Purpose:
Experiment 2 served several objectives, which are described in brief as:
To verify the design of the test profiles
To verify that the driver profiler is able to produce experimentally repeatable test profiles.
To empirically obtain battery-operating constraints under various load demands.
To empirically test the performance of the ultracapacitors as a vehicle propulsion power source.
To compare the receptiveness of ultracapacitors and battery pack to regenerative energy during vehicle regenerative braking operation.
Procedure:
With the sizing references described in Chapter 6 and [80], empirical verification of the battery and ultracapacitors constraints were performed to design a M-PEMS for the experimental vehicle. The traction drive used for this experiment and subsequent experimental work was equipped with regenerative power handling capability.
The vehicle was subjected to a predefined 4-segment velocity profile (Figure 8.4) in two sets of experiments. The first tests were conducted with only batteries followed by tests with only ultracapacitors connected to the load. In both cases the total vehicle mass, tyre pressure and ambient conditions were the same.
0 10 20 30 40 50 60
1 11 21 31 41 51 61 71
Segment 1 Segment 2 Segment 3 Segment 4
Time (s)
(km/h)
Figure 8.4 Designed test profiles
Results:
Figure 8.5 shows the results of the tests conducted with only batteries and only ultracapacitors servicing the vehicle load demand. The results of the velocity traces are presented on the same graph for ease of comparison. Form the measured velocity data, the profiles are in agreement indicating excellent experimental repeatability. In both experiments, no driver intervention for the vehicle commanded speed was required. The task of the driver was to only initiate the test programme and steady the vehicle on a level travel trajectory.
With the regenerative capability of the traction drive, no mechanical braking effort was required to decelerate the vehicle to a complete stop. Maximum deceleration was found to have increased to 1.9ms-2
Velocity trace for tests runs with only batteries supplying the load power Velocity trace for tests runs with only ultracapacitors supplying the load power (d)
(b)
(c) (a)
Segment 1 Segment 2
Segment 3 Segment 4
Time(s)
Time(s) Time(s)
Time(s)
Velocity (km/h)Velocity (km/h) Velocity (km/h)Velocity (km/h)
Figure 8.5 Comparison of measured velocity traces
Figure 8.6 shows the battery and ultracapacitor power profiles and the corresponding voltage measurement for the four tests. From the voltage plots, it is seen that the battery voltage experiences an excessive rise. An expansion of this segment is presented as Figure 8.7.
Ultracapacitor Volatge Battery Voltage
Ultracapacitor Power BatteryPower
Segment 1 Segment 1
Power (W) Voltage (V)
Time(s)
Time(s)
Figure 8.6 Battery and ultracapacitor power and voltage profiles
Regenerative Braking
events Ultracapacitor Power
BatteryPower
Ultracapacitor Volatge Battery Voltage Segment 1
Segment 1
Battery Voltage Boundary
Time(s)
Power (W)
Time(s)
Voltage (V)
Figure 8.7 Comparison of battery with ultracapacitors in a regenerative braking event
Discussion:
The top graph of Figure 8.7 shows that the power profiles for both tests are similar, indicating that the net energy transfers in both cases are almost equal. A close examination of the voltage measurement (lower graph of Figure 8.7) reveals that during deceleration events, the battery voltage rises significantly above the threshold (gassing) limit. Although the measured power flowing back to the battery indicates that the battery is being charged, the low charge acceptance rate in batteries causes a large portion of the power to be dissipated as losses. The unreceptive nature of high charging current or a rapid change from discharge to charge mode is manifested as a rise in terminal voltage. This immediate rise in the battery pack voltage is attributed to the sudden increase in density of electrolyte in the pores of the active material. In the case of lead acid batteries, repeated operation under these conditions will lead to a reduction in battery life.
Two interesting observation can be made regarding Segment 3. It is seen in Figure 8.6 that a large regenerative power event also takes place in this segment but the battery terminal voltage rise is much less compared to Segment 1. The first reason for this is due to the reduced battery state of charge in Segment 3. Since these tests were carried sequentially with a maximum battery SoC at the beginning of Segment 1, subsequent segments were executed with the SoC depleting. Recalling the discussion in Chapter 3 (Section 3.14) regarding the maximum changing power, it is empirically observed that in this case, the battery charge acceptance rate is higher at lower SoC. This demonstrates the charging power limitation as a function of the battery SoC. The second reason is not very obvious without a closer examination of the drive cycle shown in Figure 8.5. In Segment 1, the change from acceleration mode to deceleration mode is much faster compared to Segment 3. Although the regenerative power event is similar in magnitude, the change from tractive power to regenerative power in Segment 3 is more gradual. As such, the voltage rise phenomenon that occurs due to a rapid change in direction of battery power is less of a factor in this case. It is timely to mention that both these phenomena do not occur in the case of ultracapacitors.
With no slow chemical phase changes taking places within the ultracapacitors as does in the batteries, regenerative energy is harnessed more efficiency. Generally, the results empirically
From the results of the experiment, practical reference parameters can be extracted as the PMS battery operating constraints. Figure 8.8 presents the measured current and voltage waveforms for Segment 1 as projections of the desired voltage boundaries to the corresponding current levels. In the voltage plot, the lower battery voltage is first defined to be approximately 45V. The point x1 marks an instance where the battery voltage crosses beyond this lower limit. The projection of x1 on the time scale to the battery current waveform gives the corresponding current value, which is marked by intersection point x2.
From the current plot, the maximum battery current is found to be 150A, thus specifying that the battery power limit (V x I) is to be set at a value of 6.75kW. Similarly, the upper voltage boundary specifies the charging power (or regenerative power) limits. The intersection values and projections using marking points x3 and x4 gives a maximum charging current of 25A . As was discussed previously, charging the battery this way is not efficient and ideally the minimum battery power should be zero with all regenerative braking power handled by the ultracapacitor instead.
Desired Battery working voltage boundary
Battery Current (A) Battery Voltage (V)
x2 x1
x3
x4 Defined Battery working current boundary
Time (s)
Time (s)
Figure 8.8 Determining battery operating limits empirically