Study On Multi-Energy Load Collaborative Forecasting Of Integrated Energy System Considering Multi-Energy Coupling

The construction of a "comprehensive energy system" is an important part of China’s current energy development,and also one of the important ways for the country to promote supply side structural reform in the energy field and achieve the "dual carbon"goal.Internationally,with the continuous improvement of China’s economic and social development level,the energy consumption structure is showing a trend of rapid optimization and upgrading.Especially in recent years,the proposal of the "dual carbon"goal and the rapid construction of new power systems have greatly promoted the rapid development of clean energy such as new energy,energy storage,and renewable energy.At present,China’s power system is transitioning from a traditional power system to a modern power system,with the most important change being the realization of a grid as the main body and coordinated interaction of multiple power sources,enabling China to develop from highly dependent on fossil fuels to highly dependent on new energy.In 2016,with the release of relevant government documents,research and development on comprehensive energy systems focused on the construction of CCHP systems and the utilization of user side infrastructure such as industrial residual pressure and waste heat;The government actively promotes the construction of China’s comprehensive energy green development system,improves the entire process system from production to sales of energy,vigorously promotes various clean and renewable energy power generation,improves the construction of the new energy power market,further liberalizes the demand side sales market and distributed power market,improves the efficiency of power resource allocation,and encourages social capital to invest in the construction and operation of distributed power sources and microgrids,Establish a sound electricity market system and develop a sound multi energy complementary system.The comprehensive energy system is conducive to improving energy utilization efficiency,reducing energy consumption costs,and playing an indispensable role in the process of achieving green coordinated development in China.Accurate load forecasting is crucial for the stable and flexible operation of the IES system,and it can provide accurate forecasting and decision-making basis for the power grid.With the upgrading of energy coupling equipment and the promotion of energy diversification policies,the correlation between various types of energy gradually strengthens,and more complex energy coupling relationships are gradually formed.However,most of the existing scholars’ research on load forecasting methods is for load forecasting alone,without considering the coupling and complementarity between energy sources.With the development of energy technology,the interconnectivity between different energy sources is becoming stronger,and the randomness and variability of time series in integrated energy systems are also increasing,making it difficult to accurately predict multi energy time series.To address this issue,the research content of this article includes:Firstly,the basic theories required for establishing a short-term forecasting model for multiple loads in a comprehensive energy system were analyzed,providing a theoretical basis for the construction of subsequent forecasting models.Firstly,the basic principles of multi energy IES and multi energy coupling were analyzed,and their basic composition,characteristics,and specific modes of multi energy integration were explored.Secondly,based on various Copula functions such as t-Copula,Gumbel Copula,Clayton Copula,Frank Copula,Normal Copula,etc.,the basic theories and methods of correlation analysis are studied.On this basis,the related principles and experimental basis of various intelligent optimization algorithms,such as Tabu search,simulated annealing,genetic algorithm,ant colony algorithm,are introduced in depth,laying the foundation for subsequent selection of appropriate optimization algorithms.Secondly,a study was conducted on the correlation analysis and coupling degree calculation methods of multiple loads in the comprehensive energy system.Firstly,this section provides a detailed introduction to the load characteristics of IES,starting from the aspects of load rate,peak valley,and correlation measurement.The factors that affect the load of IES are identified,and based on multiple considerations,Copula theory is selected as a tool and method for multivariate load correlation analysis.This method is introduced in detail from multiple perspectives,and a calculation model for load coupling relationship is established.Thirdly,a short-term forecasting model for multiple loads in a comprehensive energy system based on SSA-LSSVM was constructed.Firstly,by analyzing the principle and composition of LSSVM,conducting in-depth research on its advantages,disadvantages,and model drawbacks,and exploring corresponding solutions,a least squares support vector machine prediction model was constructed using the least squares method,and optimized to address the drawbacks of traditional LSSVM models;Secondly,the traditional LSSVM parameter selection method using the network cross validation method is improved,and the optimal regularization parameters and kernel function parameters are automatically found using the Salp Swarm Algorithm(SSA);Finally,based on the SSA-LSSVM based comprehensive energy system multiple load prediction model,the prediction process of the comprehensive energy system multiple load system is studied,and specific prediction steps are provided.Fourthly,based on the above research,a comprehensive energy system is selected as a case to be analyzed and validated in the model,in order to test the reliability of the multivariate short-term load forecasting model.Firstly,input basic parameters such as load data and meteorological data into the model to conduct correlation analysis on the multiple loads of the comprehensive energy system in the park;Secondly,the overall error is tested and the degree of deviation between the predicted value and the true value is compared.The average absolute percentage error and root mean square error are selected as evaluation indicators for comprehensive analysis;Finally,simulate different scenarios of the IES system,conduct simulation predictions,and compare and analyze the results to verify the predictive performance of the model.