With the twin threat of climate change and air pollution high on the global agenda, IBM has launched a new initiative which leverages cognitive computing and the Internet of Things (IoT) to enable city governments, utility companies, and industry to improve their relationship with the environment. The ambitious initiative is a hi-tech contribution to addressing some of Earth’s most pressing environmental problems which stand as a serious and complicated threat to both public and environmental health.
Due to rapid industrialization over the past two decades, Chinese cities are amongst those experiencing some of the world’s most pressing environmental challenges. Beijing’s pollution woes are particularly complex being part of a highly-industrialized region and surrounded by three smog-trapping mountain ranges. With China’s recent wave of investment into green innovation and a budding new generation of environmental scientists coming to the fore, IBM’s China Research lab launched the Green Horizons initiative in 2014 with a multipronged strategy to help cities better manage air pollution while at the same time enabling utilities to improve the viability of renewable energy.
Building on a successful partnership with the Beijing Environmental Protection Bureau and forging numerous collaborations in other cities across China, IBM is now taking the initiative global with new commercial deals and research engagements across four continents including those in India, South Africa, Japan, the UK and the USA.
WHAT IS COGNITIVE IoT?
Green Horizons uses some of IBM’s most advanced technologies, namely cognitive computing and Internet of Things, drawing on IBM’s deep experience in weather prediction and climate modeling to enable some of the world’s most accurate energy and environmental forecasting systems. With thousands of sensors across physical and man-made environments, the Internet of Things generates vast amounts of real-time data complex in scale, form and meaning. IoT therefore, at its heart is a data challenge. That’s where cognitive computing comes in. Traditional programmable computing has limitations in such a fast-paced digital world. Cognitive computing has no such limitations. Rather than being explicitly programmed, cognitive systems learn from interactions with us and their experiences with their environment. This enables them to keep pace with the volume, complexity, and unpredictability of information generated by the Internet of Things.
AIR QUALITY MANAGEMENT
IBM’s air quality management system has its roots in a collaboration with the Beijing Environmental Protection Bureau which is grappling with reducing air pollution in the broader Jing-Jin-Ji (JJJ) region - shorthand for Beijing-Tianjin-Hebei – one of the world’s most populous, industrialized and polluted regions. In order to achieve its goal of reducing levels of ultrafine Particulate Matter (PM 2.5) by 25% by 2017, the Beijing Environmental Protection Bureau needed a way to fine tune its clean air action plan while protecting sustained economic growth and jobs for over 100 million people in the JJJ region.
PM2.5 is a measure of particulate emissions including carbon, nitrogen, sulphur, and heavy metals. A clear blue sky has a PM2.5 reading of 0. A typical busy city might score between 40 and 60, and the threshold of lung irritation and sensitivity is around 100 – near heavy industrial activity. Anything above 200 is hazardous to public health and published figures for Beijing have in the past exceeded 600.
IBM’s researchers helped to develop a system capable of generating predictive models showing where pollution is coming from, where it will likely go, and what will be its potential effect. Not only that, but using scenario modelling, they came up with a way to create hypothetical ‘what if’ scenarios - enabling city officials to try out the effectiveness of different action plans to achieve a balance between environmental and economic concerns.
The system works by using IoT technologies to gather data from environmental and weather monitoring stations as well as meteorological satellites and traffic cameras. Machine learning technologies ingest and learn from the data, constantly self-configuring and automatically adjusting the predictive models to different seasons and topographies. The system is able to generate highly accurate pollution forecasts, down to the nearest kilometer, 72 hours ahead of time and as well as pollution trend forecasts up to 10 days into the future.
Armed with this information, city officials can calculate the most targeted, effective and sustainable responses. Measures include short term limitations on urban traffic and construction activity as well as long term improvements to industrial production and power generation - such as switching to cleaner energy sources and installing filtering systems. Spraying water into the atmosphere and issuing public health alerts are part of the action plan for the most severe predictions. Selective, temporary reductions in industrial activity are also considered for large scale events such as the 2022 Winter Olympics to be jointly hosted by Beijing and the northern city of Zhangjiakou.
“Our environmental engineers are working on a daily basis to tackle Beijing’s complex and challenging pollution problem and protect the health of citizens,” said Dawei Zhang, Director of Beijing’s Environmental Monitoring Center, a department of the BEPB. “Through our collaboration with IBM Research – China, we are delivering on our environmental commitments with the help of some of the most advanced technologies available. Over the past year we made good progress and the joint innovation with IBM is one of the key driving forces behind it.”
RENEWABLE ENERGY FORECASTING
Alongside the focus on air quality management, IBM’s Green Horizons initiative is supporting the global shift from fossil fuels to renewable energy – which is necessary to help achieve reductions in CO2 – the biggest cause of climate change.
IBM Research has developed a renewable energy forecasting system which combines climate modeling, IoT and cognitive computing to help utility companies predict how much available energy they will have ahead of time.
Solar farms use sky-facing cameras to monitor cloud movement and calculate their potential blocking impact on solar radiation. Wind turbines are fitted with sensors 80 meters above the ground to monitor wind speed, moisture and air pressure.
Assimilating that with weather forecasting data, the system is able to predict the performance of wind and solar energy farms with 90% accuracy several days ahead – equating to savings of thousands of megawatts of energy that could otherwise be lost. Such knowledge is key to integrating more renewable energy into the grid and creating an optimum balance between supply and demand in energy markets.
The system has already been deployed at numerous wind and solar energy farms around the world. One of the first to benefit was China’s largest renewable energy initiative, the Zhangbei Demonstration Project, which is now able to integrate 10% more alternative energy into the national grid. With its renewable energy forecasting systems, IBM is helping utilities to maximize their output of clean, abundant, renewal energy.
IBM is realizing on the promise of Green Horizons by delivering critical insight to its partners. In China, IBM has entered new clean engagements in Boading - one of China’s most polluted cities; the city of Zhangjiakou - host to the 2022 Winter Olympics which is working to improve air quality for the important outdoor sporting event; and Xinjiang Province in north-west China. In India, IBM has entered a research collaboration with the Delhi Dialogue Commission, under the Delhi government, to leverage Green Horizons technology to help calculate the most effective and sustainable strategies for tackling air pollution in India’s capital. In South Africa, IBM has forged an innovation partnership with the City of Johannesburg to model air pollution trends and quantify the effectiveness of the city’s intervention programs in support of Johannesburg’s air quality targets and sustainable development.
In the area of renewable energy, IBM has built on momentum in China with an agreement with China's largest wind power solution provider - Xinjiang Goldwind Science & Technology Co., Ltd to use IoT, cloud computing, big data analytics and other advanced technologies to drive innovation and transform Goldwind's business and technological models. In the UK, energy giant SSE is piloting IBM technology to help forecast power generation at its wind farms in Great Britain. The system is able to forecast energy for individual turbines and includes visualization tools to show expected performance several days ahead. In Japan, IBM is working with the Toyo Engineering Corporation and renewable energy company Setouchi Future Creations LLC on the Setouchi solar project – one of the largest in the country. IBM’s monitoring systems will help Setouchi manage and control energy from the plant's 890,000 solar panels. Through the United States Department of Energy’s SunShot initiative, IBM is making renewable energy forecasting technology available to government agencies, utilities and grid operators across the United States to support supply and demand planning.
The purpose of the Internet of Things is to strengthen our understanding of the places we live and work in and deepen our connection with the world. But without cognitive computing, the usefulness of this information would be limited by its own complexity and scale. That’s why we believe that cognitive computing is essential in realizing the true value of the Internet of Things. That’s why Green Horizons is important – it’s a powerful example of combining cognitive and IoT technologies to create man-machine decision support systems, capable of not only furthering our understanding of what’s going on in our cities and energy systems, and crucially, predicting what will happen next, but of improving humanity’s relationship with the environment and helping to tackle some of the biggest challenges of our generation.