EV Charging Location Forecasting for Utility Load Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Determining optimal locations for electric vehicle (EV) charging stations is challenging due to various factors such as the number of EVs in an area, available installation locations, and operating costs, leading to sub-optimal usage and revenue issues for operators, which can limit EV adoption.
Innovation Solution
A computer-implemented method using machine learning models, such as neural networks or support vector models, to dynamically generate charger maps by training on historical data, considering consumer and location parameters, and updating in real-time to optimize EV charger locations based on prioritized features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional methods are used to determine EV charging station locations, then the process is simpler, but the accuracy and optimization of charger placement deteriorates
Solution Approach 1:
The patent introduces a machine learning model as an intermediary between raw data (consumer parameters, location parameters, historical data) and charger location recommendations. This ML intermediary processes complex relationships and patterns that traditional methods cannot capture, thereby improving accuracy without requiring the end user to directly handle the complexity of multiple data factors.
Solution Approach 2:
The patent replaces traditional mechanical or manual location determination systems with a machine learning-based computational system. Instead of using simple heuristics or manual assessment of location factors, the system uses trained ML models (neural networks, support vector models) to automatically analyze historical data and consumer behavior patterns, substituting complex computational processing for simpler traditional methods.
2Adaptability or versatility
If static charger maps are used, then the system is easier to maintain, but the adaptability to changing conditions deteriorates
Solution Approach 1:
The patent implements dynamic charger maps that automatically update in response to changing conditions. The system continuously receives new consumer parameters, location parameters, and historical data, retrains the machine learning model with this updated information, and generates revised charger location recommendations. This transforms the static map into a dynamic system that adapts to evolving consumer behavior, new charging infrastructure, and changing geographic conditions.
Solution Approach 2:
The system incorporates feedback loops where the performance and usage data from existing chargers are fed back into the machine learning model. The model analyzes this feedback information along with new data inputs, and uses the learned patterns to optimize charger placement decisions. This feedback mechanism enables continuous improvement and adaptation of the charger map based on real-world performance data.
3Manufacturing precision
If comprehensive data analysis is performed, then the optimization quality improves, but the processing time increases
Solution Approach 1:
The patent performs preliminary actions by pre-training the machine learning model on extensive historical data and consumer parameters before actual charger location recommendations are needed. This pre-processing and model training phase captures complex patterns and relationships in advance, so that when actual charger placement decisions are required, the system can quickly query the trained model rather than performing comprehensive analysis from scratch each time.
Solution Approach 2:
The system changes parameters by using the machine learning model to transform multiple input parameters (consumer demographics, location characteristics, historical usage data, competitor charger locations) into optimized output parameters for charger placement. The ML model learns optimal parameter relationships and transformations during training, enabling efficient computation of high-quality placement recommendations without manually processing each parameter combination in real-time.
Data Source
AI summary
A computer-implemented method of projecting a utility load demand including providing vehicle parameters for a plurality of areas and location parameters associated with the plurality of areas to a machine learning (ML) model, iteratively training the ML model to identify relationships between the vehicle parameters, the location parameters, and historical utility data associated with the plurality of areas, receiving a target area and a future target date, providing the target area and the future target date to the trained ML model, obtaining and providing target vehicle parameters and target location parameters for the target area to the trained ML model, determining, via the trained ML model, an EV charging forecast for the target area at the future target date and projecting, via the trained ML model, a utility load demand within the target area at the future target date based on the EV charging forecast.