Carbon emission measurement method of regional power system based on LSTM-Attention model | Science and Technology for Energy Transition (STET)

Decarbonizing Energy Systems: Smart Grid and Renewable Technologies

Open Access

Table 1

Comparison between this paper and related works.

Method	Typical case	Advantage	Disadvantage
IPCC method		Ouyangbin [4]Lu [5]	High facetedness, High transparency, Strong traceability, Wide applicability		High data requirements, Not applicable to small-scale projects, Lack of flexibility
Actual measurement		Zhou [6]	High accuracy, Wide applicability, Sustainability assessment		Costly, Data uncertainty, High degree of restriction
Material conservation method		Yan [10]	Simple and easy to follow, High comprehensiveness, Wide applicability		High data uncertainty, Difficult to quantify indirect emissions, High limitations
Carbon footprinting		Zhao [8]Shang [9]	Full in terms of assessment, Highly comparable		High data demand, High data uncertainty, Scope limitations
Carbon flow approach		Zhang [11]	High accuracy, Strong tracking capability, Wide applicability		High data demand, High data uncertainty, High complexity

Ref	Objective functions	Solution procedure	DSM strategy	Advantage	Disadvantage

[13]	Environment, Cost, Power	Multi approach	Load curtailment, Load shifting	The ε-constraint method guarantees a set of non-inferior solutions to the multi-objective problem	High demands on computing resources and time
[14]	Environment, Cost, Requirement	Multi approach	Demand curtailment, Demand shifting, Onsite generation	Optimization problems for complex systems with simultaneous consideration of multiple objectives	High computational complexity
[15]	Environment, Cost, Deviation	Epsilon-constraint method	Load shifting	Handling multiple conflicting objectives, applicable to various types of optimization problems	Calculation costs increase with the number of targets
[16]	Environment, Cost, Optimal shifting	Augmented epsilon-constraint	Optimal shifting and Strategic conversion as reserve	Efficient generation of multiple non-inferior solutions, flexible and independent	Computational costs increase with problem size
[17]	Cost, Emission, LOLE, Deviation	Epsilon-constraint method	Optimal shifting	Combined consideration of multiple conflicting objectives; ε-constraint method ensures a non-inferior solution set for multi-objective problems; improved practicality and robustness against renewable energy uncertainty	Computational complexity, subjectivity, and limitations in accuracy and efficiency when dealing with high-complexity uncertainties in large-scale systems
This paper	Cost, Emission, Deviation	Carbon emission measurement method of regional power system based on LSTM-Attention model	Optimal shifting	The support system has a comprehensive understanding of the distribution trajectory of carbon emissions in the power system; combined with neural networks, it can predict the fluctuation of carbon emissions over time	High data quality and quantity requirements and high consumption of computing resources

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.