Primary tab navigation

Green Horizon

Driving Sustainable Development



This page contains also video presentation which
requires a newer version of Adobe® Flash® Player.

1978 was the start of China’s economic reform. By then, 17.9% of people in China lived in cities. Fast forward to 2013, the urban share of the population has grown to 53.7% and the percentage is projected to reach 70% by 2030 with over 200 million people moving to the cities.  Consequently, urbanization is accelerating the demand for energy, housing, and transportation. Emissions from using coal as fuel, vehicles, and dust from construction sites have led to significant human costs and challenges — adverse environmental impact on air quality, public health safety, social instability, structural inefficiencies, and a growing reliance on energy imports, to name just a few. Moreover, air pollution has received significant global, national and local attention driven by media coverage, creating health concerns for the public and negative press coverage for the government. In response, the Chinese government has set aggressive targets for cleaner and more efficient energy and sustainable urbanization.

Green Horizon is a 10-year initiative launched by IBM in July 2014. It aims to empower China in transforming its national energy systems and support its needs for sustainable urbanization. Led by IBM Research – China, the initiative taps into IBM’s network of 12 global research labs to create an ecosystem of partners from across government, academia, industry and private enterprise. It covers three projects: Air Quality Management (AQM), Renewable Energy Integration, and Industrial Energy Efficiency.


Air Quality Management (AQM)

Air pollution has received significant national and local attention and has become one of the top concerns in China. It is also a matter of global importance. The composition of PM2.5 is relatively complex, including direct emissions of fine particles of the combustion process (primary PM2.5 particles), and secondary particles generated by multiphase chemical reactions of atmospheric pollutants. Air quality modeling is a very complicated process that involves various chemical and physical models, and has temporal & spatial representative problems. The average ambient PM2.5 levels in China is higher than the World Health Organization standards, which is set at 10 µg/m3. Beijing is 9x the WHO standard and at times the local index can spike up to 30-50x the standard.

Based on IBM’s unique technologies such as data assimilation and cognitive modeling, we have developed a comprehensive air quality management solution which is comprised of high-resolution air quality forecasts, emission source identification and traceability, and quantitative policy decision support. This solution enables the decision makers to make effective and timely decisions to enforce laws and regulations, and foster environment management operational excellence. The key technology differentiations of this solution include:

  1. Data Assimilation: Model forecast error often results from uncertainties in initial condition. We leverage data assimilation to combine at best different sources including surface monitoring data, weather data, emission data, satellite data and geographical data, to estimate the initial state of a model (i.e., initial condition), and then get high-accuracy air quality forecasting.
  2. Cognitive modeling based on physical-statistics integration: Air quality modeling has temporal & spatial representative problems. We utilizes multiple models including WRF-CHEM, WRF, CMAQ, CAMx, etc. As each model performs best in different conditions (i.e. temperature, wind speed/strength, geographic ), the system uses adaptive machine learning mechanism to train those models and adaptively adjust the parameters for each model, and selects the optimized one with best performance for each specific situation.
  3. Combined source identification technologies: We combined sensitivity modeling, inverse/joint method, portable sensor and social media method to get the comprehensive estimation on the pollution source identification, which could address the requirements on law enforcement and long-term emission control strategy making.

Starting from air quality management, Green Horizon initiative also covers other two topics including renewable energy integration and industrial energy conservation. As air pollution is a very comprehensive problem, the key to tackling environmental problems is not only monitoring emissions but adopting a “system-in-a-system” approach to air quality management and addressing the issues at their roots.


Renewable Energy Integration

Replacing fossil-fuel based energy generation with renewables (e.g. wind, solar) can substantially reduce emission and help sustainable development. Wind and/or solar-dependent energy is generated intermittently, at random, and barely controllable. Generators need methods to predict outputs to increase capacity utilization level. In China, as of 2013, there was over 92GW of installed wind power. As per estimates, 25~40% installed capacity is not utilized. This presents a huge opportunity for optimization.

IBM Hybrid Renewable Energy Forecasting (HyREF) Solution combines weather prediction and Big Data analytics to accurately forecast the availability of renewable energy with high variability. This technology enables utility companies to forecast the amount of energy which will be available to be redirected into the grid or stored - helping to ensure that as little as possible is wasted.

The system has already been rolled out to 30 wind, solar and hydro power sources. The biggest deployment is at China’s largest renewable energy initiative - the Zhangbei Demonstration Project managed by State Grid Jibei Electricity Power Company Limited (SG-JBEPC) in the Northern province of Hebei. Using the system, SG-JBEPC is able to integrate 10% more alternative energy (enough for 14,000 homes) into the national grid. With a prediction accuracy of 90% proven on Zhangbei’s wind turbines, it is one of the most accurate energy forecasting systems in the world.


Energy Optimization for Industry

The Chinese government has committed to reduce the country’s "carbon intensity" by 40-45% by the year 2020. This will bring tremendous industry changes to China in comparison with the country’s 2005 carbon levels (equivalent to 130 million tons of coal per year).

To support these goals, IBM is developing a new system to help monitor, manage and optimize the energy consumption of industrial enterprises that is adding up to over 70% of China’s total energy consumption. Using a Big Data and analytics platform deployed over the cloud, it will analyze vast amounts of data generated by energy monitoring devices and identify opportunities for conservation. For example, it could be used to analyze data from a steel factory and identify that an oxygen furnace wastes the most energy because the temperature of the output smoke is too high. The system will be a valuable tool for guiding decisions about optimization and investment in China’s most power hungry industries such as steel, cement, chemical and non-ferrous metal.


“Green Technology Innovation Excellent Practice” Award

“Green Technology Innovation Excellent Practice” Award

On November 2014, Green Horizon project was awarded “Green Technology Innovation Excellent Practice” by China Business News (CBN is one of most influential business media in China with substantial Chinese viewers including business decision makers, influencers and financial professionals).

It is the first time that China Business News issues Green Technology Innovation Excellent Practice. This is to recognize the achievements of Green Horizon project aiming at green innovation and sustainable growth.

Media Coverage