bijlage 3 - Indicatieve metingen geluidsemissie elektrische voertuigen (let op: indicatief, niet geschikt voor verder gebruik)

INDICATIVE MEASUREMENTS OF THE NOISE EMISSION OF ELECTRIC VEHICLES Wim van Keulen VANKEULEN advies, Schijndel, the Netherlands email: info@vankeulenadvies.nl Since rolling noise is the dominant component of traffic noise the assumption arose that EVs make more noise due to their significantly higher weight. Therefore, the noise emission of electric vehicles (EVs) has been measured on several low-noise road types. The results have been compared to those of vehicles with a conventional internal combustion engine (ICVs). This paper presents the results of indicative noise measurements and the associated statistical analysis. The results show that EVs are louder than ICVs.

Keywords: electric vehicles, noise emission, traffic noise, measurements

1. Introduction Electric vehicles (EVs) are growing in popularity in Europe as the continent takes steps to reduce greenhouse gas emissions and combat climate change. Several factors contribute to the growth of electric cars in Europe, including government incentives, stricter emissions regulations, and improved technology. In 2021, there will be more than 2.2 million electric vehicles on Europe's roads from 2021 onwards. Norway has the highest EV adoption rate, with EVs accounting for more than half of all new car registrations [1]. Other European countries with a high adoption rate of electric vehicles include the Netherlands, Iceland, Sweden, and Denmark. After 2030, no new vehicles with a combustion engine (ICV; diesel or patrol) will be permitted in the Netherlands.

Figure 1: Left panel: Percentage of the total new passenger cars and vans registrations that is full battery electric (BEV). Right panel: Total number of publicly accessible AC recharging points. 1

The European Union has set ambitious targets for reducing emissions, and the transition to EVs is a key part of its plan. The EU aims to reduce carbon emissions from cars by 37.5% by 2030 compared to 2021 levels, which requires a significant increase in EV adoption. During Statistical Pass-by and other measurements on free-flowing traffic, the increasing rate of EVs in traffic is striking. The general assumption is that EVs have a much lower noise emission than ICVs. In the Netherlands this became a political truth. Quote from a spokesman of the Dutch government: “we are convinced that EVs lead to much lower noise levels. Measurements are not necessary and, therefore, undesirable.” The purpose of this paper is to show some data from different measurements under traffic and under normal conditions.

2.

Some earlier studies

Due to major political interests, studies on the subject have often led to discussion about its objectivity (see for example the citation in Section 1). A reliable literature overview is therefore quite difficult to obtain. However, this is not the purpose of this paper. Therefore, we focus on one Dutch study from 2019. In this study it was estimated that replacing all vehicles by EVs, the overall noise reduction would range from about 2 to 3 dB(A) for light vehicles at high speed [2] depending on the road surface. The authors also estimated that the noise-reducing effect of low-noise tires is greater for EVs. An important parameter in the generation of traffic noise in general, and rolling noise in particular, is the weight of the vehicle. Much research is being done on this subject. Because the measurement results vary widely, it is rather difficult to estimate the emission effect of higher loads. Based on older data, an increase of up to 4 dB(A) can be expected upon doubling the load [3]. The second important parameter is the tire width. In general, wider tires produce more road noise than narrower tires. This is because wider tires have a larger contact patch with the road, which can cause more vibrations and noise to be generated as the tire rolls over the surface [3].

Table 1 shows some data of a popular mid-sized car in Europe with both a combustion engine and an electric one. Table 1: Some data of a popular ICV and EV from the same make.

ICV

EV

Tire dimensions

205/55R16

225/55R18

Mass

1164 kg

1712 kg

3. Generation mechanisms of traffic noise To analyze and predict the effect of EVs on traffic noise emissions, it is important to know the most important mechanisms that generate this noise. In particular, the relative weight of the various contributions needs further consideration. In this paper we will focus on both the combustion or propulsion noise and rolling or tire/road noise (for a detailed description see [4]). Fig. 2 shows the respective noise emissions from propulsion noise and tire/road noise as a function of vehicle speed for passenger cars according to Cnossos [5]. Fig. 2 shows the well-known property that tire/road noise is the dominant source at speeds above 40 km/h (from recent measurements is seems that these results for low speeds might be to high). The emission effect of EVs in traffic is first: the propulsion noise of electric engines is lower than that of internal combustion engines, and second: the accompanying tire load is significantly higher. To accurately compare the measured noise levels for ICVs and EVs, it is important to calculate the respective immisions as accurately as possible. ICSV29, Annual Congress of International Institute of Acoustics and Vibration (IIAV), 9 - 13 July 2023

Cnossos 78 76

Noise Level [dB(A)]

74 72 70 68 66 64

propulsion

62

tyre/road

total

60 30

40

50

60 70 Speed [km/h]

80

90

100

Figure 2: Noise emissions from propulsion and tire/road surface as a function of vehicle speed, for passenger cars, calculated with the EU noise mapping model Cnossos EU.

During SPB-measurements according to ISO 11819-1 [6] different vehicle types are distinguished, including new ones like EVs. However, averaging all the measurement results from EVs and ICVs and calculating the difference between these categories is statistically an incorrect method. This had led in some research to biased conclusions. First, the population sizes are very different and second, the effect on low-noise pavements is (probably) different since low-noise pavements act mainly on tire/road noise, and third, the average speeds of both categories are never the same, and finally, the ambient temperatures differ. Therefore, for each measurement the speed-dependent regression-coefficients according to ISO 11819-1 need to be compared with taking into consideration the accompanying 95% confidence intervals as weighing functions. This is an extended method based on the Dutch Croad labelling procedure. Therefore, the possible effect is a function of overall noise reduction, speed, and vehicle category. The noise measurements have been performed on a number of standard Dutch noise-reducing pavements. The average noise levels are calculated by assuming that the confidence interval is an estimate for the density of data points. This average is calculated as follows: 𝐿!",$ (𝑣$ ) =

!",$ (&$ ) (()%c.i.",$ (&$ )). / ∑" (()%c.i.",$ (&$ )).

∑"

(1)

with: Lav: average SPB level [dB(A)], L: local SPB level [dB(A)], k: location, v: speed [km/h], m: vehicle category.

The overall accuracy is calculated similarly according: 95%c.i.!",$ (𝑣$ ) =

&

(2)

/

-∑"

0()%1.2.",$ (&$ )3

.

The advantage of this method is that measurements performed at different speeds can be compared accurately without knowing the results from independent vehicle passages separately. From Eq. 1 the initial noise reductions relative to an asphaltic dense concrete (AC Surf 16) can be determined according to: 𝐶'(')'!*,$ (𝑣$ ) = 𝛥𝐿$ + 𝜏$

&+

"

log 2" $ 3 4,$

(3)

with: Cinital: initial noise reduction [dB(A)], DL: average noise reduction at the reference speed [dB(A)], t : speed dependent term [dB(A)], v: speed [km/h], v!: reference speed [km/h], m: vehicle category.

ICSV29, Annual Congress of International Institute of Acoustics and Vibration (IIAV), 9 – 13 July 2023

3

Table 2 shows the results from Eq. 3 for those speeds where the confidence interval is less than 0,3 dB(A) according to Eq. 1. Table 2: Initial noise reductions in dB(A) for ICVs and EVs on standard low-noise pavements.

Speed [km/h]

ICV

EV

40

6,2

4,5

50

6,9

5,9

From Table 2 it can be seen that EVs are more than 1 dB(A) louder than ICVs at low speeds.

4. Some Consequences 4.1 Acoustical A direct consequence of louder traffic is that noise barriers must be higher to meet legal immission levels. Indicative calculations according to the Dutch noise scheme showed that, despite that 1 dB(A) is a small effect, the respective barriers must be up to half a meter higher. The associated costs of this measure amount to approximately 10 million Euros. 4.2 Durability Mechanical damage to the road surface is a high order non-linear function of the axle load, and therefore, of the tire load. The major type of damage of low-noise pavements is rutting. Due to the heavier light vehicles (see Table 1) the lifetime will be shorter which will lead to higher noise emission during the lifetime of the road and more maintenance with associated (noise) emissions. Additionally, heavy vehicles have a maximum overall weight. That implies that the maximum cargo load will be lower due to heavy batteries which will lead to more traffic movements. This will also lead to higher noise emissions and additionally a shorter lifetime of (low noise) road surfaces. The associated costs of extra maintenance due to reduced durability amount up to approximately 20 million Euros annually.

REFERENCES 1. 2. 3. 4. 5. 6.

Observatory, European Alternative Fuels. https://alternative-fuelsobservatory.ec.europa.eu/interactive-map. 2021. Peeters, B. and E. de Graaff, Quieter traffic at last, thanks to electric vehicles. (in Dutch). Geluid, 21 June 2019. 2: p. 9-12. Sandberg, U. and J.Ejsmont, Tyre/Road Noise Reference Book. 2002, Kisa, Sweden: Informex. Keulen, W. van, Experiencing Noise-Reducing Pavements (On the Interaction of Acoustics, Road Construction, Psychology, Legislation and Policy), keynote lecture, 22nd International Congress on Sound and Vibration. 2015: Florence. Commision, European. Common Noise Assessment Methods in Europe (CNOSSOS-EU). 12-092012. ISO 11819-1, Acoustics -- Measurement of the influence of road surfaces on traffic noise -- Part 1: Statistical Pass-By method. 1997.

ICSV29, Annual Congress of International Institute of Acoustics and Vibration (IIAV), 9 - 13 July 2023