Computational Time Analysis for a Hybrid Energy Storage Power Forecasting Model

5.5 KB of data supporting a paper proposing a hybrid forecasting method for day-ahead power generation at energy storage stations. The dataset, authored by Lingzhi Xi and last updated in April 2026, likely contains analysis results from experiments using operational data from a 10 MW/20 MWh electrochemical energy storage power station. The proposed model achieved a Mean Squared Error of 6.41 MW² and a Coefficient of Determination (R²) of 0.898.

Use Cases

• Benchmarking computational efficiency of forecasting model stages based on the described analysis.
• Validating hybrid deep learning architectures (CNN-LSTM-MLP) for time-series prediction based on the described methodology.
• Comparing hyperparameter optimization techniques like Adaptive Genetic Algorithms based on the paper's framework.
• Analyzing the impact of feature extraction methods (EEMD, PFBD) on forecast accuracy based on the described process.

Strengths

• Experimental results are validated with specific performance metrics (MSE: 6.41 MW², MAE: 7.68 MW, R²: 0.898).
• Methodology is described in detail, integrating a physical loss model with data-driven deep learning techniques.
• Data is openly shared under a CC-BY-4.0 license, permitting reuse with attribution.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The 5.5 KB size indicates a very limited scope, likely containing summary results rather than raw time-series data.

Provenance

Source

Lingzhi Xi, via figshare.

Collection Method

Likely derived from experiments using actual operational data from a 10 MW/20 MWh electrochemical energy storage power station.

Time Range

null

Freshness

Last updated 2026-04-24 17:32:08; freshness should be verified.

Geography

null