Forecasting Power Demand In China Based On Multimodal Information Fusion Method,and Coal Power Exit Scale Calculation

As the industry with the highest carbon emissions in China,it is important to steadily promote the orderly exit of coal power for achieving the dual carbon goal.At the same time,due to the volatility of renewable energy supply,global energy market conflicts,and other reasons,it is necessary to ensure the safe and stable supply of electricity in China.Therefore,under the premise of dual carbon goals and power safety goals,it is crucial to scientifically plan the coal power exit work.The prerequisite for scientifically planning the withdrawal of coal based electricity is to carry out forecasting work on China’s electricity demand and renewable energy supply.However,at present,there are very few studies that provide data support for the withdrawal of coal based electricity by making a comprehensive consideration of electricity demand and renewable energy forecasts.In view of this,based on the idea of multimodal information fusion modeling,this paper proposes a combined prediction modeling method that integrates time series data and text data.At the same time,this article has carried out the construction and demonstration of China’s electricity demand forecasting model,as well as the construction and demonstration of renewable energy supply forecasting model.On this basis,the scenario analysis method is used to calculate the scale of coal power withdrawal in China.Firstly,aiming at the information complementarity between time series data and text data,the research paradigm of multimodal information fusion is used to construct a CNN-LSTM(Revolution Neural Network,Long Short term Memory)prediction modeling method,which realizes data information fusion prediction of structured and unstructured data.Secondly,collect time series data of electricity demand and related multi type text data,renewable energy supply time series data and related multi type text data,and conduct prediction model construction and empirical prediction on China’s electricity demand and renewable energy supply,respectively.The fluctuation trend of China’s future electricity demand and renewable energy supply is revealed.Finally,on the basis of the above prediction,scenario analysis method is used to predict the power demand and renewable energy supply under multiple scenarios,so as to calculate the scale of coal power withdrawal in China under multiple scenarios.The research results show that:(1)This article proposes a modeling method based on multimodal information fusion.Empirical results show that compared to single prediction models(ARIMA,CNN,LSTM)and combined prediction models(EEMD ARIMA,EEMD LSSVM,etc.),the modeling method proposed in this article achieves the best horizontal and directional accuracy in predicting China’s electricity demand and renewable energy supply.This also indicates that the method of organically integrating time series data and text data can effectively improve the predictive performance of the model.(2)Due to multiple factors such as China’s economic restructuring and energy system transformation,China’s electricity demand will undergo a phased adjustment from January 2023 to December 2024,with the growth rate of electricity demand slowing or even showing a downward trend.This finding provides an important decision-making reference for China’s low-carbon transformation of the power system.Although the forecast results show a downward trend in China’s electricity demand in 2023 and 2024,China’s renewable energy power generation will still increase in the future.Renewable energy plays an increasingly important role in ensuring energy supply.In the future,renewable energy power will become the mainstay of China’s power system.(3)Under different development scenarios,the scale of coal power withdrawal varies greatly,indicating the difficulty of coal power withdrawal.Under the constrained scenario,China’s electricity demand development presents a relatively high trend.Regardless of the supply and development mode of renewable energy,the scale of coal power withdrawal is still significantly higher than other scenarios.In the development scenario,the existing installed capacity may be difficult to meet the power demand,and even the supply of renewable energy cannot make up for this situation.In order to ensure the smooth and orderly progress of coal power withdrawal in China,we need to develop electricity demand based on actual conditions,while increasing investment in renewable energy to supplement the coal power demand that needs to be withdrawn.In general,the research results of this thesis have important significance and value in both theoretical and practical aspects.In theory,this thesis proposes a predictive modeling method based on multimodal information fusion,which organically integrates time series data and text data,making full use of the information complementarity of multi-source heterogeneous data,thereby enriching the theory of predictive modeling methods in the energy industry.In practice,we have constructed a forecasting model for China’s electricity demand and renewable energy supply,and conducted empirical predictions to reveal the future trends of both.In addition,the scenario analysis method is used to calculate the scale of coal power withdrawal under multiple scenarios.This provides corresponding theoretical and data support for power system planning and coal power exit work.This thesis has a total of 38 pictures,18 sheets,and 95 references.