DOI

Title

Multi-Horizon Short-Term Load Forecasting Using Hybrid of LSTM and Modified Split Convolution

Overview

This repository contains the code and datasets for our research on short-term load forecasting using the LSTM-SC hybrid model. We have utilized three datasets for our study:

1. AEP Dataset
2. ISONE Dataset
3. NTDC

Abstract

Precise Short-Term Load Forecasting (STLF) plays a crucial role in the smooth operation of power systems, future capacity planning, unit commitment, and demand response. However, due to its non-stationary and its dependency on multiple cyclic and non-cyclic calendric features and non-linear highly correlated metrological features, an accurate load forecasting with already existing techniques is challenging. To overcome this challenge, a novel hybrid technique based on long short-term memory (LSTM) and a modified Split-Convolution (SC) neural network (LSTM-SC) is proposed for single-step and multi-step STLF. The concatenating order of LSTM and SC in the proposed hybrid network provides an excellent capability of extraction of sequence-dependent features and other hierarchical spatial features. The model is evaluated by the Pakistan National Grid load dataset recorded by the National Transmission and Dispatch Company (NTDC). The load data is pre-processed and multiple other correlated features are incorporated into the data for performance enhancement. For generalization capability, the performance of LSTM-SC is evaluated on publicly available datasets of American Electric Power (AEP) and Independent System Operator New England (ISO-NE). The effect of temperature, a highly correlated input feature, on load forecasting is investigated either by removing the temperature or adding a Gaussian random noise into it. The performance evaluation in terms of RMSE, MAE, and MAPE of the proposed model on the NTDC dataset are 500.98, 372.62, and 3.72\% for multi-step while 322.90, 244.22, and 2.38\% for single-step load forecasting. The result shows that the proposed method has less forecasting error, strong generalization capability, and satisfactory performance on multi-horizon.

Network Architecture

Hybrid of LSTM and Modified Split Convolution (LSTM-SC)

Datasets

1) AEP Dataset

The AEP dataset is publicly available and can be accessed directly for training and testing purposes.

2) ISONE Dataset

The ISONE dataset, another publicly available dataset, is included for comprehensive analysis and comparison.

Data Processing

Processed-Data Folder

The datasets available in the Processed-Data folder have been meticulously processed based on insights from 50 research papers. These processed datasets are ready for direct use in training the LSTM-SC hybrid model.

Unprocessed-Data Folder

For those interested in exploring raw data and applying custom preprocessing techniques, the Unprocessed-Data folder contains all the raw datasets. Feel free to apply any preprocessing methods suited to your research needs.

Getting Started

1. Clone the Repository: git clone https://github.com/[SyedHasnat]/[Papers].git

2. Navigate to Processed-Data or Unprocessed-Data Folder:

   • For direct usage, explore the Processed-Data folder.
   • For customization and preprocessing, delve into the Unprocessed-Data folder.

3. Run Experiments:

   • Utilize the processed datasets for training and evaluating the LSTM-SC hybrid model.
   • Apply your preprocessing techniques if using the unprocessed datasets.

4. Algorithms Soon Algorithm 1 and 2 will be availible on PyPI, I will update readme file accordingly.

   • Algorithm 1: Pseudo-code of load data generator for single step.
   • Algorithm 2: Pseudo-code of load data generator for multi step.

Datasets Used

• Independent System Operator New England (ISO-NE)
• American Electric Power (AEP)
• National Transmission and Dispatch Company (NTDC) ## Forecasting single-step and multi-step STLF ## Code Files All the code files for the paper "Multi-Horizon Short-Term Load Forecasting Using Hybrid of LSTM and Modified Split Convolution" are availible in the folder "PeerJ" as well in Code-Files, both the folders have the same code. ## Note to Researchers We encourage fellow researchers to explore and build upon our work. If you find this work interested we can callobrate.

Feel free to contribute, report issues, or suggest improvements. Let's collectively advance the field of short-term load forecasting using innovative techniques!

Contact

If you have any questions or need further assistance, you can reach me via:

• Email: eng.syed.its@gmail.com
• LinkedIn: Syed Muhammad Hasanat

Happy Forecasting!