Back in 2018, I wrote an article explaining how an automated chilled water plant is not necessarily a fully optimised one. Well, I believe this needs to be addressed again.

I visit many plants throughout the GCC region and the United States, and plant owners and operators make such statements as: “Our plant is fully optimised. We have a remote monitoring service, we have a command-and-control centre, we have a SCADA system, etc., etc., etc.” When you dig deeper, what you find is that data is being collected and analysed, alarms are being generated, but no true real-time optimisation is taking place at these plants.

My team and I at US Chiller Services were recently involved in a project where a plant was being constructed and the owners said they had a robust SCADA system but wanted to optimise the plant with a chiller management programs. These CMS systems are also not true optimisation programs and still fall short of “Chiller Plant Optimisation”. BMS, CMS, SCADA as well as most so-called chilled water plant management systems, chiller plant energy saving programs and plant flow control systems are not truly optimisation-enabling systems. These systems are designed to respond to written sequences that measure an input, make a decision and then instruct the output to be implemented. Each sequence for a control loop is focused on one or two variables and the one output. What’s being employed in much of the industry is packaged software and VFDs, which fall woefully short of a fully optimised chilled water plant. For example, if the chilled water plant demand exceeds 90% of the operating chiller’s capacity or the chilled water supply temperature exceeds setpoint for five minutes, the next standby chiller shall be started. This is exactly how most energy or BMS/CMS and SCADA systems are programed.

In this example, the control system monitors two variables: chilled water system demand and supply temperatures. The control system compares the current values to the maximum operating capacity and chilled water temperature setpoints. Then, if either condition is true, the BMS initiates one output – start the next chiller. For example, a control system is programed to maintain a cooling tower to a fixed 78 degrees F. Chillers are sequenced on once they reach full capacity and chilled water set point is adjusted based on a fixed time of day schedule and chilled water pumps operated to maintain a fixed pressure.

Strategies, such as the aforementioned example, have been proven to consume as much as 75% or more energy at part load, when compared to a true CPECS (fully optimised) plant system.

There are many other control loops operating simultaneously in the control system that are monitoring and controlling the chilled water pumps, the condenser water pumps, cooling tower fans, valve positions, water temperature setpoints and the pressure differentials across piping, but these are all based on the same design to respond to a written sequence that measures an input, makes a decision and then instructs the output to be implemented. In other words, these sequences are static. A chilled water plant comprises many parts that make the whole plant, and to look at independent pieces of equipment/loops and not the system as a whole makes these systems not effective in optimising the plant performance.

Now, let’s look at a true optimisation system. A CPECS (optimised) software system has complete knowledge of compressor, tower and pump performance characteristics, which it uses in real time to modulate control levels to all VFDs and provide the maximum level of system performance while respecting chiller flow, temperature limits and occupant comfort. The result is a chilled water central plant that operates in synchrony to deliver the highest possible total performance. Unlike other static optimisation strategies, as I explained previously, CPECS has the ability to self-correct chiller, tower and pump performance maps such that regardless of wear and tear, inaccuracies in manufacturer’s data or off-design conditions, the chilled water plant will always run at peak efficiency. A CPECS system goes past the central plant and out to the heat exchangers or AHUs, continuously scanning specific data points in order to balance central plant performance with air side performance without sacrificing occupant comfort. Anytime a variable speed chiller plant operates at a capacity less than its maximum there exists a huge opportunity for optimisation of set points and flows without compromising occupant comfort or process temperatures. ASHRAE studies conclude that air conditioning applications operate at part load over 96% of the time. The CPECS web interface displays actual plant performance in real time to the operator, placing maximum focus on efficiency. All CPECS control systems have the ability to remotely warn of equipment failures or poor efficiency through a built-in email server. Each installation deploys with a full enterprise SQL database that resides on the site – all data owned and under control of the plant owner. Performance is directly related to cooling tower temperature and flow rate. An increase of 1 degree F (0.56 degree C) in condenser water inlet temperature may impact chiller performance by as much as 2.8%. Accurate cooling tower control and an optimised total system energy approach is essential in an efficient chiller plant.

Typical cooling tower control neglects tradeoff between fan energy and chiller energy at part load. A CPECS system takes complete system wide view of the chilled water system. It not only monitors all of the various control loops for each particular device or piece of equipment as the BAS/SCADA does, but it also has the capability to understand how the various loops affect each other and can make adjustments to the control loops, based on that understanding. This capability to make changes automatically to control loops – to make a control sequence “dynamic” – is what differentiates a chiller energy optimisation system from a BMS/CMS, SCADA or other so-called optimisation systems.

I was recently at a District Cooling conference in June and was surprised to see that many companies involved in the construction and/or engineering of District Cooling plants are attempting to design and build their own optimisation programs. They stated they are doing this, because they found most control companies do not offer very good “optimisation programs”. This statement I agree with 100%. The problem is when these people were explaining their system architecture, it was also based on responses to written sequences, so their solution is nothing more than the same old, same old. It’s obvious the industry understands there is a dire need for chilled water plant optimisation and that almost everything available falls well short of this goal. That is, except CPECS.

CPECS achieves optimisation through:

• Optimised cooling tower control and sequencing
• Chiller sequencing that seeks lowest kW for the capacity
• Variable speed, variable set point chilled water pumping
• Optimised VFD condenser water pump control
• Optimised pump sequencing
• Chilled water reset based on actual HEX or AHU demand

The CPECS simulation engine contains chiller performance models, cooling tower performance maps created from full- and part-load design conditions, pump curve design information, and heat exchanger thermal models. The simulation engine uses inputs of actual cooling capacity measured by a BTU meter, installed in the chilled water supply lines, supply chilled water temperature setpoint and outside air wet bulb temperature. With the installation of power meters on all plant equipment for real-time Kw data collection, the CPECS can monitor the individual power consumption for every separate piece of equipment and can attempt to trim the power used for a specific piece of equipment. The total amount of kW used by all the plant equipment is the sum of the simultaneous demands of all the plant equipment – the overall chiller plant kW/ton. This is the purpose of a CPECS, overall chiller plant kW/ton reduction in real time. The CPECS engine iterates through each available chiller sequence, tower combination, operational mode, condenser water flow and cooling tower setpoint to find the combination that results in the lowest instantaneous plant power/kW demand.

The CPECS also has Machine learning/AI capabilities, and thus the system “learns” about the installed plant system and then uses that data to select the most energy-efficient combination of equipment and settings. The CPECS uses predictive methods at first to determine equipment operation, observes the resulting energy usage and then adapts to fine tune the settings.

The CPECS’ features include:

• Energy report generation and condition monitoring
• Seven years onsite and offsite historical data storage
• Internet remote monitoring and control
• Chilled water flow monitoring as standard
• Open programming language and open protocol
• Easy integration into existing BAS/CPM/SCADA network
• Carbon usage reporting

It’s a strong statement but a very true one – a CPECS system just has better technology and greater savings versus anything available in the chilled water industry. CPECS optimisation systems have achieved annual total plant operating efficiencies of 0.48kW/Ton (7.3 COP) and better. These results far exceed today’s energy efficiency code requirements and defy conventional thinking. Each plant is installed with its own flow, electrical and temperature metering, which enables the plant owner to view plant efficiency in real-time. In addition to electrical savings, reductions of up to 10% in tower water consumption can be realised. CPECS plants have connectivity to Modbus RTU, Modbus TCP/IP or BacNet.

Measurable results

CPECS optimisation allows immediate evaluation of savings through the use of a real-time baseline calculation. Baseline performance can be programmed as a 90.1 code-compliant plant or a custom baseline that evaluates actual performance against a pre-retrofit value. The CPECS web interface, unlike any other chiller optimisation software, delivers to the end user actual performance baseline and, most important, target performance indicators. Other “optimisation” providers and programs cannot deliver a real-time performance target; as a result, performance shortcomings are not realised until too late. CPECS optimisation graphical interfaces are user friendly and can be viewed from anywhere through the Internet.

A CPECS virtual panel can also be installed in any chilled water plant and overlaid on the existing BMS/CMS or SCADA system; and the CPECS system will operate for three months and allow you to see what your savings would be with CPECS software operating the chilled water plant versus the current system. In other words, the predicted savings can be proven and a determination can be made based on the savings to have a fully optimised CPECS system installed in the plant before even having it installed.

If the goal is true chiller plant optimisation, then the way to achieve that today is with a CPECS. I have seen many systems and have done extensive research, and no programs were able to save as much Energy/kW/ton as a CPECS.