Application of artificial intelligence technology in the field of energy conservation
The application of artificial intelligence technology in the field of energy conservation
The principle of AI-driven digital intelligence in energy conservation
Smartwatts AI calculates the optimal operating parameters of the central air conditioning system in real time, dynamically adjusts the balance between cooling supply and demand, reduces cooling waste, and ultimately achieves energy savings.
Core platform architecture: Cloud-edge-end AlOTarchitecture
Digital simulation + AI model
Through the digital simulation-assisted verification platform, real mechanism data perception and AI-based load prediction are rapidly modeled mathematically, enabling real-time optimization of system operations.
MaaS model as a service
IIt supports API calls to AI algorithm models, calculates and generates optimal paths in real time without requiring downloads or live iterations, and enables seamless integration with multiple Building Automation (BA) systems.
Flexible deployment
IIt supports API calls to AI algorithm models, calculates and generates optimal paths in real time without requiring downloads or live iterations, and enables seamless integration with multiple Building Automation (BA) systems.
Digital Twin Mechanism Simulation Platform
Based on digital twin technology, the multidisciplinary mechanism model of the air conditioning system is integrated to enable dynamic simulation, real-time monitoring, performance optimization, and early fault detection throughout the entire lifecycle of the central air conditioning system.
A highly accurate digital twin is built to mirror the actual air conditioning system—including chillers, air handling units, air ducts, and terminal equipment—ensuring precise simulation and control.
Through virtual-real data interaction and mechanism-based simulation, the system’s operating state, energy transfer dynamics, and response characteristics are accurately reproduced.
Provides scientific support for system design, operational management, and energy optimization.
Smartwatts AI global intelligent optimization engine
Leveraging artificial intelligence, optimization theory, and the operational characteristics of air-conditioning systems, the optimal control strategy is identified by dynamically analyzing system parameters—such as chiller outlet temperature, pump and fan speeds, damper and valve positions, and unit status—alongside environmental variables.
While ensuring indoor comfort (including temperature and humidity), adhering to equipment safety constraints, and complying with environmental safety regulations, the system optimizes energy consumption, enhances equipment longevity, and improves operational efficiency.
Facilitate intelligent and low-carbon operation of the central air conditioning system to maximize energy efficiency, minimize emissions, and promote sustainable building management.
Autonomous perceptual algorithm model for central air conditioning systems
Framework for autonomous cognitive system mechanisms in intelligent building management.
Without relying on manually preset rules, the system autonomously identifies the internal logic governing equipment operation characteristics, energy flow dynamics, and environmental load responses in central air conditioning systems. This is achieved through the integration of data-driven insights and physical laws—for example, recognizing the mechanism behind COP variations with condensation temperature, and understanding the relationship between water system flow rates and pressure losses.
Dynamically adapt to changes in enviromental conditions
The system continuously perceives real-time changes in outdoor weather, indoor load, and equipment status, and interprets their influence on system mechanisms. For instance, it can detect how elevated summer temperatures lead to increased condensation pressure, which subsequently reduces unit efficiency—capturing such chain reactions to enable intelligent decision-making and adaptive system control.
Supports interpretable decision-making
Control strategies and diagnostic results derived from mechanistic cognition are fully traceable. For example, a recommendation to increase the outlet water temperature is based on the observed reduction in system load, where a higher outlet temperature improves the unit's COP. This interpretability helps mitigate the risks associated with 'black-box' decision-making inherent in purely data-driven models.
Independent iterative learning
As the system operates, it continuously accumulates data and incrementally improves the accuracy of its mechanism models. For example, it can automatically update the efficiency degradation curve caused by equipment aging, allowing the system to adapt intelligently to changes throughout its entire life cycle.