DeepCtrls Showcases Physical AI for Smart Facilities at the 9th Green Factory Conference

The 9th Green Factory Facilities Management Conference recently took place in Wuxi. Under the theme “Empowering Green Factories Through Energy-Efficient, Low-Carbon Facilities Management,” the event provided a national platform for exchanging ideas on industrial sustainability, attracting more than 30,000 attendances from industry professionals, technical experts, and business representatives.

As China advances its carbon peaking and carbon neutrality goals, the transformation of factory facilities is entering a more demanding phase. Growing adoption of Physical AI is moving facilities operations toward a new model built on physics-based frameworks, integrated data, and autonomous optimization.

As a leader in Physical AI-powered energy optimization, DeepCtrls showcased three areas of innovation for advanced manufacturing and computing infrastructure: zero-carbon campuses, predictive maintenance, and the DeepBot AI algorithm platform. Powered by proprietary PhyAI technology, these solutions address coordinated energy and carbon management, equipment reliability, and intelligent control, creating an AI-native facilities management framework spanning energy optimization, maintenance, and carbon governance.

Core Technology and Solutions

Zero-Carbon Campuses: Integrated Energy and Carbon Management

Green factories increasingly require coordinated energy and carbon management alongside integrated operations. DeepCtrls combines generation, grid, load, storage, and carbon management with Physical AI optimization to create a unified management hub for advanced manufacturing campuses, including semiconductor and electronics facilities.

PhyAI dynamically optimizes distributed generation, flexible loads, and energy storage charging and discharging. This enables precise peak shaving and valley filling while maximizing renewable electricity use, with campus green power utilization reaching over 85%.

An AI-powered comprehensive carbon accounting model automatically generates carbon ledgers, emissions alerts, and tailored reduction strategies, supporting average annual reductions in carbon intensity of 15–25%.

Real-time monitoring across energy systems, equipment, and emissions provides operational visibility and rapid anomaly response. By coordinating production processes with energy loads, the solution reduces unnecessary consumption and supports reductions of more than 15% in energy use per unit of output value.

Predictive Maintenance: Toward Uninterrupted Facilities Operations

Reliable facilities equipment is essential to continuous production and efficient energy use. Conventional scheduled maintenance and reactive repairs struggle to meet modern factories’ requirements for reliability and precision maintenance.

Deployment and Key Capabilities

Combining PhyAI algorithms with Failure Mode and Effects Analysis (FMEA), DeepCtrls evaluates equipment health using operating thresholds and trends. Analysis of vibration, temperature, pressure, and other data enables early fault warnings, health assessments, and predictive maintenance.

The system correlates abnormal parameters, fault symptoms, and historical data to identify root causes quickly and automatically generate standardized maintenance work orders. This creates an intelligent workflow spanning sensing, prediction, root-cause analysis, and resolution.

The approach enables proactive fault prediction and targeted maintenance, helping prevent unplanned downtime and extend equipment life. It supports businesses in bringing downtime losses close to zero and halving maintenance costs.

Practical Benefits and Industry Value

DeepBot AI Algorithm Platform: Intelligent Energy Optimization and Decision-Making

DeepCtrls also demonstrated the built-in AI algorithm platform for DeepBot, the industry’s first L4-level energy efficiency agent.

Combining PhyAI with reinforcement learning, the platform autonomously manages the complete cycle of data collection, intelligent analysis, strategy generation, and closed-loop optimization. Core capabilities include:

• AI-powered energy consumption analysis.

• Intelligent equipment management and diagnostics.

• Real-time operational monitoring.

• Energy-saving potential assessment.

Lightweight, plug-and-play deployment and rapid strategy validation help facilities teams overcome delayed decision-making and reactive optimization. Designed for complex industrial operating conditions, the platform supports high energy efficiency, safe and stable operation, and low-carbon performance.

DeepCtrls’ three solutions work together to address equipment reliability, energy savings, precise maintenance, and zero-carbon compliance. Their integrated capabilities attracted widespread industry attention at the conference.

DeepCtrls will continue advancing Physical AI to help industrial facilities transition from fragmented management to integrated, autonomous optimization, accelerating the development of next-generation intelligent green factories.