Login 
This article provides information and best practice on noise impact assessments to support the development of Battery Energy Storage Sites (BESS). Whilst this relates primarily to UK developments, many of the same principles apply elsewhere.
Battery storage technology
The current demand for carbon reduction to achieve the UK’s net zero targets has led to a drive for developments in energy storage, grid balancing, solar parks, and reserve power facilities in the UK. This has led to several developers investing in BESS facilities to store and supply electrical energy either as a standalone development, or increasingly, to tie-in with naturally volatile energy sources such as wind and solar. The simplest BESS configuration typically consists of four main components; battery containers (often based upon standard shipping containers), along with cooling systems, inverters, and transformers.
Batteries can be housed in containers of variable sizes, each requiring a cooling system to regulate the internal temperature. The containers connect to inverters and low-voltage transformers; these are further stepped up by a primary transformer connected to the national electrical grid. The number of transformers, inverters, containers etc. are dependent upon the size of the site and environmental constraints such as the impact of noise upon surrounding sensitive receptors.
Other technologies such as smaller, modular battery ‘cubes’, and synchronous condenser (flywheel) systems, can also be incorporated into a BESS site. Synchronous condensers are used to stabilise the grid power supply through the use of a flywheel, smoothing the power fluctuations in the gird.
“noise levels have to be assessed for day and night periods which can give different levels of impact”
A battery cube is a small container of battery arrays that has a liquid cooling system to manage the battery temperature. The liquid cooling system reduces the necessary size of the container and is generally more suited to smaller developments.
Main noise sources
A BESS operates on a continuous automated basis, as such any noise sources can in theory operate at any time. As such, noise levels have to be assessed for day and night periods which can give different levels of impact given the sensitivity of the period. The main noise sources from a BESS are:
- The cooling system for the battery container/cubes
- Inverters
- LV to MV transformers
- Primary transformer
Where a synchronous condenser is proposed for the site, the flywheel can be a significant noise source with typical Sound Power Levels up to 110 dB(A). As such, the flywheels and associated plant are often housed within a building with an appropriate acoustic performance depending on site context and distance to sensitive receptors.
As part of a planning application, a noise assessment for a BESS needs to consider the worst-case scenario, namely a full capacity charging/discharging period, which means assessing the maximum noise emission from all plant operating simultaneously and at full capacity. However, this can overestimate the development’s noise emissions as, in practice, BESS facilities will not charge batteries to full capacity (to preserve long battery life) and generally operate at 40-50% charging capacity, as such, the level of impact tends to be lower in practice than those presented in assessments.
Planning applications
The noise assessment for every site varies depending on the context of site, the location, and existing acoustic environment of the area. Understandably, BESS are typically situated adjacent, or in close proximity to a national grid substation, these substations are often in rural locations and as such the background acoustic environment of the site becomes an important factor when considering impact.
Local Planning Authorities in the UK typically ask for noise impact assessments to be undertaken to British Standard BS4142:2014+A1:2019 ‘Methods for rating and assessing industrial and commercial sound’ (‘BS4142’). Although BS4142 is generally considered most appropriate for such developments, there are some councils that may impose planning conditions with different limits (such as the use of NR curves). As such, it is good practice to undertake consultation with the Environmental Health Officer (EHO) of the relevant Council to agree methodology and criteria at an early stage of the assessment and prior to any surveys.
A brief methodology of a BS4142 assessment is summarised below.
BS4142
BS4142 outlines methods for rating and assessing sound from industrial developments in order to provide an indication of its likely impact upon nearby premises (typically residential dwellings).
The ‘Specific’ sound emitted by a development (dB, LAeq) is predicted and converted to a ‘rating’ level by taking into account both the level and character of the sound (i.e., tonal elements, impulsivity, intermittency and distinctiveness), the resulting rating level is then assessed against the existing prevailing background sound level (dB, LA90) at that location to determine a likely level of impact. Therefore, a survey to establish the local acoustic background is necessary as part of the noise impact assessment.
The level by which the rating level exceeds the prevailing background sound level indicates the following potential impacts as specified in BS4142:
- A difference of 10 dB or more is likely to be an indication of a significant adverse impact, depending on the context
- A difference of around 5 dB is likely to be an indication of an adverse impact, depending on the context
- Where the rating level does not exceed the background level, this is an indication of the specific sound source having a low impact, depending on the context
A difference of no more than 5 dB above prevailing background is the commonly accepted noise limit, as this reflects the onset of a potential adverse impact. However, some Local Authorities often seek a lower level where possible. Again, good consultation/communication is key to agreeing a reasonable approach, which should be undertaken on a case-by-case basis.
Background monitoring
An integral part of the BS4142 assessment is the background sound survey to determine the prevailing background levels. The aim of the survey is to establish the existing background noise environment at the nearest sensitive receptors (NSRs); it is important to ensure that all NSRs are represented, appropriate periods such as quiet evening times and sleeping periods are measured, and periods that may exhibit lower background levels are included where necessary
“a difference of no more than 5 dB above prevailing background is the commonly accepted noise limit”
We generally consider it good practice to measure over 24 hours or more on a representative day (without atypical activities) to present a comprehensive sound profile of the location. Where shorter durations may be necessary due to security concerns, weather conditions or access restrictions, a quieter period should be measured with care taken to avoid rush-hour periods and any other times where background levels may be elevated.
It is good practice to present the proposed monitoring locations and duration of survey to the Local Planning Authority for agreement in advance, occasionally a particular residence or other NSR may require additional monitoring due to a historical noise complaint or sensitive nature.
Defining a representative monitoring location
It is important to identify representative locations to establish a comprehensive background profile around the development site. The monitoring locations need to be as close as reasonably practicable to the identified receptors; ideally placed at the perimeter of the receptor boundary. When determining representative background sound it is important to note the sources at receptor locations. Some examples to note:
- A large farm will have higher sound levels due to the farm activities as opposed to a single residence as such measured levels at a farm may not be representative.
- Dwellings next to a main road, railway line or river will tend to experience higher background levels than those dwellings located further away.
- A specific receptor may have plant such as heat pumps, generators, or workshop activity. Therefore, whilst the measured levels may be representative of that property, they may not be representative of other local receptors.
- Look out for receptors near overhead power lines and transformer poles which may affect the measured sound levels (usually picked-up as a peak at ~100Hz).
- Residences close to existing substations will not normally provide a representative background for all nearby NSRs. In such cases, monitoring should be taken either at a more distant receptor where background level is not dominated by the substation (a worst-case approach), or two or more monitoring locations should be used to provide sufficient coverage.
Careful analysis and consideration of the surrounding environment and expected acoustic climate is necessary to ensure the chosen monitoring location are representative of all sensitive receptors surrounding the development.
Rural context
Most BESS developments are located in rural areas, either as standalone battery storage sites or combined with solar/wind farm developments. Background levels in a rural context are typically characterised by road traffic noise, which exhibits an obvious diurnal pattern as shown below.
Night-time sound levels can be as low as 20 dB(A) in rural areas (a recording studio is generally around 25 dB); the night-time assessment can consequently become a significant hurdle for the development. It is therefore important for the consultant to take a reasonable approach that reflects the reality of noise perception and level of impact to the residents of the dwellings.
“the night-time assessment can become a significant hurdle for the development”
The essence of assessing noise impact at night is to consider the potential disturbance to sleep or adverse impact on sleep quality of the residents in the NSR. As such, BS4142 permits the use of alternative absolute criteria or internal limits such as: BS8233: 2014 ‘Guidance on sound insulation and noise reduction in buildings’, NR Curves, or WHO: ‘Guideline for community noise: 1999’ where background levels are particularly low, thereby providing some flexibility and ensuring developments are not unreasonably constrained.
Acoustic character – rating corrections
In addition to the predicted noise levels from the Development BS4142 states that corrections should be applied to predicted noise levels to account for certain acoustic features which have the potential to increase the level of noise impact at nearby dwellings.
The four acoustic features to be considered in the application of rating corrections are as follows:
1 Tonality:
“… a penalty of 2 dB for a tone which is just perceptible at the noise receptor, 4 dB where it is clearly perceptible, and 6 dB where it is highly perceptible.”
2 Impulsivity:
“… a penalty of 3 dB for impulsivity which is just perceptible at the noise receptor, 6 dB where it is clearly perceptible.”
3 Other sound characteristics:
“Where the specific sound features characteristics that are neither tonal nor impulsive, though otherwise are readily distinctive against the residual acoustic environment, a penalty of 3 dB can be applied, and 9 dB where it is highly perceptible.”
4 Intermittency:
“… If the intermittency is readily distinctive against the residual acoustic environment, a penalty of 3 dB can be applied.”
A BESS is commonly dominated by cooling fan noise from battery storage containers, which tend to be broadband in nature (non-tonal). They are also usually controlled via individual thermostats and therefore the development does not have clearly perceptible on/off states (i.e. non-intermittent). A rating correction for tonality may be necessary when receptors are located particularly close to the transformers due to a low-frequency (100 Hz) ‘hum’ that can
be audible at receptors. However generally speaking, this is often masked by other plant, and/or the existing background noise.
Operational monitoring
Following consent of the development, it is relatively common for planning conditions to require an operational noise survey to be undertaken once the development is operational to affirm the predicted noise levels and impact presented in the application stage. A similar approach may also be required in the event of a noise complaint. This process is much the same as that described above, but with the Specific noise of the BESS facility being measured, rather than predicted.
The simplest method is to monitor at the same locations as the background survey, ideally over the same days of the week and duration. This method allows for a clear comparison between predicted and operational noise levels in practice, and also allows for observation of any applicable acoustic corrections at the NSR, thereby ensuring the level of impact remains compliant with the accepted criteria.
Calculating tonality
We recommend measurement of third octave frequency bands (1/3 octaves) when monitoring operational noise. The measurement of 1/3 octaves will allow an objective assessment of tonality at the receptor locations over appropriate intervals using the method described in BS4142 Annex C:
“… For a prominent, discrete tone to be identified as present, the time-averaged sound pressure level in the one-third-octave band of interest is required to exceed the time-averaged sound pressure levels of both adjacent one-third-octave bands by some constant level difference.
The level differences between adjacent one-third-octave bands that identify a tone are:
- 15 dB in the low-frequency one-third-octave bands (25 Hz to 125 Hz);
- 8 dB in the middle-frequency one-third-octave bands (160 Hz to 400 Hz); and
- 5 dB in the high-frequency one-third-octave bands (500 Hz to 10 000 Hz).”
Figure below presents a 1/3 octave chart of an operational noise measurement interval showing a tone at 500 Hz in accordance with the BS4142 method.
It should be noted than BS4142 also describes an alternative ‘Reference’ method to be used if the presence of tones remains
in dispute. This is known as the Joint Nordic Method 2, as found in ISO 1996-2. This is typically a more labour-intensive approach requiring a narrow-band frequency analysis, often performed by recording high-resolution audio for later processing/analysis.
In conclusion
Noise impact assessments for BESS developments can be a varied science; each site will bring its own constraints and special considerations. The essence of the noise assessment is to reflect the likely level of adverse impact from developments to the environment and people, and a robust assessment that prioritises this will effectively be more economically and environmentally beneficial. Therefore, consultation and good communication with all invested parties/authorities is imperative.
