Process Optimization
From Utilizing Expert System to AI Modelling
We can further enhance your system efficiency to achieve more with less
HVAC Optimization
HVAC systems accounts for a significant proportion of a building's energy consumption. A typical system accounts for more than 40% of total energy consumed by the building. Hence, being able increase the efficiency of HVAC systems through system optimization can bring long term benefits including energy savings and lowered carbon emissions.
​
Through analyzing a number of different variables within the HVAC system such as occupancy, indoor/outdoor temperature and humidity, usage pattern, equipment data and etc...and applying different types of AI modelling, the cooling load can be accurately estimated and system control setpoints can be optimally set to match the required cooling/heating demand.
W/WW Treatment Optimization
Water and Waste Water (W/WW) Treatment comprise of a number of different processes where optimization can be carried out to either reduce material consumption or enhance energy saving. Some processes where optimization can be applied are:
​
1) Dosing Process
By utilizing model predictive control and AI modelling, the optimal setpoint for different chemicals can be achieved resulting to a more efficient dosing process. This can reduce excessive chemical consumption and sludge production.
​
2) Pump Optimization
By anlayzing water inflow and outflow rates as well as historical consumption pattern, the operation of pumps can be optimized to minimize electricity consumption whilst being able to effectively discharge content based on predicted operating conditions.
Compressed Air Optimization
Compressed Air system used in industrial applications consume high amounts of energy. In order to ensure there is adequate air provided to the various systems and processes, the compressed air level is often over supplied to avoid undersupply when there is peak demand. This results to excessive energy consumption and wastage.
​
By analyzing the system and equipment characteristics, we are able to utilize machine learning models that can forecast local air demand for each process and conduct dynamic control to respond to changing demands. Not only are compressed air levels optimally tuned, but also the sequencing of multiple compressors being started and stopped to achieve the highest plant efficiency.
Steps to Successful Optimization
1
Assessment and Analysis
Outline the goal and success criteria of the system based on existing information and knowledge. Gather as much system data as possible including system performance, bottlenecks and other measurable metrics. Identify the variables and understand its impact on the system.
2
Strategy and Implementation
Generate potential strategies for optimization including adding additional sensors for data collection, upgrading equipment and modifying control process based on the use of various modelling techniques. This may include expert system modelling, AI modelling as well as model predictive controls.
3
Monitoring and Continuous Improvement
Continuous tracking of system performance against baseline and objectives. Ensure there is a reliable and systematic method to analyze system performance. Fine tune system models and parameters based on feedback.
