Neural Network-Based Model Reference Control of Braking Electric Vehicles

Abstract

The problem of energy recovery in braking of an electric vehicle is solved here, which ensures high quality blended deceleration using electrical and friction brakes. A model reference controller is offered, capable to meet the conflicting requirements of intensive and gradual braking scenarios at changing road surfaces. In this study, the neural network controller provides torque gradient control without a tire model, resulting in the return of maximal energy to the hybrid energy storage during braking. The torque allocation algorithm determines how to share the driver’s request between the friction and electrical brakes in such a way as to enable regeneration for all braking modes, except when the battery state of charge and voltage levels are saturated, and a solo friction brake has to be used. The simulation demonstrates the effectiveness of the proposed coupled two-layer neural network capable of capturing various dynamic behaviors that could not be included in the simplified physics-based model. A comparison of the simulation and experimental results demonstrates that the velocity, slip, and torque responses confirm the proper car performance, while the system successfully copes with the strong nonlinearity and instability of the vehicle dynamics.

The Impact of the Available Infrastructure on the Electric Vehicle Market in Poland and in EU Countries

“…The greatest production of electricity from photovoltaics occurs when most users are away from home (e.g., at work). The use of an electric car for energy storage may, however, become an interesting idea to increase the self-consumption of energy produced by photovoltaics and thus further accelerate the return on investment in both the electric car and photovoltaics [ 81 , 82 ].…”

Optimizing regenerative braking on electric vehicles using a model-based algorithm in the antilock braking system

“…Friction in the braking process can be used to charge the battery instead of being discharged as heat [23]. Vehicle energy is lost about 20 to 70% during braking [24]. Modeling and simulation are carried out to create performance and safety in automotive [25].…”

Critical review on optimal regenerative braking control system architecture, calibration parameters and development challenges for EVs

“…Compared with an ANN model, FLC is much more dependent on the membership functions but ANN is particularly used to maximize the available energy for recovery. 74 In this approach, the deceleration and speed are considered as input parameters and a portion of the front wheel-braking force is obtained as output. 75 Further, the number of layers and neurons varies in different modelling approaches, such as in NN-based SRM drive control strategy, 76 the ANN has three layers and the hidden layers have five neurons.…”

“…Similar to other controls in the neural network, there are various modelling approaches like NN‐based drive control, 69 NN with sliding mode control, 70 ANN‐based methodology, 71 fuzzy neural network strategy, 72 ANN with PI controller, 73 as presented in Table 2. Compared with an ANN model, FLC is much more dependent on the membership functions but ANN is particularly used to maximize the available energy for recovery 74 . In this approach, the deceleration and speed are considered as input parameters and a portion of the front wheel‐braking force is obtained as output 75 .…”