2019 IEEE Vehicle Power and Propulsion Conference (VPPC) 2019
DOI: 10.1109/vppc46532.2019.8952549
|View full text |Cite
|
Sign up to set email alerts
|

Predictive Equivalent Consumption Minimization Strategy for Hybrid Electric Vehicles

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
2

Citation Types

0
12
0

Year Published

2020
2020
2025
2025

Publication Types

Select...
4
2

Relationship

0
6

Authors

Journals

citations
Cited by 6 publications
(12 citation statements)
references
References 18 publications
0
12
0
Order By: Relevance
“…However, it relies on not only the present vehicle states, but also the historical values [113]. While, the NNbased prediction, such as chaining neural network (CNN) [30], BP neural network [31], radial basis function (RBF) neural network [42,114], and recurrent neural network (RNN) [115] only requires the historical velocity sequences without compromising the accuracy of speed prediction. For ECMS applications, Li et al [12] presented comprehensive comparisons among aforementioned speed predictors.…”
Section: Speed Predictionmentioning
confidence: 99%
See 1 more Smart Citation
“…However, it relies on not only the present vehicle states, but also the historical values [113]. While, the NNbased prediction, such as chaining neural network (CNN) [30], BP neural network [31], radial basis function (RBF) neural network [42,114], and recurrent neural network (RNN) [115] only requires the historical velocity sequences without compromising the accuracy of speed prediction. For ECMS applications, Li et al [12] presented comprehensive comparisons among aforementioned speed predictors.…”
Section: Speed Predictionmentioning
confidence: 99%
“…In addition, the regenerative energy can be calculated by the predicted speed and introduced into the EF adaptation as the correction factor [33]. Moreover, the EF adjustment can be regulated by a speed-related variable, called the probability of electric energy consumption [29][30][31][32][33][34][35][36]. This EF adaptation method has been discussed in detail in Section 4.1 in this study.…”
Section: Speed Predictionmentioning
confidence: 99%
“…The EF correction was optimized by using the particle swarm optimization method and the future driving cycle was predicted by using an artificial neural network, and the EF estimation model was updated online. Zhang et al [29] established an ECMS prediction model by predicting vehicle speed through back propagation (BP) neural network and designed dynamic adaptive EF according to the future driving cycle. Zhang et al [30] proposed a chain neural network to predict the speed in different time ranges for application to ECMS.…”
Section: Introductionmentioning
confidence: 99%
“…A key control challenge in a HEV is to allocate its total propulsive/braking demand to its actuators such that either the total fuel consumption or the energy consumption is optimised for a driving mission [2], [3], [9]. To address this problem, several control strategies have been proposed so far, such as 1) deterministic or rulebased strategies [10], [11] and fuzzy-logic based control strategies [12]- [14], which rely on empirical relationships and/or pattern recognition, 2) Equivalent consumption minimisation strategies (ECMS), which rely on minimising a Hamiltonian function at each time step to find a local optima [3]- [5], [9], [15]- [17] and 3) Dynamic programming based strategies, which find the global optima but are computationally intensive, since they perform brute-force search of the entire state-space of feasible solutions [9], [11], [18].…”
Section: Introductionmentioning
confidence: 99%
“…Among these strategies, a standard approach to model the energy conversion dynamics of the powertrain actuators is to approximate them as steady state efficiency maps for control allocation [1]- [7], [9], [16], [17]. Under high transient load demands (a typical use case in HEV's), this method creates mismatch between plant output and controller prediction as it ignores 1) the actuator dynamics and 2) the potential cost for transition between two operating points, hence resulting in a sub-optimal control policy leading to increased fuel and energy consumption [19]- [21].…”
Section: Introductionmentioning
confidence: 99%