ABSTRACT

Using the laboratory-based fuel consumption models for predicting real-world fuel consumption requires the measurement of data under certain conditions to obtain high accuracy of predicted result. Therefore, it is necessary to develop a logging device for measuring the real-time fuel consumption and speed of vehicle on the road. This article presents a study on developing the on-board data logging device to collect real-world data of fuel consumption and speed for motorcycles with the update rate of 1 Hz. The instantaneous speed of the motorcycle was determined based on the rotational speed of the wheel and the wheel radius. Another module was used to determine the instantaneous fuel consumption rate (FR) though measuring the duration injection pulse. The relationship between the duration injection pulse and the injected amount of fuel was established with high correlation coefficient of 0.997. In addition, a filter was designed to remove noise in the dataset collected using the data logging device. The random errors in the speed and the FR profiles were detected and replaced, the percentage of these errors is 1.8% and 2.4%, respectively. The developed method is a precise one for transient fuel consumption and speed measurement. In chassis dynamometer test, the average deviation between steady speed measured by the chassis and the data logging device is only approximately 0.35%. At transient state, the biggest deviation between these two datasets is less than 3.5%. The average FR at steady speed measured by the developed method is slightly different from the one measured by the carbon balance method. The difference is 0.9%, 2.5%, and 0.25% at the speeds of 30 km/h, 50 km/h, and 70 km/h, respectively. Following the transient test cycle, the fuel consumption measured by the developed method is 4.35% lower than that determined by the carbon balance method.

Implications: A robust method for collecting and processing the on-road instantaneous data of fuel consumption and speed was developed for motorcycles. The proposed method can record well the real-world driving data for motorcycles, including the fuel consumption and speed, with the update rate of 1 Hz. The filter was designed to minimize noise while maintaining data integrity of the collected dataset, the percentage of errors in the the speed and the FR profiles is 1.8% and 2.4%, respectively. The proposed method, therefore, can be used as effective tools for future studies relating to the fuel consumption and emission of motorcycles on the road.

摘要

使用基于实验室的油耗模型预测实际油耗需要在特定条件下测量数据，以获得高精度的预测结果。因此，有必要开发一种实时测量道路上车辆油耗和车速的测井装置。本文介绍了开发车载数据记录设备的研究，以收集摩托车油耗和速度的真实世界数据，更新率为 1 Hz。摩托车的瞬时速度是根据摩托车的转速确定的。车轮和车轮半径。另一个模块用于通过测量持续喷射脉冲来确定瞬时燃油消耗率 (FR)。喷射脉冲持续时间与燃料喷射量之间的关系以0.997的高相关系数建立。此外，还设计了一个过滤器来去除使用数据记录设备收集的数据集中的噪声。检测并替换了速度和 FR 曲线中的随机误差，这些误差的百分比分别为 1.8% 和 2.4%。所开发的方法是一种精确的瞬态油耗和速度测量方法。在底盘测功机测试中，底盘测得的稳定速度与数据记录装置的平均偏差仅为0.35%左右。在瞬态时，这两个数据集之间的最大偏差小于 3.5%。由开发的方法测量的稳定速度下的平均 FR 与通过碳平衡方法测量的略有不同。在30公里/小时、50公里/小时和70公里/小时的速度下，差异分别为0.9%、2.5%和0.25%。在瞬态测试循环之后，所开发方法测得的燃料消耗比碳平衡法测得的低 4.35%。

影响： 开发了一种用于收集和处理摩托车油耗和速度的道路瞬时数据的稳健方法。所提出的方法可以很好地记录摩托车的真实驾驶数据，包括油耗和速度，更新率为 1 Hz。在速度和 FR 配置文件中，分别为 1.8% 和 2.4%。因此，所提出的方法可用作未来有关摩托车在道路上的燃料消耗和排放的研究的有效工具。