Hybrid Energy Storage Scheduling With CNN Load Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing hybrid energy storage systems lack predictive capabilities, leading to poor controllability and inefficiency, with inaccurate adjustments due to rough training data and insufficient temperature management, resulting in shortened lifespan.
Innovation Solution
A hybrid energy storage system utilizing a convolutional neural network to predict scheduling instructions based on historical data, combining power-type and energy-type storage media, with temperature-adaptive battery configurations and improved loss functions for enhanced accuracy and flexibility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If hybrid energy storage systems are configured to smooth power voltage and current fluctuations, then power quality is improved, but system complexity increases
Solution Approach 1:
The system divides energy storage functions into two independent modules: power-type energy storage (supercapacitors) for high-power fluctuations and energy-type storage (batteries) for sustained energy supply. This segmentation allows each module to specialize in specific functions, improving power quality while managing complexity through functional decomposition
Solution Approach 2:
The control system integrates multiple functions including prediction, scheduling, and coordination of different energy storage types into a single unified platform. The system can adaptively switch between different energy storage combinations based on grid conditions, achieving multi-functionality without proportionally increasing complexity
2Measurement precision
If artificial intelligence algorithm training is used for scheduling, then scheduling accuracy is improved, but data quality requirements increase system complexity
Solution Approach 1:
The system performs preliminary data processing and feature extraction before feeding data into the AI algorithm. Historical scheduling data is pre-processed to extract relevant features, and the system maintains a database of pre-processed data that can be quickly queried, reducing the computational burden during real-time scheduling while maintaining high accuracy
Solution Approach 2:
The system introduces an intermediate layer between raw data and the AI algorithm, including data cleaning, normalization, and feature selection modules. This intermediary layer transforms rough historical data into high-quality training data, enabling accurate scheduling without requiring complex data processing in the core algorithm
3Device complexity
If ambient temperature effects are not managed, then system simplicity is maintained, but energy storage lifespan decreases
Solution Approach 1:
The system incorporates temperature sensors that continuously monitor the thermal environment of energy storage devices. The collected temperature data is fed back to the control system, which adjusts scheduling strategies to avoid extreme temperature conditions and reduce thermal stress on batteries, thereby extending lifespan while adding only minimal monitoring complexity
Solution Approach 2:
The system performs preliminary thermal assessment before executing scheduling commands. By predicting temperature changes based on historical data and current conditions, the system can pre-adjust operational parameters or scheduling timing to avoid temperature extremes that would accelerate degradation, preventing damage before it occurs
Data Source
Figure 1~2
AI summary
The invention discloses a hybrid energy storage system scheduling system and method, which includes steps S1: the CPU detects whether a scheduling instruction is received in real time, and detects and obtains the maximum power of the power type energy storage medium; S2: when the power value required by the scheduling instruction is greater than When the maximum power of the power-type energy storage medium is reached, the capacity-type energy storage medium is turned on for supplementation, otherwise the power-type energy storage medium is used to respond to the scheduling instruction; S3: While executing step S2, the convolutional neural network output predicts the amplitude of the next scheduling instruction. and duration; S4: Coordinate the two energy storage media to respond to the scheduling instructions according to the predicted amplitude and duration. This hybrid energy storage system scheduling method predicts the amplitude and duration of the next scheduled power command based on the statistical characteristics of historical data, which greatly enhances the controllability of power scheduling. It can also arrange and set the application public power in advance, improving power scheduling. efficiency.