①Acquisition of data
(1) Data setting
When planning the types of data collected for many working conditions in foreign countries, in order to meet multiple purposes at the same time, a large number of collection parameters are usually set: auto speed, engine speed, engine oil temperature and water temperature, driving time, mileage, road gradient, throttle position , Fuel consumption, ambient temperature, energy consumption of the electrical system, and the use of braking devices, and even the use or operation of lights, wipers, rear window heaters, and engine fans. But for the development of a specific auto driving condition, not all of the above parameters are necessary. Excessive pursuit of details has no specific meaning in the development process of driving conditions based on the principle of statistical characteristics. From the perspective of the development process and performance results of working conditions, it is necessary to record auto speed, engine speed, fuel consumption, and parameters related to fuel consumption. With the three-dimensionalization of urban traffic, a slope should be added to the driving conditions so that the driving conditions can reflect the actual characteristics of urban road traffic. In addition, the slope directly affects fuel consumption.
(2) Pulse number selection and sampling interval setting
In order to obtain more realistic data, the researchers hope to use the auto’s own sensors (modern autos provide this possibility); and commonly used external high-precision speed sensors such as microwave type and photoelectric type may not be normal due to rain and snow. use. The number of speed signal pulses are generally: 6, 24 and 48 per wheel revolution, or even higher (such as for ABS, TCS, etc.). Starting from the consistency of the proportional distribution of the parameters of the working conditions, it is recommended to use as many pulses as possible (such as 48 pulses/week) to obtain the actual driving conditions data. The sampling frequency (generally 5Hz, 2Hz and 1Hz) of the on-board data recording equipment is also a very important factor. The longer the time interval, the smaller the data fluctuation; however, using a too large sampling interval will smooth out larger acceleration values, and will also underestimate the proportion of low speed. Since larger acceleration values have a greater impact on auto design and evaluation, it is necessary to avoid such errors. According to the current auto sensor configuration (that is, the number of pulses per wheel revolution), a sampling frequency of 2Hz (0.5s) is recommended.
(3) Determination of the amount of data
There are a lot of relevant studies on domestic working condition surveys, but the results are quite different. The reason is that on the one hand, the traffic flow survey is not scientific enough, and the planned test route is not representative. On the other hand, the amount of raw data collected is different.
The amount of data collected and the accuracy of the final derived result have a relationship as shown in Figure 1. In theory, the more data collected, the more accurate the results will be. But when the amount of collected data reaches a certain value n, even if the amount of data is increased, the accuracy will not be greatly improved. At the same time, due to the limitations of objective conditions, the amount of data collected is also certain. Where conditions permit, collect as much data as possible. When obtaining massive amounts of data, it is necessary to adopt advanced statistical methods and means.
②Analysis and processing of data
There are two main methods for data analysis and processing: one is to analyze the entire driving process as continuous facts and phenomena with statistical methods, before constructing driving conditions, artificially classify the Driving Cycles according to the test area, and artificially synthesize them; The other is to start from the road traffic conditions, through the research and classification of the various kinematic segments that constitute the entire driving process, and then construct the working conditions. The latter is also the latest research method currently used abroad. autos depart from the start to the destination to stop, affected by road traffic conditions, during which they will go through many start and stop operations. The movement of the auto from the start of one idle speed to the start of the next idle speed is defined as a kinematic segment (hereinafter referred to as a segment). As shown in Figure 2, the entire journey can be regarded as a combination of various segments. The traffic conditions reflected in some of the segments may be consistent. The same fragments may appear at different times, locations, and road types. Sometimes the characteristics of the fragments on a busy highway may be similar to those on a crowded city road. Link these types of segments with traffic conditions, analyze the patterns of low-speed, medium-speed, and high-speed movement in a targeted manner, and construct working conditions on this basis.
Analyze the auto speed curve as a function of time, and the characteristic parameters of this curve can be used as a function of traffic conditions.
The kinematic segments are continuously segmented from the original data, and the characteristic parameters of these segments, such as duration, segment length, speed, acceleration, etc., are analyzed by the principal component score. On this basis, the cluster analysis method is used to classify the fragments to obtain the class set corresponding to the traffic condition; finally, the probability is used to construct a suitable representative working condition with a moderate length of time. The analysis of short stroke characteristics is mainly considered from the following aspects: short stroke length, idle time, short stroke duration, average speed, operating speed (not including average speed during idle time), maximum speed, and standard deviation of speed and acceleration, etc. .