Scenario Generation for Cooling, Heating, and Power Loads Using Generative Moment Matching Networks

TL;DR

This paper introduces a novel deep generative network based on GMMN for modeling and generating realistic cooling, heating, and power load scenarios, capturing complex patterns without assuming explicit probability distributions.

Contribution

The paper proposes a GMMN-based approach with auto-encoder for scenario generation, offering improved universality and accuracy over traditional explicit density models.

Findings

GMMN accurately models multi-class load distributions.

Generated scenarios replicate real load shape and temporal characteristics.

Energy consumption of generated data closely matches real data.

Abstract

Scenario generations of cooling, heating, and power loads are of great significance for the economic operation and stability analysis of integrated energy systems. In this paper, a novel deep generative network is proposed to model cooling, heating, and power load curves based on a generative moment matching networks (GMMN) where an auto-encoder transforms high-dimensional load curves into low-dimensional latent variables and the maximum mean discrepancy represents the similarity metrics between the generated samples and the real samples. After training the model, the new scenarios are generated by feeding Gaussian noises to the scenario generator of the GMMN. Unlike the explicit density models, the proposed GMMN does not need to artificially assume the probability distribution of the load curves, which leads to stronger universality. The simulation results show that the GMMN not only fits the probability distribution of multi-class load curves well, but also accurately captures the shape (e.g., large peaks, fast ramps, and fluctuation), frequency-domain characteristics, and temporal-spatial correlations of cooling, heating, and power loads. Furthermore, the energy consumption of generated samples closely resembles that of real samples.