From toothbrushes to automobiles, earbuds to mobile devices, and toys to semi-trucks, lithium-ion (Li-ion) battery technology is finding its way into just about everything. As the demand for battery-powered devices is ever-increasing, and more utilities, and commercial and industrial enterprises, turn to battery energy storage as a source of electricity, more battery technology is making its way into the supply chain, user base, and waste stream.

In recent years, battery storage, charging, and recycling centers have experienced increased fire activity caused by Li-ion battery thermal runaway. Li-ion batteries can experience thermal runaway and catch fire from manufacturing defects or external misuse like overcharging, overheating, puncturing, or being crushed.

Battery energy storage systems, warehouses that store batteries and battery-powered devices, charging stations, and recycling centers are finding ways to mitigate and prevent fire damage using early detection technologies like thermal imaging cameras. Thermal cameras “see” the formation of battery hotspots providing the earliest warning of a possible runaway condition. Connecting cameras via the Internet of Things (IoT) with cloud-based monitoring and notification software creates an early warning notification system, keeping the Li-ion battery stream running safely.

1. Both external and internal causes can initiate a thermal runaway event in a lithium-ion battery. Courtesy: MoviTHERM 

Thermal runaway occurs when the temperature of the Li-ion battery reaches a critical state such that the reaction rate of an exothermic reaction increases the temperature, which in turn leads to further acceleration of the reaction rate. Figure 1 summarizes the external and internal initiation events that trigger thermal runaway in a Li-ion cell.

Internal causes of spontaneous ignition include coating defects at the electrode surface, contamination particles, and poor welds. Typically, these defects cause electrical shorts during operation that generate heat.

External causes include electrical abuse from overcharging, mechanical abuse via crushing or puncture, and thermal abuse from exposure to high-temperature environments. External initiating events are related to each other. For example, mechanical abuse from a puncture of the Li-ion cell causes a short circuit, which is electrical abuse. The electrical abuse creates heating, which increases the cell temperature, causing thermal abuse, which can trigger thermal runaway.

Fire Detection Devices

Various fire detection sensors are available today that alert of fire formation with varying detection timing during the progression of a fire. Figure 2 shows the relative detectability of fire detection devices at different stages of fire development with corresponding damage levels.

2. Different stages of fire development can cause varying levels of damage to the surrounding area. Courtesy: MoviTHERM 

Early fire detection (EFD) and infrared (IR) camera systems operate on the heat transfer principle of radiation. The IR camera has a focal plane array of detector elements that sense infrared light from object surfaces. The radiation captured by the IR camera detector is digitized, converted to data, and displayed as a viewable image (Figure 3).

3. Here’s an example of an infrared (IR) image with spot and area regions of interest. The white areas of the motorcycle show hot areas. Courtesy: MoviTHERM 

Calibrated IR cameras can report temperature measurements from specific spots, lines, and areas on live or recorded images. IR cameras are available in different wavebands, pixel resolutions, lens configurations, and communication protocols to meet various installation requirements.

IR camera systems are the first to alert before a fire develops. They “see” a warming of material early in the fire development process before forming smoke particles or flames. These warming materials appear as hot spots in a thermal image and are quantified with regions of interest (ROIs) like spots, lines, or areas that report temperature values. Applying multiple ROIs to an image and setting temperature thresholds per ROI allows monitoring and alarming multiple locations within the camera’s field of view. When the threshold condition of an ROI is satisfied, alarms trigger notifications to the appropriate personnel.

What Is IoT?

The Internet of Things (IoT) refers to interconnected sensors, instruments, and other devices networked into software applications that use predictive analytics and artificial intelligence (AI). These connected networks create systems that can monitor, collect, exchange, analyze, and deliver valuable insights into a system or process. IoT revolutionizes automation by using cloud computing to simplify integration and enhance process control.

IoT works with thermal imaging and EFD to improve safety and reduce fire risk for battery storage, charging, and handling. By connecting IR cameras that alert at the earliest stages of fire development, potential fires can more readily be detected and prevented. Safety alerts are sent to hundreds of people quickly and effectively with IoT. Communication options include voice calls, texts, and emails to targeted recipients to establish quick and effective awareness.

Another advantage to cloud-based EFD is scalability. Facility managers can connect multiple facilities into a central monitoring and alarming dashboard. Understanding the situation at all facilities improves the oversight and management of multiple systems from a single control point.

IoT-based EFD systems can improve emergency planning by using algorithms and analytics to help prepare better emergency and evacuation plans quickly. For example, analytics can consider factors such as the number of people in the facility, facility maps, location of the fire, and the rate at which the fire is spreading to develop better evacuation plans. Analytics-based evacuation plans can prevent congestion by guiding workers to different locations for optimum evacuation routing.

IoT EFD systems are less expensive to install and maintain than traditional detection systems. As the EFD software resides in the cloud, there is no need for a dedicated facility computer server. Any potential for operating system software conflicts is eliminated as access to the cloud-based application only requires an internet connection. Users access the EFD system anywhere and anytime with any internet-connected device. And with the appropriate credentials, control and alarm settings can be modified remotely to optimize performance.

4. This image shows a sample map view display from a cloud-based Internet of Things (IoT) early fire detection (EFD) program. Courtesy: MoviTHERM 

Another key advantage to a cloud-based EFD system is the ability to share dashboards and map views (Figure 4). For example, sharing a live map view with first responders allows for scene assessment before arriving on-site, saving time and optimizing safety. These map views identify the alarm sensor location, monitored area, alarm conditions, facility entry, and exit points.

IR Camera IoT Early Fire Detection for Battery Monitoring

IR camera IoT EFD systems for battery monitoring can integrate multiple detection technologies to track temperatures and detect smoke particles at critical locations. The most common detection sensors for battery monitoring EFD include:

■ IR cameras for quantitative and qualitative monitoring of hot spots.

■ Visible cameras for identification of smoke or flame.

■ Aspirating smoke detectors for detection of smoke particles.

5. Here’s an example of an IR camera IoT EFD configuration for battery monitoring, showing the various connections to optimize the system’s capability. Courtesy: MoviTHERM 

Correct sensor selection and placement for battery monitoring (Figure 5) are critical to ensure optimum detection performance. For example, IR cameras require a direct line of sight to the area of interest to provide detection. Critical areas obscured from the camera’s field of view could be monitored by smoke detectors, thereby augmenting the camera’s detection. For outdoor or high airflow installations, IR sensors are best for detection as dilution effects may limit the performance of smoke detectors.

It’s important to note that IR camera IoT EFD systems do not replace existing detection and response protocols. Instead, the system functions as an early warning system—detecting areas in the facility where ignition may occur. New detection methods for heat, smoke, and fire are continually developing. Many new detection devices include wireless capabilities that make integrating IoT EFD a straightforward exercise.

Beyond alarms and notifications, IoT EFD systems can provide automation controls like initiating and directing an extinguishing system. Because IoT EFD systems leverage cloud computing, they require less hardware with a reduced installation burden. Available communication technology can be added to existing detectors, making IoT retrofitting existing systems easy. By warning earlier on the pathway to ignition, managers of the battery chain avert costly and potentially life-threatening fires before they are permitted to start and spread.

David C Bursell is vice president of business development at MoviTHERM, a company that provides thermal imaging solutions for remote monitoring, automation, and non-destructive testing.