TY - RPRT
TI - Triton: Environmental Monitoring Technology Development, Collision Risk Data Collection
AU - Packer, A
AU - Acker, T
AU - Staines, G
AB - Current energy converters like tidal and in-river turbines are a promising technology for clean electrical power generation in a variety of locations in the US. The environmental risk of most concern in relation to these turbines is collisions between animals such as fish or marine mammals with moving parts of the turbines. Monitoring technologies like video cameras and active acoustic sensors can collect data about animal interactions with turbines that regulators and stakeholders can use to assess this risk. Video and acoustic cameras can generate compelling images, allow identification and classification of targets, and give insight into the behavior of animals near a turbine. However, analyzing data from these sensors, particularly from long-term continuous monitoring, is time consuming and costly. The goal of this project was to develop software tools that will reduce the time and cost of analyzing data from these sensors. In this project, we assembled a sensor suite that combines a scientific echosounder (sonar) system with video and acoustic cameras (secondary sensors). The sensor suite generates data that is amenable to automated target detection algorithms and can provide inputs to animal encounter models. We developed software (archiving software) to collect data simultaneously from all sensors in the suite, analyze the sonar data automatically in near-real-time to identify time periods when targets of interest were present, and automatically archive data from the secondary sensors for these time periods. The product of the archiving software is a data set for the secondary sensors, containing data only for the times when targets of interest were determined to be present by the sonar. We performed controlled field testing to verify the operation of the sensor suite and the archiving software. In real-world applications we expect that the secondary sensor data set produced by the archiving software will be significantly smaller than the full continuous data set for those sensors and will therefore be faster and cheaper to analyze. The exact reduction in data accumulation, and the associated time and cost savings, will be specific to each site, determined by such factors as number of animals present and water column dynamics like entrained air or debris drift. Further work is required to quantify the reduction that can be achieved at sites of interest. The sensor suite used for this project consists entirely of sensors that are available commercially from established manufacturers. The software developed for this project leverages proven commercial off-the-shelf software applications and software development kits. It is designed to be robust, for long-term, continuous, unattended monitoring deployments. The software allows for the sensors in the suite to be collocated or not and is extensively configurable to allow use near different sized turbines, in various environments (rivers, dams, marine) where the targets of interest may be of different types, such as large individual animals, or dense groups of small animals. At the conclusion of this project, the sensor suite and software are at Technology Readiness Level 4. The software works reliably as designed, and successfully archives data from video and acoustic cameras when the sonar detects targets of interest. Our recommendation for the next step is to deploy and beta-test the system near an operational turbine, to demonstrate that the system can reliably collect useful monitoring data in an operational environment and to quantify the reduction in secondary sensor data volume that can be achieved at a realistic site.
CY - Richland, WA
DA - 2022/09//
PY - 2022
SP - 62
PB - Pacific Northwest National Laboratory (PNNL)
SN - PNNL-33475
UR - https://www.pnnl.gov/main/publications/external/technical_reports/PNNL-33475.pdf
LA - English
KW - Marine Energy
KW - Tidal
KW - Collision
KW - Fish
KW - Marine Mammals
ER -