Work

Research on multi objective dynamic control strategy of electric vehicle air conditioning for energy-consumption comfort air-quality collaborative optimization

Year: 2026

Type: article

Abstract: Aiming at the limitations of traditional single-objective regulation in vehicle air conditioning control, the paper proposed a multi-objective dynamic control scheme for Electric Vehicle Air Conditioning (EVAC) based on energy-saving, comfort, and health collaborative optimization. Firstly, a multi-objective collaborative optimization control scheme for EVAC was based on the Predicted Mean Vote (PMV) of thermal comfort indicators and Indoor Air Quality (IAQ) of air quality indicators. Secondly, the Improved Non-dominated Sorting Genetic Algorithm-II (I-NSGA-II) algorithm was proposed to perform multi-objective optimization on energy consumption, PMV, and IAQ. The Latin Hypercube Sampling (LHS) and Differential Evolution (DE) crossover strategy in the algorithm improved the global search efficiency and solution distribution uniformity during the solving process. During the process of finding the optimal solution using the I-NSGA-II algorithm, the Long Short-Term Memory neural network (DE-QPSO-LSTM) algorithm proposed in the paper, could solve the problems of poor prediction accuracy in adjusting long-term series data and difficulty in determining hyperparameters in prediction models, thus better achieving priority adjustment of air conditioning power allocation. Finally, the simulation and experimental results showed that the scheme can effectively optimize the requirements for comfort, health, and energy efficiency during the use of air conditioning in vehicles. • Dynamic collaborative optimization of PMV and IAQ, constructing a comprehensive optimization model for energy consumption, comfort, and air quality. • Proposed the DE-QPSO-LSTM battery capacity prediction technology, which improved accuracy and efficiency. • Proposed the I-NSGA-II optimization algorithm, which improved the global search efficiency in multi-objective optimization. • Designed and conducted comparative experiments.

Source: Energy Reports

DOI: 10.1016/j.egyr.2025.109012

Authors: Ning Li, Kai Xie, Yuhao Hu, Ziping Wang, Chunlin Wang, Hao Qi

Institutions: Weifang University of Science and Technology, Shandong Huayu University of Technology

Language: English

Topic: Building Energy and Comfort Optimization

Subfield: Building and Construction

Field: Engineering

Domain: Physical Sciences

SDGs: Affordable and clean energy, Industry, innovation and infrastructure, Sustainable cities and communities, Climate action

Open Access status: gold

Estimated APC paid: $3,040