Customized Load Profiles Synthesis for Electricity Customers Based on Conditional Diffusion Models

TL;DR

This work addresses data shortage for electricity customer analytics by introducing a conditional diffusion-based framework to synthesize customized load profiles. By framing synthesis as conditioned data generation, a server–customer architecture with cross-learning trains a global conditional diffusion model and a time-series–aware noise estimator to produce exclusive, high-fidelity load sequences conditioned on each customer's characteristics. The approach demonstrates superior performance over baselines in both data generation and data augmentation scenarios, and yields tangible improvements in downstream tasks such as load forecasting, while preserving data privacy. The results suggest a practical, scalable path for personalized analytics in power systems using synthetic data.

Abstract

Customers' load profiles are critical resources to support data analytics applications in modern power systems. However, there are usually insufficient historical load profiles for data analysis, due to the collection cost and data privacy issues. To address such data shortage problems, load profiles synthesis is an effective technique that provides synthetic training data for customers to build high-performance data-driven models. Nonetheless, it is still challenging to synthesize high-quality load profiles for each customer using generation models trained by the respective customer's data owing to the high heterogeneity of customer load. In this paper, we propose a novel customized load profiles synthesis method based on conditional diffusion models for heterogeneous customers. Specifically, we first convert the customized synthesis into a conditional data generation issue. We then extend traditional diffusion models to conditional diffusion models to realize conditional data generation, which can synthesize exclusive load profiles for each customer according to the customer's load characteristics and application demands. In addition, to implement conditional diffusion models, we design a noise estimation model with stacked residual layers, which improves the generation performance by using skip connections. The attention mechanism is also utilized to better extract the complex temporal dependency of load profiles. Finally, numerical case studies based on a public dataset are conducted to validate the effectiveness and superiority of the proposed method.

Figures (8)

• Figure 1: Customized load profiles synthesis framework.
• Figure 2: Illustration of conditional diffusion models for customized load profiles synthesis.
• Figure 3: The neural network architecture of noise estimation model.
• Figure 4: Load profiles synthesis results of two examples under data generation scenario in (a) summer and (b) winter.
• Figure 5: Two-dimensional visualization of synthesis performance comparison under data generation scenario. A data point represents one daily load profiles after dimensionality reduction. The curves on the axes depict the density curves for each type of load profiles in the two reduction dimensions.