在常规和可再生能源下使用人工智能技术对电力大量能源消耗进行有效的控制和管理

祝贺Tanveer Ahmad同学的论文《Effective bulk energy consumption control and management for power utilities using artificial intelligence techniques under conventional and renewable energy resources》被《International Journal of Electrical Power and Energy Systems》期刊录用。

论文信息

论文题目：Effective bulk energy consumption control and management for power utilities using artificial intelligence techniques under conventional and renewable energy resources

作者: Tanveer Ahmad, Huanxin Chen

第一单位：Department of Refrigeration and Cryogenics, Huazhong University of Science and Technology, Wuhan, China

期刊名：International Journal of Electrical Power and Energy Systems

影响因子：3.610

中科院分区：二区

论文介绍

论文摘要

Increasing sustainability demands initiateestimating various design and control opportunities for classifyingenergy-efficient plan ever more significant. These conditions demand simulationalgorithms which are not only fast but also accurate. Artificial intelligence(AI) enables efficient mimicry of bulk energy consumption control whileproducing results much faster than data-mining and machine learning models.This study proposes two AI based approaches for utilities bulk energyconsumption prediction, control and management. Two different zones actualenvironmental and energy consumption data are obtained for input featuresselection and modeling analysis. Each zone is categorized into five featuresparameter selection (PS) states. Each PS state is further divided into fourdifferent hidden neurons (HD) and hidden layers of the model’s network. Theforecasting duration is based on 1-month and 1-year ahead intervals formedium-term (MT) and long-term (LT) respectively. Further the current proposedmodel’s performance is compared with three existing models. One of thepromising finding in this research is that substantial improvement inprediction accuracy applying features extracted by PS-3 and PS-5. Results showthat AI models are powerful in solving complex and nonlinear patterns of rawdata. This study renders optimal decisions can be projected while utilitiesenergy supply strategy & control, capacity expansion, capital investmentresearch market management, revenue analysis and future load requirement forecasting.

论文简介

Fig. 1. ACF and PACF of net energy usage ofZone-1 and Zone-2.

Fig.2. Proposed forecasting methodology

Fig. 4. Aggregated actual and forecastedenergy requirement for long-term load management.

Fig. 5. The gradient of PRGBNNs and GDALBNNs models.

Fig. 6. Forecasting results compared withthe existing models.

作者介绍

Tanveer Ahmad (韩珂)

博士研究生

华中科技大学

能源与动力工程学院

制冷与低温工程