A new tool developed by Lawrence Berkeley National Laboratory in the United States can help automate fault detection and diagnostic software, minimizing the need for human-computer interaction, thereby increasing efficiency and reducing carbon emissions.
Today, building automation and energy management systems are becoming increasingly necessary in facilities management, which has a direct impact on building operations as graphs enable owners and operators Enable greater efficiency, flexibility and resilience in the face of climate change. But with these sophisticated tools comes increased complexity and the introduction of errors, often at the expense of the efficiencies these technologies provide.
As a result, Building Fault Detection and Diagnostics (FDD) technology is growing in popularity, saving property owners millions of dollars in building costs each year, with a payback period of typically less than two years. FDD tools automate the process of detecting HVAC system failures and sub-optimal performance to help diagnose potential causes. FDD typically sits on top of existing building automation systems (BAS), according to a February 2022 report from Lawrence Berkeley National Laboratory (LBNL) in Berkeley, California.
However, while commercial FDD tools appear to be a panacea for improving energy efficiency and thus reducing carbon emissions, there is still a small problem: a human solution is required. The LBNL report states that "once a fault is detected, human intervention is required to repair the fault, which often results in delays or even inaction, resulting in additional operating and maintenance costs and impact on comfort conditions within the building."
In other words, a building’s efficiency, energy savings and carbon footprint still depend heavily on people.
According to LBNL, automated fault correction for commercial FDD applications shows great promise in closing the loop between passive diagnostics and proactive control. In some cases, these tools can integrate artificial intelligence (AI) for predictive maintenance, giving facility managers more flexibility and freedom than ever before.
According to statistics, buildings use 70% of the electricity in the United States, account for nearly 33% of global carbon emissions from fuel combustion, and account for approximately 10% of total greenhouse gas emissions. 20% of the amount. Therefore, buildings must become increasingly efficient and predict problems with their systems before they occur.
However, FDD tools are not foolproof. In fact, studies estimate that legacy equipment failures and control issues can significantly increase greenhouse gas emissions and energy bills to the tune of $17 billion and 90 million tons of CO equivalent annually, according to LBNL and the U.S. Department of Energy (DOE).
“It turns out that the most energy-impacting opportunities we encounter most often can be addressed through automated fault correction and control optimization,” LBNL said.
These opportunities to improve energy performance include:
"We are now working to extend our suite of best-in-class trouble-free control solutions to a wider range of FDD partners and include additional strategies such as automated commissioning/functional testing and requirements flexibility, "Granderson said.
In 2016, LBNL launched the Smart Energy Analytics movement in partnership with the U.S. Department of Energy and various industry partners. This is a public-private partnership that has produced the largest data set on building analysis, costs, benefits and usage. In the years since, LBNL has also partnered with leading domestic FD technology market vendors to expand state-of-the-art technology beyond what was previously available. Granderson said her team has developed and implemented additional programming capabilities to automatically correct faults once they are identified by the existing FDD software.
In a 2020 field study with two end-user partners, LBNL developed and deployed a set of seven fault correction algorithms for HVAC systems that used existing BAS vendor platforms in real-world Testing was carried out in the building. Variables corrected by the algorithm cover schedules, set points, sensor readings, commands, heating/cooling requests, and proportional, integral, derivative (PID) parameters.
Historically, FDD technology has been integrated with building automation systems to capture operational data for system and equipment operations in a “read-only” format. “The first thing we did was enhance the interface so that the FDD system could also ‘write’ commands back to the BAS,” Granderson explains.
The team then developed a library of engineering logic that defined how to solve various control-related problems by modifying control system parameters typically accessible through the BACnet protocol.
Finally, the team integrated the correction logic into the FDD platform and operator-facing user interface. Now, once the FDD system detects and diagnoses a fault, the operator is notified of the problem along with recommended corrective actions. After operator approval, corrective actions will be implemented and the fault resolved.
Granderson provided the following example: A zone temperature set point that is too aggressive may be flagged for operator attention and correction with the message "The cooling set point for this zone is 66 degrees, which is lower than recommended." Would you like to return the set point to the recommended 68 degrees?" With operator approval, the FDD system is able to write the revised 68 degrees Fahrenheit set point back to the zone controller through its interface to the BAS. Once this action is completed, the fault is resolved and the FDD system returns to problem detection and diagnosis.
In addition to fault correction, LBNL also extends FDD system capabilities to control optimization. First, it developed and tested a method to implement best-practice adjustments and responsive reset strategies for air handling unit static pressure and supply air temperature in accordance with ASHRAE Guide 36: High-Performance Operating Sequence for HVAC Systems. Among these solutions, LBNL's technology is suppressing "special" areas that experience increased energy use due to unmet heating or cooling needs. Granderson noted that while LBNL is not currently using AI in the fault correction methods it develops, some FDD vendors are using AI in certain parts of their technology stacks.
Building IQ, based in Sydney and Fargo, North Dakota, has launched what it calls an Outcome-Based Failure Detection (OFD) service, which combines artificial intelligence, energy analysis and human expertise to overcome many FDD services Shortcomings. “Outcome-based fault detection is a comprehensive solution that takes fault detection in a better and broader direction,” the company’s then-president and CEO Michael Nark said in a June 2018 press release express.
"It does this by embracing the critical role played by facility experts and augmenting it with machine learning and cutting-edge artificial intelligence. OFD works regardless of whether the data is good or bad, and leverages machine learning Shifts the burden of data analysis to the cloud. The result is that building operators don’t have to waste valuable time and resources searching through hundreds of daily fault tables. Instead, with OFD, operators can focus on what really needs to be fixed, their tenants and the bottom line."
Advantages of Automated FDD Systems
She said, Automated FDD technology can continuously analyze operational data to identify problems for building operators and energy managers, noting that “the benefits are substantial. Our work shows that organizations using FDD systems across their portfolio can save an average of 9% on investment.” The payback period is two years.” Adding automatic fault correction extends the benefits even further, she continues. Instead of waiting weeks or months for issues to be resolved, issues can be resolved within hours and valuable staff expertise can be put to work solving the toughest problems.
“In addition, the ability to write control commands back to the BAS also allows us to implement supervisory control optimizations,” she said. "Providing supervisory optimization control through an FDD system allows for scalable implementation across different years and brands of BAS without the need for expensive upgrades, whereas more traditional approaches may require direct modifications to BAS programming."
Based on automation and Artificial intelligence-powered BAS and BEMS solutions have been adopted in the commercial construction sector globally. For example, ABB’s Ability BE Sustainable with Efficiency AI currently manages more than 275 buildings totaling more than 100 million square feet. Collectively, these installations reduce CO2 emissions by more than 1 million metric tons per year, all by leveraging investments already made in building automation.
Good data is the foundation of building automation and management systems, and the more data that can be fed into energy management and information systems, the better. As FDD tools and automation software evolve, smart building implementation, scalability, and reliability will continue to improve—and building owners and facility managers looking to start this journey will have the tools at their disposal.
In October 2020, LBNL released an Application Showcase to help stakeholders understand how to get started, highlight best practices from Smart Energy Analytics event participants, and provide examples of innovation happening in the industry.
“We have tested these new capabilities in a number of buildings and BAS products,” Granderson said. “Results to date indicate that they can be scalable across different controllers, with modest additional development and implementation lift provided by FDD vendors. As these emerging technology capabilities are provided by their partners through our product features or modules, LBNL will be able to track incremental costs relative to traditional FDD systems.
“This is all very new and still maturing, but what’s exciting about this work is what it shows us about smart buildings. future. We are increasingly asking our buildings to become net-zero greenhouse gas emitters, integrate an increasing number of distributed energy resources, and provide healthy and comfortable indoor environments while harmonizing with renewable grids.
"The only way to achieve this at scale is to leverage the modern software-based infrastructure provided by FDD and other smart building software. It provides us with a channel to continuously 'drive' improved control and analytics solutions .”
The above is the detailed content of Can artificial intelligence or automation solve the problem of low energy efficiency in buildings?. For more information, please follow other related articles on the PHP Chinese website!
The difference between vivox100s and x100
What are the commonly used functions of informix?
Ethereum browser query digital currency
What is a Bitcoin wallet
The specific process of connecting to wifi in win7 system
What is the use of docker image?
How to install linux system
What are the production methods of html5 animation production?
![[Web front-end] Node.js quick start](https://img.php.cn/upload/course/000/000/067/662b5d34ba7c0227.png)
