Home ExpertSpeak Artificial Intelligence & Energy Efficiency in Building

Artificial Intelligence & Energy Efficiency in Building

‘BPS is generally being performed for building in India after the architectural design has progressed to a reasonably advanced stage and there is not much scope at this stage for incorporating corrective changes in design based on evaluations made using BPS.  There is also very limited decision-making support available at present to quickly assess and evaluate the Energy related performance of multiple design options in the early stages’, says Ar. Aman Batish

The trends in the field of energy efficiency of buildings display a remarkable versatility and promise lately. Developments in energy efficiency have bought energy preservation and management to the centre stage of building construction and design. The reason is that buildings consume a large part of end-use energy, particularly in the residential and commercial sectors.

The figure for India is set to rise further as the building sector grows in line with a projected area of 20 billion sq.m by 2030. A related rise in the Energy consumption in these sectors is already visible in their high Compound Average Growth Rate. Therefore, management and optimisation of energy consumption in buildings assume prime importance, in order to improve their Energy Efficiency.  This is very crucial, if the projected growth in these sectors is to be achieved in a sustainable manner, without adversely impacting the environment.

Government Regulations and Beyond

Realising the importance of energy efficiency in buildings, the Government of India launched the Energy Conservation Building Code (ECBC) under the Energy Conservation Act 2002, to be implemented on a voluntary basis. Under the same Act, implemented on March 1, 2002, the Bureau of Energy efficiency (BEE) was established with a directive to ‘spearhead improvement in energy efficiency through various regulatory and promotional measures and implement the provisions of the act’.

As a result of such initiatives taken in the recent years, the awareness about the matter has increased and the demand for ‘Green Buildings’ in India has also risen. As a matter of fact, India has crossed the milestone of achieving 500 million sq.m of registered green building footprint which has put the country in second place in the world in this regard. Though it is a remarkable milestone, the share of green buildings in the overall building stock is still low, particularly in small scale projects, at around 5% of the total building stock.

Efforts are therefore required to broaden the footprint of energy efficient buildings. Whereas at present, the provision of ECBC are currently applicable only to projects with a connected load of 100kVA, or a contract demand greater than 120kVA, which for all practical purposes makes it applicable only to buildings with an air-conditioned floor area of more than 1000sq.m. Thus, in the absence of a mandatory compliance with energy efficiency norms, other smaller projects in India often depend on the onus and initiative of the respected owners to incorporate the suggested measures of energy efficiency on a voluntary basis.

Moreover, even in case of ‘Green Buildings’, the ratings and certifications are issued on the basis of a theoretical comparison between the proposed design and the ‘Base Case’ for the buildings using Building Performance Simulation tools like ‘Design Builder’ (based on the EnergyPlus engine). There are also prescriptive measures incorporated in the recommendations to achieve a minimum level of energy efficiency in buildings. Mechanisms to monitor the real time performance of the buildings in the operational phase, w.r.t. the rating or certification issued, are generally not in place at present in India.

Prospects of Application of Current Technologies

In the light of these facts, it would be interesting to evaluate the future prospects in the area of Energy Efficiency of Buildings in context of the recent advancements made in the field of computational science, particularly Artificial Intelligence (AI), machine learning and data mining techniques. In order to have clarity on how and in which specific areas, these techniques can be applicable, it is necessary to have a better understanding of the stages involved in efforts aimed at enhancing energy efficiency in buildings, as mentioned below:

Builidng Design Stage:

  1. Early Design Stage decision making
  2. Later Design Stage

Post construction (operational phase) monitoring and control acquires great significance in this scheme of things. The typical design process of a building requiring ‘Green Building’ certification/rating, in the current Indian scenario, involved preparation of multiple design alternatives in the early design stage’ selecting a preliminary design out of these alternatives to work further for building approvals etc. in the later design stage and moving ahead to assess the building’s energy performance using BPS tools. Once the simulations have been performed, ideally there should be scope for the design to be optimised to enhance energy efficiency through passive means and thereafter, the good for construction drawings can be prepared.

However, BPS is a specialised task involving 3D modelling of building designs and requiring detailed physical information like building geometry, weather conditions etc. It is usually outsourced by Architectural Consultants costing them substantially.  BPS is generally being performed for building in India after the architectural design has progressed to a reasonably advanced stage and there is not much scope at this stage for incorporating corrective changes in design based on evaluations made using BPS.  There is also very limited decision-making support available at present to quickly assess and evaluate the Energy related performance of multiple design options in the early stages.

This presents a great opportunity for using AI / machine learning / data-mining based techniques to assess and predict energy consumption and performance of buildings in early design stages. Sine these techniques are data-driven, empirical techniques, they do not require any information about building geometry, but rely instead on the input-output relationships between predictor or independent variables in the data used for training the models. AI based techniques have been shown to outperform even BPS when it comes to predicting building energy consumption of operational buildings. ‘Smart Cities’ and ‘Sustainable Development’ have received government’s attention; ensuring access to ‘affordable, reliable, sustainable and modern energy for all’ has been included as sustainable development goal- in its vision.

As energy becomes more accessible to all, monitoring and management of energy consumption in buildings will assume even greater significance. ‘Smart metering’ is an important component of ‘Smart City infrastructure & services and has been recommended as an action required to gain a fuller understanding of energy consumption patterns in residential buildings by Rawal and Shukla (2014).

Possible Implications for Architects

As stated earlier, compliance with ECBC regulations is presently not mandatory for a large part of the current building stock. However, in the near future that may change and Energy Conservation measures may be incorporated in the local building bye-laws and their compliance enforced through local bodies like Municipal Corporations as part of their duties. That would require the architects to be aware about these measures and also become more familiar with the tools and techniques available for evaluating building performance. They may be required to perform compliance related to self certification on the same lines as that for compliance with the building bye-laws. AI based tools shall provide much needed support for early design decision making for Architects and stake-holders to make comparative assessments of various design and material options in terms of initial cost and operational costs.

Once the buildings have been constructed and operational, the monitoring of their energy performance in real time shall be done to compare performance against bench-marks for the respective categories. The bench marks, which might be adapted initially from energy related codes like ECBC, will be regularly updated as per the ground situation using energy consumption models based on AI techniques to assess trends and patterns in real time data obtained from ‘smart meters’.

AI to Revolutionise Energy Efficiency 

Technologies like AI/Machine learning and data mining using Big Data on energy consumption have the potential to revolutionise energy efficiency of buildings and ultimately the way buildings are being designed in India. These smart systems could be integrated together in a smart city environment to assess the demands at the city or regional level to improve the overall management of energy related resources, thereby making the entire process more efficient and environmentally sustainable. And in this way, these techniques can contribute in realising the border vision of achieving a sustainable growth in India.




latest news