UC Berkeley

Data-driven techniques that extract insights from sensor data reduce the cost of improving system energy performance through fault detection and system health monitoring. To lower cost barriers to widespread deployment, a methodology is proposed that takes advantage of existing sensor data, encodes expert knowledge about the application system, and applies statistical and mathematical methods to reduce the time required for manual configurations.

Renewable energy technologies as well as building energy management systems have upwards of hundreds of existing sensor data points used for control and monitoring. Furthermore, innovations in "Internet of Things" (IoT) devices have further led to connected power meters, lights, occupancy sensors, and appliances that are capable of data collection and communication. This data presents a valuable opportunity to extract meaningful information and take data-driven action.

The motivation for transforming data from these devices into actionable information is to improve operations, monitor system health, increase energy generation, and decrease energy waste. The development and widespread use of energy conservation and renewable energy technologies are critical to minimizing negative environmental consequences. To that end, increasing profitability for users and decreasing costs of these technologies enables market penetration and widespread adoption. On the energy generation side, operations and maintenance accounts for up to 30% of the cost of wind generation, and unexpected failures on a wind turbine can be extremely expensive. On the energy demand side, commercial buildings consume 19% of US primary energy. Of this, an estimated 15\% to 30% of energy used in commercial buildings is wasted by poorly maintained, degraded, and improperly controlled equipment.

However, one cannot achieve scalable deployments of analytics and applications across systems if deploying solutions requires vendors and domain experts to install sensors and information technology infrastructures that require tailoring each solution for each deployment. Today, even well-established commercial offerings are not deployed at scale because costs are prohibitive. Thus, a major challenge to scalability is reducing hardware and software installation costs, manual configuration requirements, and manual monitoring.

To address this challenge, a methodology is proposed that leverages machine learning techniques to configure automated fault detection systems and controls. The approach combines sensor data points and encodes engineering knowledge that is generic to the application system but independent of a particular deployment. The resulting data can be input into numerous machine learning and optimization algorithms. Furthermore, the procedure selects data points and demonstrates that only a small number of sensors are necessary for fault detection with high accuracy rates. Applications to a wind turbine, a commercial building chiller plant, and residential buildings demonstrate the proposed methodology. Implementation is possible and the results are realizable using off-the-shelf algorithms, libraries, and tools. The goal is to enable an application that can be written once and then widely deployed with little additional cost or effort. The results of analysis can also inform policy decisions for stakeholders.