Multi-objective optimization scheduling of integrated energy systems considering regional time-of-use electricity prices and weight sensitivity

• Published in: Electric power systems research Vol. 236; p. 110905
• Main Authors: Li, Jianlin, Wu, Yiwen, Ma, Suliang, Zhang, Jianhui, Sun, Xinzhe
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01-11-2024
• Subjects: Flexible load; Heating network; Multi-objective weight optimization; Natural gas network; Power network; Power network; Flexible load; Natural gas network; Multi-objective weight optimization; Heating network
• DOI: 10.1016/j.epsr.2024.110905
• ISSN: 0378-7796

•Propose a zonal price optimization mechanism based on regional energy redundancy. Targeting the issue of different regional energy imbalances in IES, different EV charging prices are used in different regions to guide EV charging demand at different locations and times. Compared to Time of Use pricing (TOU), the use of Regional Time of Use pricing (RTOU) reduces EV charging costs by 48.31 %. Simulation results demonstrate the superiority of RTOU in guiding EV charging demand.•Addressing the multi-objective optimization issue in IES optimization dispatch, propose a multi-objective weight optimization approach based on single-indicator weight sensitivity. The key focus of the proposed method lies in considering the sensitivity of individual indicators as the weight increases. By allocating weights based on the differences in indicator sensitivity to weights, the Energy Input-Output (EIO) increases by 1.73 %, validating the effectiveness and superiority of this optimization approach.•Introduce an EW-TOPSIS method for multi-index weight optimization. Utilizing the concept of entropy weight to characterize the sensitivity of a single indicator as the weight changes, the EW-TOPSIS algorithm combines the TOPSIS method to compare and evaluate multiple indicators and ultimately quantitatively assign weights to the indicators. Compared to the weights optimized using EW, the EW-TOPSIS method increases the EIO by 1.66 %, demonstrating that the EW-TOPSIS algorithm effectively evaluates the differentiated results of multiple indicators as the weight increases and has a good application effect in multi-objective weight optimization problems.
The integration of Electric Vehicles (EV) and diverse energy users in an Integrated Energy System (IES) poses a key challenge in IES optimization dispatch. This paper presents an IES optimization dispatch method that incorporates EV zoning electricity prices and multi-objective weight optimization. Firstly, the power flow calculation is integrated into the IES optimization dispatch, establishing a comprehensive IES network power flow model that considers electricity, gas, and heat. Next, an innovative Regional Time of Use Price (RTOU) is developed based on regional energy redundancy and the division of IES regions. This RTOU guides the charging demand of EVs in different regions and at different times. Subsequently, an approach for multi-objective weight optimization is proposed, which takes into account the differences in weight response and the incongruity of indicator meanings in the IES context. This approach utilizes a single-indicator weight sensitivity matrix to optimize the weights and incorporates the EW-TOPSIS evaluation method to achieve multi-dimensional weight optimization in IES. Finally, the second-order cone programming method is employed to solve the IES power flow model. This verifies the practical effectiveness of the proposed weight optimization approach based on the single-indicator weight sensitivity matrix and analyzes the application effect of the EW-TOPSIS algorithm in multi-objective optimization. Simulation results demonstrate the positive outcomes of this research: (1) The use of RTOU reduces the additional charging cost of EVs by 5.24 % compared to the use of Time of Use (TOU), validating the effectiveness of the improved pricing optimization mechanism. (2) The weight optimization method based on weight sensitivity increases the weight utilization index by 1.73 % compared to the use of normalized weights. The findings in this paper serve as a valuable reference for addressing multi-objective optimization problems in future IES dispatch work.