Search

The ringing and the resulting asymmetry in the propagation delay are the main causes of bandwidth degradation and communication errors in a controller area network (CAN) with flexible data rate. Therefore, in this paper, the mechanism of ringing is analyzed, and the ringing suppression circuit is proposed based on the analysis. The proposed circuit includes an impedance matching resistor and a forward-biased diode for terminating the repetition of reflections and limiting the magnitude of the ringing. These components are temporarily connected by operating a switch that is controlled by the TxD signal of a CAN controller so that the ringing suppression circuit only engages when necessary and at the correct time. The performance of the ringing suppression circuit is verified by experiments. The experimental results show that the proposed circuit keeps the ringing adequately suppressed and the good symmetry in the propagation delay across the physical bus is possessed.

Signal distortion, called ringing, in a controller area network with flexible data rate (CAN-FD) causes asymmetry in propagation delay throughout a network, which is dependent on its physical bus design and becomes a critical issue in limiting the data rate. Thus, in this paper, the bus structure of a CAN-FD is investigated and a transfer function of ringing is modeled by considering the physical bus configuration and features of a controller area network transceiver. The effects of the main parameters of the physical bus, such as the stub line length and number of stub lines connected to a joint connector, are then quantitatively analyzed using the presented model. When a practical bus structure is considered, additional signal distortion owing to a neighboring connector in a physical bus is modeled and analyzed. These analytical results are confirmed experimentally. Their comparison shows that the presented model-based analysis can closely approximate the size and time response of ringing and the extent of bit time distortion. The presented model-based analysis facilitates reliable communication performance by estimating the ringing and resulting asymmetry of the propagation delay for a given physical bus design and, if necessary, adjusting the design.

Hybrid electric vehicles (HEVs) provide a promising alternative to conventional engine-powered vehicles, with less emission and better fuel economy. This study proposes a hierarchical control for a power-split HEV over a driving cycle, featuring pre-trip optimisation and en-route speed adaption. Constraints including vehicle powertrain boundaries, road gradients, and speed limits, are taken into consideration. In the first stage, the HEV operating conditions, including the optimal vehicle SOC, speed profiles, and total driving time, are generated for the entire trip before departure. Based on the pre-trip results, the second stage adapts the vehicle speed for a short horizon when driving, while taking the safety spacing to the preceding vehicle into consideration, which acts as an indicator of actual traffic conditions and guarantees safe driving. Both optimisations are conducted under the distance domain for realising localisation in the optimal speed profile due to frequent changes in traffic conditions. An estimation of distribution algorithm is used to run the simulation so that the feasibility, robustness, and effectiveness of the proposed hierarchical control can be demonstrated.

This paper proposes a distance-based two-stage energy management strategy for power-split plug-in hybrid electric vehicles (PHEVs). One stage is for long-term optimization and the other is for short-term adaptation to actual traffic conditions. Energy consumption in PHEVs depends on the characteristics of the drivetrain as well as the operating conditions such as power demands and their split. Thus, prior to departure, the operating conditions for a whole trip are optimized for the drivetrain characteristics and trip information, which generates optimal speed and state-of-charge profiles. While driving, the operating conditions are adapted to current traffic conditions for a short horizon on the basis of long-term optimization results. In consideration of the changeability of traffic conditions, the proposed energy management strategy is performed in a distance domain, which localizes the effects of changes in traffic conditions on the long-term optimization results. Therefore, this distance-based two-stage strategy improves the balance between the optimality and the real-time computing time, which is suitable for online management. A model for the propulsion system in a PHEV and the energy management strategy were formulated in a distance domain. An estimation of distribution algorithm was used for long-term optimization and local adaptation.

The response time of a comparator in a time-pickoff circuit used for time measurement depends on the slope and the overdrive of its input signal. Thus, the generation time of the timing pulse in a time-pickoff circuit is dependent on the amplitude of the input signal, which is the main cause of a timing error. In this study, a time-pickoff method, which employs an automatic gain control (AGC) circuit, is proposed for reducing the timing error called time walk. The AGC circuit in front of the comparator renders the amplitude of the input signals to the comparator to be uniform and the proposed method can be arranged to be relatively independent of the properties of the comparator. The performance of the proposed time-pickoff method was tested and its effectiveness was confirmed through the experiments. It is found that the time walk of the proposed time-pickoff method is less than 227 ps for input signals with a rise time less than 1 ns and a dynamic range of 40 dB.

The response time of a comparator involved in a time-pickoff circuit depends on the overdrive and the slope of its input signal, which is the main cause of a timing error. For eliminating the dependency, a time pickoff method employing an automatic gain control (AGC) circuit renders the amplitude of the input signal to the comparator to be uniform. However, its application is limited to the low count rate measurement due to relatively long delay of the nonlinear bias circuit for controlling the gain in the AGC circuit. In this study, a time pickoff method, which employs an automatic gain control circuit with a shorter delay, is proposed for fast count rate measurements. According to the amplitude, the input signal is divided into 4 steps. Then, the gain in the AGC circuit is set to be same for the input signals of the same step, which is basically inversely proportional to the amplitude. This gain setting is performed at one time by amplitude comparison, which is similar to a flash type analog-to-digital converter (ADC) structure, and has a smaller propagation delay. The performance of the proposed method is analyzed through the experiments for input signals with a dynamic range of 40dB.