{
  "id": 12452455,
  "title": "From food to fuel, AI helps pinpoint how to grow 'microalgae' at scale",
  "url": "https://urgent.news/2026/10/06/from-food-to-fuel-ai-helps-pinpoint-how-to-grow-microalgae-at-scale",
  "topic": "science",
  "section": "Science",
  "published": "2026-10-06T19:20:15.000Z",
  "source": {
    "name": "Phys.org",
    "slug": "phys-org",
    "url": "https://phys.org/news/2026-10-food-fuel-ai-microalgae-scale.html"
  },
  "original_language": "en",
  "account": "Microalgae, microscopic organisms that utilize light and carbon dioxide for growth, have potential applications in producing various ingredients, food for animals, pigments, and even renewable fuels. However, cultivating microalgae efficiently at an industrial scale remains a challenge. My research focusing on AI-assisted monitoring and imaging aims to analyze microalgal growth and improve cultivation.\n\nMicroalgae, found abundantly in seas, rivers, and lakes, serve as primary producers through photosynthesis, generating proteins, fats, carbohydrates, and other organic compounds. Cultivation in photobioreactors, industrial systems tailored for optimal conditions of photosynthetic microorganisms, necessitates managing several parameters such as light, temperature, pH, and oxygen concentration. These parameters often have interdependencies, for instance, insufficient light leading to limited photosynthesis, while excessive light can cause photoinhibition, reducing photosynthetic efficiency. Moreover, as the microalgae population increases, cells can shade each other, reducing the light available.\n\nAddressing the complexities of cultivation in laboratory settings, my research delves into AI-assisted monitoring and imaging to analyze microalgal growth and optimize cultivation conditions. AI and machine-learning methods are already employed to monitor and optimize certain conditions. Sensors continuously measure parameters like temperature, pH, dissolved oxygen, light, and carbon dioxide; machine-learning models analyze the gathered data to identify patterns related to the culture's condition and growth.\n\nIn my work, I have reviewed the applications of AI in photobioreactors and explored its potential benefits. Machine-learning methods have been utilized to predict microalgal growth and optimize cultivation conditions. Image-based approaches are also under study for monitoring microalgae without solely relying on manual sampling. By analyzing images of a culture, researchers can estimate changes in biomass or culture density, a practice that my colleagues and I are developing within our microalgae laboratory.\n\nSome biological changes are challenging to measure continuously. Conditions leading to stress for microalgae and inhibiting growth or productivity may develop before clear changes in the culture become observable. Integrating various sensor measurements with machine-learning models assists researchers in identifying these subtle changes and understanding the underlying processes during cultivation. The insights gained can then inform decisions regarding operating conditions such as carbon dioxide supply, pH, and light.\n\nCurrent microalgae production primarily exists at laboratory or small pilot scales, though AI-based control has been tested in larger outdoor cultivation systems. For instance, a 2026 study conducted in Almeria, Spain, utilized reinforcement learning in an 80 m² (860 square feet) open raceway for microalgae cultivation. An open raceway is an open-air system where the water containing microalgae circulates through channels, typically employing a paddlewheel. The AI system, connected to the cultivation process, regulated pH by controlling carbon dioxide injection. Leveraging measurements such as temperature, sunlight, and dissolved oxygen, the AI-based controller decided on the required amount of carbon dioxide to add. Under changing outdoor conditions over eight days, the study demonstrated that the controller effectively managed the process while responding to operational changes. The research illustrates AI-based control being applied to an operating microalgae cultivation system, as opposed to being evaluated through modeling or laboratory experiments.\n\nSeveral challenges remain before widespread adoption of AI-based control in commercial production. AI models rely on reliable data; sensors in photobioreactors can become fouled or exhibit measurement drift over time. Additionally, microalgae strains may respond differently to various cultivation conditions, potentially leading to model performance variations when applied to different systems. As my research focuses on both biological and engineering aspects of microalgae cultivation, it is crucial to recognize that AI should be viewed as an adjunct tool for researchers and process operators, rather than a replacement for conventional process engineering. AI excels at analyzing vast amounts of cultivation data and supporting decisions on adjusting operating conditions.",
  "summary": "Microalgae (microscopic organisms that use light and carbon dioxide to grow) can be used to make ingredients and food for animals, pigments and potentially renewable fuels. But growing microalgae efficiently at an industrial scale is difficult.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}