ESS Charge Scheduling Using Corrected Power Demand Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current power demand prediction methods for energy storage systems are inaccurate due to their reliance on past data alone, failing to consider abnormal states and recent usage patterns, leading to inefficiencies in charge/discharge scheduling and increased energy costs.
Innovation Solution
A method that collects and analyzes weather and power usage data, generates long-term prediction data using weighted averages, and corrects it based on immediate power usage patterns to improve prediction accuracy, allowing for optimized charge/discharge control of energy storage systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If power demand prediction uses only past data, then the prediction method is simple, but the prediction accuracy is poor
Solution Approach 1:
The system performs preliminary classification of historical data into normal and abnormal states before prediction, and pre-calculates statistical parameters (mean, standard deviation) for different states. This preliminary preparation enables more accurate prediction by considering both normal and abnormal patterns without excessive real-time computational complexity.
Solution Approach 2:
The prediction method segments historical data into different states (normal and abnormal) based on power demand patterns. By dividing the data into distinct categories and applying different prediction approaches for each state, the system achieves higher accuracy while maintaining manageable complexity through targeted analysis of specific data segments.
2Reliability
If charge/discharge scheduling is based on peak cut rate margins, then the scheduling is simple, but the system cannot meet unexpected peak usage
Solution Approach 1:
The system continuously monitors actual power usage and compares it with predicted demand, then adjusts charge/discharge scheduling in real-time. This feedback mechanism enables the system to respond to unexpected peak usage by dynamically modifying the schedule, ensuring reliability while keeping the control logic relatively simple through iterative adjustment rather than complex pre-planning.
Solution Approach 2:
The charge/discharge scheduling transitions from static peak-cut-based margins to dynamic adjustment based on real-time power usage patterns and predictions. This dynamic approach allows the system to adapt to changing conditions and unexpected peaks, improving reliability while the underlying algorithm remains computationally efficient.
3Loss of energy
If energy storage system operates with low daily battery usage, then energy costs increase, but the system lacks complexity for advanced prediction
Solution Approach 1:
The system changes key parameters including the introduction of abnormal state detection thresholds, statistical parameters (mean, standard deviation), and state-specific prediction models. These parameter changes enable more effective battery utilization by accurately identifying when abnormal consumption occurs, thereby reducing energy waste and costs through targeted charge/discharge operations.
4Reliability
If power demand prediction does not consider abnormal states, then the prediction process is simple, but the prediction reliability deteriorates
Solution Approach 1:
The system performs preliminary classification of historical data into normal and abnormal states before prediction, and pre-calculates statistical parameters (mean, standard deviation) for different states. This preliminary preparation enables more accurate prediction by considering both normal and abnormal patterns without excessive real-time computational complexity.
Solution Approach 2:
The system introduces statistical parameters (mean, standard deviation) as intermediaries to objectively identify and classify abnormal states. These intermediary metrics enable reliable detection of abnormal power consumption patterns without requiring complex analysis, thereby improving prediction reliability through quantitative thresholds.
Data Source
AI summary
A method for predicting consumer power demand uses power consumption data measured over a long term and a power usage pattern immediately before a target time and for controlling ESS charge/discharge of an ESS based on the predicted power demand. A power demand prediction apparatus using the method includes respective components for collecting weather data and data on power used by the consumer; selecting data points according to a preset condition from among data collected by the data collector based on a specific time span; generating long term prediction data for the power demand in the specific time span based on the selected data points; analyzing a power usage pattern immediately before the specific time span and comparing the power usage pattern with the long-term prediction data, to determine whether prediction data correction is required; and correcting the prediction data based on the power usage pattern when correction is required.