The next transformation in HVAC control in large buildings is AI

Dr. Amit Chaudhari, Associate Director, KPM Design Services Pvt. Ltd., deliberates that HVAC upgrades that work better with AI are airside economizers and energy recovery ventilators.

 

HVAC systems are the largest energyconsuming loads in residential and commercial buildings. For this reason, energy efficiency measures that target HVAC can achieve significant savings. However, HVAC upgrades that save energy should not affect indoor environmental quality, or they can cause discomfort and health issues. In business settings, this also hurts productivity.

Artificial intelligence has promising applications in HVAC since it can improve energy efficiency and indoor environmental quality. AI can optimize variables like airflow, temperature and humidity – conserving air quality with the lowest possible energy consumption.

 

HVAC Control: A Technical Challenge 

Controlling an HVAC system effectively is difficult because building conditions are constantly changing. Hence, considering the following aspects is essential. The outdoor temperature continuously changes, affecting the heating or cooling required to keep a suitable indoor temperature. Another vital point is that people are constantly entering and leaving buildings, affecting ventilation needs and temperature control. Further, the activities performed indoors also affect HVAC needs. For example, a commercial kitchen needs more ventilation and cooling than an office of the same size.

The ventilation, heating and cooling needs of a building are constantly changing. Only an intelligent control system can process all this data in real-time and adjust the HVAC system accordingly. An HVAC system with manual controls cannot reach peak performance due to variable working conditions.

 

Smart Controls for ventilation systems 

Air handlers use less energy than air conditioners and space heaters. However, the amount of outdoor airflow a building provides impacts its heating and cooling requirements. For example, if a ventilation system increases airflow during the winter, more heat is required to heat a larger volume of cold outdoor air.

The same applies to air conditioning when the ventilation system increases airflow during the summer. Many buildings have ventilation systems that run at total capacity all of the time. For the reasons stated previously, this represents a significant waste of energy. When ventilation controls are AI-enabled, they can determine the optimal airflow required by the building. The system can also monitor the number of occupants in the building and the concentration of key air pollutants such as VOCs and particulate matter.

Over-ventilation represents a waste of energy, while under-ventilation is detrimental to indoor air quality. Ventilation control with AI helps prevent both of them. There is only one crucial requirement: the ventilation system must never reduce airflow below the minimum value required by local building codes.

 

Optimizing space heating and air conditioning 

There are now furnaces, boilers, chillers, mini-splits and other HVAC equipment with built-in artificial intelligence. As a result, significant energy savings can be expected simply by upgrading old equipment. However, improved performance is possible when intelligent controls are used for the entire HVAC system.

Heating and cooling controls, like ventilation controls, must balance energy consumption and air quality. According to a study conducted by Lawrence Berkeley National Laboratory (LBNL), human productivity peaks at around 21°C – 22°C, and as the temperature rises or falls, it causes discomfort. When the temperature rises to dangerous levels, it hurts health and can be fatal. Extreme temperatures are unlikely in building interiors, even with a faulty HVAC installation. However, the indoor environmental quality (IEQ) worsens gradually as the temperature deviates from the ideal range.

 

AI implementation with energy efficiency measures 

Energy efficiency measures can become more effective when AI is added to the mix, achieving more significant savings. Two HVAC upgrades that work better with AI are airside economizers and energy recovery ventilators.

 

Airside economizers can save plenty of energy in some climates. When outdoor air has a suitable temperature and humidity for “free cooling”, the economizer increases ventilation rates while reducing the air conditioning output. Electricity is saved because fans are less expensive to operate than air conditioners. Intelligent controls can optimize the airside economizer to maximize these savings.

 

Energy recovery ventilation also achieves synergy with intelligent ventilation. When the outdoor airflow is optimized, extra air is avoided by heating or cooling energy waste. The ERV system can then exchange heat between the supply air and the exhaust air, reducing the HVAC workload even more. ERV works with air conditioning and space heating: indoor air precools outdoor air in summer and preheats outdoor air in winter. Building certifications such as LEED and WELL have demanding heating, cooling and ventilation performance requirements. Automatic controls with AI can help meet these requirements, optimizing the operation of HVAC systems.

 

Advantages of implementing AI technology in HVAC systems for building controls 

The essence of Artificial Intelligence is for thermal comfort requirements. An orchestration of intelligent entities (aka. rational agents) can be used to map the decisions made by AI to the physical environment. A rational agent has four defining aspects: a performance measure, the environment, actuators, and sensors (PEAS). The performance measure is the criterion that measures the agent’s success, the environment dictates the restrictive factors an agent may encounter in its operating space, and the actuators and the sensors work together continuously, as sensors process information and relay information to the actuator to carry out the physical changes in the environment. For example, in building environments, thermostats and HVAC systems are traditionally interlinked to accomplish the same goal. 

While thermostats only detect temperature changes, HVAC systems provide heating and cooling, but despite their different immediate primary functions, both are geared at maintaining comfortable indoor temperatures. Rational agents are used to carrying out the actions dictated by artificially intelligent systems. To function effectively, components that control the agent’s problem-solving process, rational decision-making capability, and the agent’s ability to learn must all function correctly. In computer science, these three components can be generalized by the following concepts: Search Algorithms, Logic

Inference, and Machine Learning. To understand this paper, a brief explanation of only these three subsections of AI is explored. Implementation examples from the literature accompany each subsection.

 

AI techniques enhance the control of HVAC systems for occupant comfort 

HVAC systems are the enormous energy-consuming loads in residential and commercial buildings. For this reason, energy efficiency measures that target HVAC can achieve significant savings. However, HVAC upgrades that save energy should not affect indoor environmental quality, or they can cause discomfort and health issues. In business settings, this also hurts productivity. Artificial intelligence has promising applications in HVAC since it can improve energy efficiency and indoor environmental quality. AI can optimize variables like airflow, temperature and humidity – conserving air quality with the lowest possible energy consumption.

AI is always awake to take these inputs into account and adjust settings. Thanks to it, the HVAC system monitors multiple parameters, specifies set points, turns units on/off, and performs several other actions. All this happens autonomously, and no energy efficiency manager is needed. When any environmental parameter changes: weather, occupancy, energy consumption, and so on, the HVAC control regulates itself right away.