4.7 Article

A knowledge-enhanced graph-based temporal-spatial network for natural gas consumption prediction

Related references

Note: Only part of the references are listed.
Article Energy & Fuels

A new energy consumption prediction method for chillers based on GraphSAGE by combining empirical knowledge and operating data

Zhiwen Chen et al.

Summary: This study proposes a new energy consumption prediction method based on graph sampling aggregation network, which constructs association graphs by combining operating data with empirical knowledge, achieving accurate prediction of chiller energy consumption.

APPLIED ENERGY (2022)

Article Thermodynamics

Dynamic simulation of natural gas pipeline network based on interpretable machine learning model

Dengji Zhou et al.

Summary: This paper proposes a novel dynamic simulation method for natural gas pipeline network using an interpretable shortcut Elman network (Shortcut-ENN) model. The method addresses the limitations of mechanism modeling and achieves better results compared to other methods.

ENERGY (2022)

Article Energy & Fuels

A safety management system for natural gas pipeline in subsidence area of coal mine

Yue Shi et al.

Summary: This paper proposes a safety management system for natural gas pipelines in subsidence areas of coal mine, which includes technical modules such as surface and pipeline displacement monitoring, risk assessment index system, and early warning model. By utilizing data collection, analysis, and real-time monitoring, the safety performance of gas pipelines in subsidence areas is effectively enhanced.

ENERGY SOURCES PART A-RECOVERY UTILIZATION AND ENVIRONMENTAL EFFECTS (2022)

Article Thermodynamics

A hybrid deep learning framework for predicting daily natural gas consumption

Jian Du et al.

Summary: This paper proposes a novel method that addresses the deficiency of conventional time series prediction for natural gas consumption by designing a hybrid deep learning framework to extract comprehensive information. The proposed model enables accurate and efficient prediction of natural gas consumption by capturing static and dynamic information and integrating long-term dependency patterns.

ENERGY (2022)

Article Computer Science, Artificial Intelligence

Identification of spreading influence nodes via multi-level structural attributes based on the graph convolutional network

Yang Ou et al.

Summary: This paper proposes a new graph convolutional network algorithm, M-RCNN, which identifies spreading influence nodes by considering structural information at micro, community, and macro levels. Experimental results demonstrate that M-RCNN achieves 9.25% higher accuracy compared to RCNN. The evaluation also shows that M-RCNN has a similar computational complexity to RCNN, indicating high efficiency.

EXPERT SYSTEMS WITH APPLICATIONS (2022)

Article Engineering, Industrial

A systematic method for the optimization of gas supply reliability in natural gas pipeline network based on Bayesian networks and deep reinforcement learning

Lin Fan et al.

Summary: This study proposes a method based on Bayesian networks to optimize the reliability of gas supply in natural gas pipeline networks. The method integrates probabilistic safety analysis with preventive maintenance and converts the system maintenance problem into a Markov decision process, solved by using deep reinforcement learning. The proposed method outperforms others in identifying optimal maintenance strategies by considering the randomness of unit failures and the uncertainty in gas demand profiles.

RELIABILITY ENGINEERING & SYSTEM SAFETY (2022)

Article Engineering, Environmental

Deeppipe: Theory-guided LSTM method for monitoring pressure after multi-product pipeline shutdown

Jianqin Zheng et al.

Summary: In this study, a deep learning algorithm Long Short-Term Memory (LSTM) was utilized to construct the pipeline shutdown pressure prediction model TG-LSTM, integrated with scientific principles and engineering theory for improved accuracy. The proposed model was validated on three real-world multi-product pipelines, achieving better accuracy compared to other prediction models with MAPE of 0.246%, 0.186%, and 0.143% respectively. Sensitivity analysis of different hyper-parameters further illustrated the robustness of TG-LSTM in pipeline shutdown pressure prediction.

PROCESS SAFETY AND ENVIRONMENTAL PROTECTION (2021)

Article Transportation Science & Technology

Multi-community passenger demand prediction at region level based on spatio-temporal graph convolutional network

Jinjun Tang et al.

Summary: A multi-community spatio-temporal graph convolutional network framework was proposed for region-level passenger demand prediction, exploring spatial and temporal correlations among regions. Real-world taxi order data from Shenzhen City and New York City were used for model validation, showing the superiority of the new model over traditional and deep learning methods. Discussions on various aspects highlighted the advantages and characteristics of the proposed model.

TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES (2021)

Article Energy & Fuels

Optimal operation of compressor units in gas networks to provide flexibility to power systems

Yongning Zhao et al.

Summary: Compressor stations are crucial in maintaining pressure levels in natural gas networks, with the ability to switch between different fuel-driven compressor units to achieve desired outlet pressure and flow throughput. The flexibility provided by compressors to the power system is influenced by the allowable range for variation of linepack.

APPLIED ENERGY (2021)

Article Energy & Fuels

Daily natural gas load forecasting based on the combination of long short term memory, local mean decomposition, and wavelet threshold denoising algorithm

Shanbi Peng et al.

Summary: With the opening of the third energy revolution, the booming development of the natural gas industry in China has led to the existing pipeline network facilities no longer meeting the gas demand of its residents. How to adjust pipeline network layout and design peak-regulating gas storage facilities to meet the gas demand of residents has become a challenge. Accurate load forecasting can help solve this problem. This study proposes a novel combined model, LMD-WTD-LSTM, which shows excellent performance in daily natural gas prediction, especially when the forecasting time length is 20 days.

JOURNAL OF NATURAL GAS SCIENCE AND ENGINEERING (2021)

Article Engineering, Environmental

Deeppipe: A semi-supervised learning for operating condition recognition of multi-product pipelines

Jianqin Zheng et al.

Summary: Intelligent operating monitoring of pipelines is crucial for detecting anomalies in time to ensure safety. A proposed semi-supervised learning approach for operating condition recognition shows high accuracy and performance, outperforming other machine learning models. The method proves effective for overcoming challenges in accurately identifying changing operating conditions of multi-product pipelines.

PROCESS SAFETY AND ENVIRONMENTAL PROTECTION (2021)

Article Engineering, Industrial

An integrated gas supply reliability evaluation method of the large-scale and complex natural gas pipeline network based on demand-side analysis

Weichao Yu et al.

Summary: This study proposes an integrated method based on demand-side analysis to assess the supply reliability of large-scale and complex natural gas pipeline networks. By analyzing market demand and user importance, the feasibility of the method is confirmed and suggestions for improving supply reliability are provided.

RELIABILITY ENGINEERING & SYSTEM SAFETY (2021)

Review Energy & Fuels

Natural gas consumption forecasting: A discussion on forecasting history and future challenges

Jinyuan Liu et al.

Summary: The paper reviews the history of natural gas consumption forecasting technology and categorizes it into different stages based on the characteristics of forecasting models used in each period. It identifies the fastest-growing forecasting horizon as short-term forecasting, mainly influenced by temperature variables, and the best models for long-term forecasting as time series models. Additionally, AI-based models show the best performance for medium-term and short-term forecasting.

JOURNAL OF NATURAL GAS SCIENCE AND ENGINEERING (2021)

Article Energy & Fuels

Time series prediction for output of multi-region solar power plants

Jianqin Zheng et al.

APPLIED ENERGY (2020)

Article Automation & Control Systems

Short-term natural gas consumption prediction based on Volterra adaptive filter and improved whale optimization algorithm

Weibiao Qiao et al.

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (2020)

Article Computer Science, Information Systems

An online real-time estimation tool of leakage parameters for hazardous liquid pipelines

Jianqin Zheng et al.

INTERNATIONAL JOURNAL OF CRITICAL INFRASTRUCTURE PROTECTION (2020)

Article Energy & Fuels

Short-Term Forecasting of Natural Gas Consumption Using Factor Selection Algorithm and Optimized Support Vector Regression

Nan Wei et al.

JOURNAL OF ENERGY RESOURCES TECHNOLOGY-TRANSACTIONS OF THE ASME (2019)

Article Chemistry, Physical

An MILP model for the reformation of natural gas pipeline networks with hydrogen injection

Bohong Wang et al.

INTERNATIONAL JOURNAL OF HYDROGEN ENERGY (2018)

Article Computer Science, Software Engineering

Wound intensity correction and segmentation with convolutional neural networks

Huimin Lu et al.

CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE (2017)

Article Construction & Building Technology

Daily natural gas consumption forecasting based on a structure-calibrated support vector regression approach

Yun Bai et al.

ENERGY AND BUILDINGS (2016)

Article Energy & Fuels

Energy consumption modeling using artificial neural networks: The case of the world's highest consumers

Gokhan Aydin et al.

ENERGY SOURCES PART B-ECONOMICS PLANNING AND POLICY (2016)

Article Energy & Fuels

Energy consumption modeling using artificial neural networks: The case of the world's highest consumers

Gokhan Aydin et al.

ENERGY SOURCES PART B-ECONOMICS PLANNING AND POLICY (2016)

Proceedings Paper Business

Model estimation of ARMA using genetic algorithms: A case study of forecasting natural gas consumption

Beyzanur Cayir Ervural et al.

12TH INTERNATIONAL STRATEGIC MANAGEMENT CONFERENCE, ISMC 2016 (2016)

Article Computer Science, Artificial Intelligence

A fuzzy nearest neighbor neural network statistical model for predicting demand for natural gas and energy cost savings in public buildings

James A. Rodger

EXPERT SYSTEMS WITH APPLICATIONS (2014)

Article Construction & Building Technology

Forecasting of daily natural gas consumption on regional basis in Turkey using various computational methods

Fatih Taspinar et al.

ENERGY AND BUILDINGS (2013)

Review Green & Sustainable Science & Technology

Forecasting of CO2 emissions from fuel combustion using trend analysis

Aylin Cigdem Kone et al.

RENEWABLE & SUSTAINABLE ENERGY REVIEWS (2010)

Article Energy & Fuels

Artificial neural network modeling for forecasting gas consumption

FB Gorucu et al.

ENERGY SOURCES (2004)