods are considered in the integrated decision model at the same time,and the comprehensive better solution is used as the final winning operation plan.Furthermore,considering that the single-objective operation rule extraction method is applied to the multi-objective optimization operaiton rule extraction,there is an ideal solution decisionmaking problem.Therefore,this research proposes a framework for multi-objective operation rule extraction under multiple uncertainties,which is driven by decision-making results.Finally,a case study was carried out with the Yalong River Basin complementary base as the research object,and a comprehensive operation plan that took into account benefits,stability and safety was made from the multi-objective operation plan. |
PDF Full Text Request