Imagine that when you walk into a house, the light suddenly lights up, all kinds of furnishings in the house interact with you as if they were conscious, and all kinds of electrical appliances and machinery start magically… If you encounter such a house in your childhood, I’m afraid people will think it’s haunted. But now, this place is not haunted at all, so there is no need to bother the Ghost Busters.
In fact, controlling objects in space is a perfectly logical thing. As engineers, we are well aware of the development of artificial intelligence (AI) in recent years. Artificial intelligence is a technology that enables computers and machines to imitate human behavior. An important part of it is machine learning (ML), which combines artificial intelligence technology with statistical processing technology to continuously improve computers and machines in the process of gradually moving towards their goals, so as to achieve the purpose of “learning”.
Artificial intelligence affects almost every field and application, and the cities and spaces we occupy are no exception. In the smart city, intelligent sensor and processing network, together with artificial intelligence and machine learning technology, will be committed to improving our living environment, family, work and entertainment environment. Smart city will affect the lighting infrastructure, buildings, utilities, transportation, environment and communication throughout the city, which is expected to change the way we interact with the surrounding world.
Some forecasts indicate that lighting accounts for one sixth to one fifth (1 / 6-1 / 5) of the current total energy consumption. With the advent of AI, building lighting can be controlled and automated to save costs, reduce energy consumption and waste, and improve service quality and customer satisfaction. Artificial intelligence is like an invisible intelligence worker, providing help for various decisions and making the future intelligent building a reality. So, how does invisible AI control the visible lighting environment? Let’s break it down (Figure 1).
Figure 1: artificial intelligence will affect the future of intelligent lighting control and automation
Introduction to intelligent lighting
Intelligent lighting system integrates communication and control systems, which can bring greater automation and flexibility. In addition, wireless communication function can be added to make the whole intelligent lighting system cover a longer distance. They all help to improve the following three aspects:
·Overall (macro level)
·Edge (local level)
·Specific (device level)
After these problems are improved, it will help to improve the overall response speed and efficiency of the lighting system.
Smart phones, computer systems or wall mounted devices can be used as control stations and switching stations for lighting systems. By adjusting the combination of red, green, blue and white light, the color or white level of light can be adjusted to provide the required specific wavelength and related color temperature (CCT). By adjusting the output light level, the optical power emitted from each position can be controlled. Ordinary lamps, embedded lamps, building (indoor or outdoor) lighting, signs and landscape lighting can be coordinated in one system.
Machine learning makes lighting better
If I want to learn all kinds of knowledge, I have to study for several years. Some things, such as the basic grammar of speaking, I mastered when I was very young, but more in-depth contents, such as quantum physics and dealing with Laplace transform like arithmetic addition, need me to study for many years.
Artificial intelligence is an important disruptive technological innovation. It creatively brings learning ability to intelligent lighting system to improve its performance like feedback in electronic circuit. This learning and improvement function is machine learning.
Machine learning often requires the use of large amounts of data. When analyzing data, the computer needs to make its own decisions. These decisions are called “inference”, which is a conclusion based on evidence and logical reasoning. This type of processing is well suited to be performed by a computer.
Computer systems can be learned in three ways:
Supervised learning is carried out by providing the best correct answer (output) required and comparing it; In contrast, unsupervised learning as a supplementary means does not provide any information about the best correct answer (output) required. Reinforcement learning provides appropriate positive or negative feedback based on the best correct answer (output) that should be obtained. Because computers have strong computing power, compared with humans without the help of computers, computers can often achieve extremely rapid improvement in reinforcement learning (Fig. 2).
Figure 2: the future intelligent lighting system will imitate human behavior and learn how to operate independently with the help of artificial intelligence
Application of artificial intelligence in lighting system
Nowadays, all walks of life are actively adopting artificial intelligence, among which banking, retail, automobile and medical treatment are the first industries to adopt artificial intelligence in their respective fields. Obviously, although artificial intelligence will be everywhere, its pace of adoption in different industries is not the same. Over time, the knowledge and lessons learned in the above areas will benefit the industrial application field.
The field of industrial control (including intelligent lighting) covers a huge breadth and scope. If an organization has a unique understanding of intelligent lighting control and automation parameters, it will adopt artificial intelligence faster than those organizations that entrust this responsibility to external companies. It will be easier for organizations that know how to solve challenges and achieve goals through learning algorithms from the beginning. Understanding the limitations and interrelationships of existing systems will undoubtedly help to identify specific areas that need to focus on and apply artificial intelligence in building lighting control and automation solutions. By customizing artificial intelligence, organizations can solve the problems of control and automation in specific application fields. Artificial intelligence is a versatile tool, just like a handyman with many sophisticated tools, which can deal with a variety of applicable environments and applications.
There are many activities in the industrial field, and common advanced functions with high return on investment (ROI) will become the main market entry point. The first areas where artificial intelligence is widely used are likely to be areas that need to invest a lot of money in personal safety, overall security and risk prevention. Secondly, for industrial applications that can quickly adapt to various uses with a little modification, it is also natural to apply artificial intelligence. There is no doubt that any organization should immediately study and formulate strategies on how artificial intelligence can improve efficiency.
A fact that is difficult to explain but no doubt is that even if each specific operation changes, or will be affected by different response methods and subsequent operations, people usually do things in a characteristic predictable way. Psychologists and sociologists often earn their income by learning people’s reactions and behaviors in various situations.
Many organizations acknowledge that they are trying to integrate artificial intelligence and machine learning technology into their processes in order to better understand what is happening. One advantage of artificial intelligence is that it will help humans analyze their own behavior and its relationship with other dynamic events.
After adopting artificial intelligence and machine learning technology, organizations can learn and respond to various changing conditions with the help of data generated by customer habits. The past statistical data model can be compared with the current operation to adjust and optimize (Figure 3).
Figure 3: AI enables organizations to perform behavior analysis by analyzing big data and customer behavior
The demand for building lighting control and automation is constantly changing, which is a dynamic trade-off process. If you want to use the data generated in the past and simulate the potential expected demand consumption, this process may benefit from the artificial intelligence enhancement model. If organizations want to improve productivity and reduce costs as much as possible, they will be eager to try artificial intelligence, hoping that artificial intelligence can bring economic advantages. The data generated in the past is extremely complex, but if it can be combined with the predicted model, it can meet the needs of today in real time and flexibly. Because artificial intelligence will learn by itself, it will continue to improve its performance. The sooner artificial intelligence is adopted and incorporated into the building lighting control and automation process, the sooner it can achieve excellent decision-making and response level. In this process, a high degree of flexibility runs through, because AI has learned how to adjust the lighting control and automation variables of buildings according to the deviation between current performance and optimal performance. Finally, the overreaction of the system will be alleviated by using artificial intelligence.
In addition, artificial intelligence can correctly deal with great changes in usage. Since the system has learned how to respond from various conditions, when there is any significant increase or decrease in each input, the system can obtain an approximate correct response with the help of the first-order approximation, and then make further adjustment according to the enhanced feedback. Finally, the artificial intelligence system provides a controllable, self correcting and automatic decision-making tool to control the lighting of buildings.
Office lighting automation
Artificial intelligence can help all kinds of things realize automation, which means that it can make the system operate with little or no direct manual supervision or control, so it can better adjust the lighting control and automation of buildings, so as to save costs, reduce energy consumption and waste, and improve service quality and customer satisfaction. In the office lighting environment, because the position of the sun is always changing throughout the day, a major application example of artificial intelligence is to synchronously and automatically control the building lighting according to the indoor measured sunlight intensity, and adjust it according to the different positions of multiple customers and the requirements of output light intensity.
For some people, new technological progress seems as terrible as haunting, but I don’t think so. We don’t need to worry too much about artificial intelligence. With the advent of the era of artificial intelligence, this technology will eventually be applied to the automatic control of building lighting, which will help us move in a positive direction, so as to save expenditure, reduce energy consumption and waste, improve service quality and improve customer satisfaction. This is not a ghost story imagined out of thin air.