Research On Regional Flexible Load Low Carbon Dispatching Based On Cloud Model

Low-carbon technologies are still in the research,development,and demonstration stage,with relatively high implementation costs.At the same time,renewable energy generation technologies such as microgrids and distributed power generation are still in the development stage.With the widespread deployment of renewable energy generation equipment and the integration of demand-side resources into new microgrid systems,the operation of the power system is gradually moving away from the traditional top-down hierarchical structure,resulting in increasing uncertainty and making optimization scheduling of the power system difficult to effectively carry out.Based on these issues,this thesis proposes a new scheduling scheme,which is mainly described as follows:(1)A model is constructed for the units within the microgrid,and a wind and solar power output model is established.For wind and solar units with high uncertainty,Monte Carlo methods are used for scenario generation and reduction to generate typical output curves and improve the accuracy of wind and solar power output.Fuel cells and diesel generators are modeled as backup units for power generation,and energy storage devices are modeled and analyzed for their classification and application scenarios.(2)To improve the traditional threshold-based electricity pricing which has the disadvantages of inaccurate selection and inability to follow real-time dispatch instructions,a dynamic peak-valley-flat electricity pricing strategy based on fuzzy c-means clustering is proposed for the supply side to partition the daily load.For the demand side,a large amount of flexible loads,such as shiftable loads,transferable loads,and reducible loads are modeled and analyzed for their transfer characteristics.By specifying transfer time periods,transfer start time ranges,and user incentive compensation methods,demand-side resources are integrated to improve the incentive response of flexible loads.Cloud models are used to analyze the uncertainty of demand response,and the relationship between fuzziness and randomness is revealed using numeric features,expectation descriptions of qualitative concept points,entropy descriptions of uncertainty,and hyper-entropy descriptions of random occurrences of concept samples.Finally,the uncertainty of demand response is analyzed for cases with no flexible loads participating and one,two,and three types of flexible loads participating,and it is experimentally proven that demand response is more stable and conducive to regulation with multiple types of flexible loads participating.(3)Under the supply-demand strategy with multiple types of flexible loads and dynamic time-of-use electricity pricing incentives and low-carbon goals,a multi-source unit combination output model is established based on cloud model analysis to minimize the power generation cost and carbon emissions while ensuring the load is met,expressing the reduction of carbon emissions in terms of carbon cost.The cost is further reduced by considering carbon trading and system purchasing and selling characteristics.Finally,three cases are simulated: no supplydemand strategy and low-carbon goal,supply-demand strategy but no low-carbon goal,and supply-demand strategy with low-carbon goal,which resulted in a significant reduction in system cost.This model enables various flexible loads to work together,the time-of-use electricity pricing is more reasonable,and the multi-source unit output is more stable and orderly,resulting in a more stable and efficient power system.