Abstract:We propose a forecasting technique based on multi-feature data fusion to enhance the accuracy of an electric vehicle (EV) charging station load forecasting deep-learning model. The proposed method uses multi-feature inputs based on observations of historical weather (wind speed, temperature, and humidity) data as multiple inputs to a Long Short-Term Memory (LSTM) model to achieve a robust prediction of charging loads. Weather conditions are significant influencers of the behavior of EV drivers and their driving patterns. These behavioral and driving patterns affect the charging patterns of the drivers. Rather than one prediction (step, model, or variables) made by conventional LSTM models, three charging load (energy demand) predictions of EVs were made depending on different multi-feature inputs. Data fusion was used to combine and optimize the different charging load prediction results. The performance of the final implemented model was evaluated by the mean absolute prediction error of the forecast. The implemented model had a prediction error of 3.29%. This prediction error was lower than initial prediction results by the LSTM model. The numerical results indicate an improvement in the performance of the EV load forecast, indicating that the proposed model could be used to optimize and improve EV load forecasts for electric vehicle charging stations to meet the energy requirements of EVs.

What problem does this paper attempt to address?

This paper attempts to solve the accuracy problem of electric vehicle (EV) charging station load forecasting. Specifically, the author proposes a method based on multi - feature data fusion to improve the accuracy of deep - learning models in predicting the load of electric vehicle charging stations. The following are the main problems and solutions in this paper: ### 1. Research Background and Problems With the popularization of electric vehicles, the power demand of charging stations has become more and more uncertain, which poses a challenge to the stability and reliability of the power system. The traditional Long - Short - Term Memory network (LSTM) model has certain limitations in predicting the charging load of electric vehicles, especially when dealing with complex and changeable environmental factors (such as weather conditions). ### 2. Proposed Solutions To improve this situation, the author proposes a new prediction method that combines multi - feature data fusion and deep learning. This method uses historical weather data (including wind speed, temperature and humidity) as multiple input features and is trained by the LSTM model to achieve more robust charging load prediction. The specific steps are as follows: - **Multi - feature Input**: Multiple environmental features (such as temperature, humidity and wind speed) are introduced. These features will affect the behavior patterns and driving habits of electric vehicle drivers, thereby affecting the charging mode. - **Data Fusion**: Through data fusion technology, multiple prediction results under different feature inputs are optimally combined, and finally a more accurate prediction result is generated. ### 3. Method Innovation Points Compared with the traditional LSTM model, the innovation points of this research are: - **Multi - feature Input**: It does not rely solely on a single variable or time step, but comprehensively considers multiple environmental factors. - **Data Fusion Optimization**: The Dempster - Shafer evidence theory (D - S theory) is used for data fusion, which improves the reliability and accuracy of the prediction results. ### 4. Experimental Results The experimental results show that the prediction error of this model is 3.29%, which is lower than the initial prediction error of the traditional LSTM model. This indicates that by introducing the multi - feature data fusion technology, the accuracy of electric vehicle charging station load prediction can be significantly improved, which helps to better meet the energy needs of electric vehicles and optimize the operation of the power system. ### Summary This paper aims to improve the prediction accuracy of deep - learning models for electric vehicle charging station loads by introducing multi - feature data fusion technology. Research shows that this method can effectively deal with the uncertainty problems in electric vehicle charging load prediction and provides strong support for the future development of smart grids.

Multi Feature Data Fusion-Based Load Forecasting of Electric Vehicle Charging Stations Using a Deep Learning Model

Prediction of Electric Vehicles Charging Load Using Long Short-Term Memory Model

Feature-Enhanced Deep Learning Method for Electric Vehicle Charging Demand Probabilistic Forecasting of Charging Station

Short-term Load Forecasting Based on Multi-model Fusion of CNN-LSTM-LGBM

A Deep Learning Approach Based on Novel Multi-Feature Fusion for Power Load Prediction

Multi-Encoder Spatio-Temporal Feature Fusion Network for Electric Vehicle Charging Load Prediction

Deep Information Fusion for Electric Vehicle Charging Station Occupancy Forecasting

Probabilistic Forecasting of Electric Vehicle Charging Load Using Composite Quantile Regression LSTM

Multiscale Spatio-Temporal Enhanced Short-term Load Forecasting of Electric Vehicle Charging Stations

A Deep Learning Based Approach for Predicting the Demand of Electric Vehicle Charge

Multi-step electric vehicles charging loads forecasting: An autoformer variant with feature extraction, frequency enhancement, and error correction blocks

Prediction of EV Charging Load Using Two-Stage Time Series Decomposition and DeepBiLSTM Model

Charging Load Forecasting of Electric Vehicle Based on Monte Carlo and Deep Learning

Electric Vehicle Charging Load Prediction Based on Weight Fusion Spatial–Temporal Graph Convolutional Network

Multi-Model Fusion Short-Term Load Forecasting Based on Random Forest Feature Selection and Hybrid Neural Network

Deep-Learning-Based Probabilistic Forecasting of Electric Vehicle Charging Load with a Novel Queuing Model

Multi-Convolution Feature Extraction and Recurrent Neural Network Dependent Model for Short-Term Load Forecasting

Load forecasting of electric vehicle charging station based on grey theory and neural network

Load Forecasting of Battery Electric Vehicle Charging Station based on GA-Prophet-LSTM

Location based Probabilistic Load Forecasting of EV Charging Sites: Deep Transfer Learning with Multi-Quantile Temporal Convolutional Network

Short-Term Forecasting of EV Charging Load Using Prophet-BiLSTM