An integrated monitoring system for disaster-causing organisms in the water intake areas of coastal nuclear power plants

Chao Li, Jian-ling Huo, Yu-ze Song, Lei Yang, Song-tang LiuFrontiers in Marine Science 2023 · DOI

This research paper, published in Frontiers in Marine Science, details the development and field testing of a seafloor-based monitoring system designed to protect coastal nuclear power plants from biological threats.

TL;DR

Researchers have developed IMSDCO, an integrated seafloor monitoring system that uses high-resolution optical imaging and hydrological sensors to provide real-time alerts for "disaster-causing organisms" (DCOs) like jellyfish and shrimp. Tested over six months at a Chinese nuclear plant, the system successfully identified threats and predicted their movement, offering a high-tech alternative to slow, manual water sampling.

What Problem Does This Paper Address?

Coastal nuclear power plants rely on massive volumes of seawater for cooling. However, sudden "blooms" of marine life—ranging from microscopic algae (Phaeocystis globosa) to swarms of small shrimp (Acetes chinensis) and jellyfish—can clog intake screens. These blockages cause unit shutdowns, massive economic losses, and potential safety hazards.

Traditionally, plants monitor these threats through manual netting and lab analysis, which is infrequent and provides no real-time warning. While sonar is sometimes used, it struggles to identify specific species and is often "blinded" by the noisy, shallow-water environments near intake structures. There is a critical need for an automated, in-situ system that can "see" what is in the water and predict where it is heading.

Methods

The authors proposed the Integrated Monitoring System for Disaster-Causing Organisms (IMSDCO). The system consists of a seafloor node connected to a shore station via a photoelectric composite cable, providing both high-speed data (1 Gbps) and stable power.

Key technical components include:

  • Optical Microscopic Imager (OMI): An IPP05 sensor capable of identifying particles between 500 µm and 20 mm with an automated zoom and a recognition rate of over 80%.
  • Hydrological Suite: An Acoustic Doppler Current Profiler (ADCP) for current tracking, a CTD sensor (Conductivity, Temperature, Depth), and multi-parameter water quality detectors.
  • SCHISM Modeling: Hydrological data is fed into the Semi-implicit Cross-scale Hydroscience Integrated System Model to predict the 3D migration tracks of detected organisms.
  • Robust Engineering: To survive the seafloor, the system includes a custom surge-protection device to handle power-on spikes and a specialized aluminum radiator system to dissipate heat from the electronics into the seawater.

Datasets

The study does not rely on a pre-existing public dataset. Instead, it generated a novel dataset during a six-month field trial (September 2021 to March 2022) at the Changjiang nuclear power plant in Hainan, China. The system was placed on the seabed 800 meters from the shore. This trial produced thousands of high-resolution images of DCOs and continuous streams of salinity, temperature, and current data.

Key Insights and Ideas

  • The "Night Shift" Threat: The system revealed that Acetes chinensis (shrimp) abundance peaked sharply between 00:00 and 09:00. As temperatures rose during the day, the organisms migrated to deeper, cooler water, suggesting that intake risks are highly diurnal.
  • Predictive Power: By combining imaging with ADCP current data, the system doesn't just count organisms; it uses the SCHISM model to forecast their arrival at the intake gates, giving operators hours of lead time.
  • Hardware Reliability: A significant portion of the paper focuses on the "boring but essential" engineering—surge protection and thermal management—proving that the system can operate autonomously for half a year without maintenance.

Key Results

  • Stability: The DC/DC power conversion remained stable within 1% variation even during load shifts.
  • Automated Detection: The software successfully identified and estimated the abundance of Phaeocystis globosa colonies, Acetes chinensis, and small fish.
  • Real-time Alarming: The system successfully demonstrated an automated alarm trigger when DCO abundance crossed a pre-set threshold.
  • Diurnal Correlation: Results showed a clear correlation between water temperature and organism migration, providing a biological basis for future risk modeling.

Limitations and Future Work

  • Size Constraints: The current OMI is optimized for organisms up to 20 mm. Larger threats, like large jellyfish or floating macroalgae, still require complementary systems like sonar or wide-angle cameras.
  • Spatial Resolution: A single seafloor node provides a "point" observation. The authors acknowledge that a large intake area requires a multi-point monitoring network to reduce estimation errors.
  • Counting Errors: The authors noted that ocean currents can cause the same organisms to be counted multiple times if they drift back and forth past the sensor, a problem they aim to solve with more sophisticated image-tracking algorithms.
  • Biofouling: While not a focus of this paper, long-term optical sensors in the ocean eventually face "biofouling" (growth on the lens), which will need to be addressed for multi-year deployments.

Summary updated

Abstract

Nowadays, nuclear power plays an important role in the energy structure of many countries. However, A bloom of a disaster-causing organism (DCO) in the cold-water intake area of a coastal nuclear power plant can block the water cooling system and seriously affect the operational safety of the nuclear power unit. Currently, the traditional method of protection is to estimate the DCO abundance by regular manual investigation and sampling, but that method cannot give continuous real-time data. Instead, proposed and implemented here is a seafloor in situ integrated monitoring system for DCOs (known as IMSDCO), which is equipped with an optical microscopic imager (OMI) and hydrometric sensors to monitor automatically the DCO abundance and hydrology. All the data are transmitted to a terminal in the shore station through a photoelectric composite cable for real-time display. When the DCO abundance reaches a preset threshold, software automatically raises an alarm. Since placing IMSDCO at the cold-water intake of the Changjiang nuclear power plant, a six-month field trial has been completed, during which large amounts of hydrology data and DCO images were obtained. IMSDCO successfully identified and estimated the abundances of various DCOs (e.g., Phaeocystis globosa , Acetes chinensis , and small fish) and predicted their movements based on hydrology data. Based on the analysis of the experimental data, we discussed the reasons for the error in the abundance estimation of DCO and the methods to reduce the error. The experimental results show that the OMI-based IMSDCO can monitor and give early warning of DCOs in the water intake areas of costal nuclear power plants and is worthy of long-term deployment.
BibTeX
@inproceedings{li2023integrated,
  author = {Chao Li and Jian-ling Huo and Yu-ze Song and Lei Yang and Song-tang Liu},
  title = {{An integrated monitoring system for disaster-causing organisms in the water intake areas of coastal nuclear power plants}},
  booktitle = {Frontiers in Marine Science},
  year = {2023},
  doi = {10.3389/fmars.2022.1089699},
}
Figure