Plant case

Pharmaceutical plant

AI operating-period average plant efficiency improved from 0.74 to 0.65 kW/ton, with BMS-compatible control bands, equipment sizes, daily operating data, frequency behavior, and calculation detail shown below.

Plant efficiency improvement
12.11%

Calculated for the AI operating period.

AI operating window
37 days

Sep 1 to Oct 7 AI control compared with Jun 1 to Jul 7 autonomous control.

Average plant efficiency
0.74 -> 0.65 kW/ton

Baseline average to AI-period average.

Plant inventory

Equipment brands, capacities, and active plant topology

The site uses a secondary-pump variable-flow chilled-water system. The central plant includes four chillers with one standby unit, four chilled-water pumps, four condenser-water pumps, three air-side chilled-water pumps, and four process-side chilled-water pumps. Normal operation is one chiller paired with one pump.

EquipmentBrandSpecificationStatus
Centrifugal chillerTraneCooling capacity 650RT / Power 402kWRunning
Chilled-water pumpGrundfosHead 20m / Power 30kW / Flow 350m³/hRunning
Condenser-water pumpXylemHead 22m / Power 45kW / Flow 190m³/hRunning
Cooling towerBACFlow 500m³/h / Power 22kWRunning

Control and BMS compatibility

The AI layer stayed compatible with existing group control

The platform supports one-click switching between existing group control and AI control. Operators can also apply AI recommendations to group-control setpoints for condenser-water pump frequency and cooling-tower frequency.

Condenser-water pump band
42Hz to 50Hz, constrained by low-flow protection.
Cooling-tower fan band
22Hz to 50Hz, optimized against plant energy and indoor-temperature response.
Current optimization scope
Condenser-water pumps and cooling towers.
Reserved future interfaces
Chilled-water supply temperature and chilled-water pump frequency.
Data source
BMS operating records, daily system kW/ton, outdoor dry-bulb temperature, and outdoor wet-bulb temperature.

Control logic

Cooling-water pump and cooling-tower optimization loop

The control loop reads BMS and weather context, applies explicit operating boundaries, selects a DDQN frequency command, and returns measured feedback to the policy.

  • Boundary checks keep every command inside approved BMS-compatible frequency bands and protection thresholds.
  • Operator review and M&V results close the loop back to the DDQN policy without bypassing the native BMS authority.

  1. BMS + Weather Data

    Plant state, trends, operating context

  2. Boundary Checks

    Frequency bands, load threshold, protection

  3. ClimaMind DDQN Policy

    Frequency command selection

  4. Approved Commands

    Condenser-water pump and tower fan frequency

  5. Operator + M&V Feedback

    Visibility, fallback, COP and kW/ton review

Cooling-tower fan frequency variation on July 2 under baseline control mode.

July 2 Cooling-Tower Frequency: Baseline Control ModeTower #1 FreqTower #2 FreqTower #3 FreqTower #4 Freq55504540353000:0003:0006:0009:0012:0015:0018:0021:0024:00HzTime of day

A brief single-sample transient around 08:30 was excluded because it did not represent sustained cooling-tower frequency operation.

Cooling-tower fan frequency variation on September 28 under software-optimized mode.

September 28 Cooling-Tower Frequency: Software-Optimized ModeTower #4 FreqTower #1 FreqTower #3 Freq51494745434139373500:0003:0006:0009:0012:0015:0018:0021:0024:00HzTime of day

The software-optimized mode modulated tower commands across the day instead of holding the baseline near-fixed tower frequencies; tower running energy fell by about 30% in the comparison window.

Calculation detail

Plant efficiency improvement and operating-period calculation

Baseline period
Jun 1 to Jul 7 baseline operation; 27 valid daily records after missing-power-data exclusions.
Baseline averages
0.74 kW/ton, outdoor dry-bulb 22.66C, outdoor wet-bulb 19.36C.
AI period
Sep 1 to Oct 7 AI control; 37 daily records.
AI averages
0.65 kW/ton, outdoor dry-bulb 25.64C, outdoor wet-bulb 22.43C.
AI-period highest plant efficiency
0.73 kW/ton, from the daily operating data.
  1. 1. Baseline average plant efficiency = 0.74 kW/ton.
  2. 2. AI-period average plant efficiency = 0.65 kW/ton.
  3. 3. Reduction = 0.74 - 0.65 = 0.09 kW/ton.
  4. 4. Plant efficiency improvement = (0.74 - 0.65) / 0.74 * 100% = 12.11%.

Case result

Plant efficiency improvement
=Baseline average kW/ton-AI average kW/tonBaseline average kW/ton*100%
=0.74-0.650.74*100%
=12.11%

Efficiency benchmark

U.S. reference context for plant efficiency

U.S. chilled-water plant practice usually compares measured plant performance in kW/ton. The table gives ASHRAE reference context without treating it as a certification claim.

MetricBaselineAI periodU.S. reference context
Plant efficiency0.740.65Lower is better; this is the main U.S. operating-efficiency view for chilled-water plants.
ASHRAE reference context0.61-0.70 kW/ton0.61-0.70 kW/tonASHRAE District Cooling Guide lists electric centrifugal chillers in this typical range.
Measurement boundarySystem kW/tonSystem kW/tonUsed as reference context, not a U.S. certification score or whole-building energy benchmark.

Daily operating data

Daily plant efficiency, dry-bulb, and wet-bulb records used for the scatter plots

The table below lists daily operating records, and the two plant-efficiency weather scatter plots use kW/ton values from these points.

Plant efficiency vs wet-bulb temperature

29262219150.550.650.750.85kW/ton
Baseline operationAI control

Plant efficiency vs outdoor dry-bulb temperature

33292521170.550.650.750.85kW/ton
Baseline operationAI control
ModeDatekW/tonDry-bulb CWet-bulb C
Baseline operationJun 10.7121.2718.27
Baseline operationJun 20.7322.4617.78
Baseline operationJun 30.6720.3615.11
Baseline operationJun 80.6921.2917.67
Baseline operationJun 90.7023.2318.43
Baseline operationJun 100.7123.0617.79
Baseline operationJun 110.6921.2417.54
Baseline operationJun 140.6920.8017.49
Baseline operationJun 150.6921.5716.96
Baseline operationJun 160.6820.7116.87
Baseline operationJun 170.6918.2517.24
Baseline operationJun 180.7319.3018.77
Baseline operationJun 190.7620.5219.80
Baseline operationJun 200.7722.3119.73
Baseline operationJun 210.7523.5818.96
Baseline operationJun 220.7222.6917.77
Baseline operationJun 230.7020.2417.18
Baseline operationJun 240.7118.1117.63
Baseline operationJun 250.7621.1719.72
Baseline operationJun 260.7621.6219.60
Baseline operationJul 10.7923.8620.88
Baseline operationJul 20.8224.2721.66
Baseline operationJul 30.8224.6721.84
Baseline operationJul 40.8226.1221.82
Baseline operationJul 50.8125.1121.78
Baseline operationJul 60.8531.3927.41
Baseline operationJul 70.8532.5426.95
AI controlSep 10.6426.1021.77
AI controlSep 20.6327.0622.70
AI controlSep 30.6426.9623.74
AI controlSep 40.6728.6925.27
AI controlSep 50.7028.1825.15
AI controlSep 60.6728.7124.63
AI controlSep 70.6428.1524.29
AI controlSep 80.6328.1723.43
AI controlSep 90.6327.5722.85
AI controlSep 100.6227.7822.56
AI controlSep 110.6428.8024.47
AI controlSep 120.6727.4824.90
AI controlSep 130.6625.5023.95
AI controlSep 140.6324.6523.29
AI controlSep 150.6324.4223.51
AI controlSep 160.6525.3023.86
AI controlSep 170.6527.7323.87
AI controlSep 180.6929.2425.74
AI controlSep 190.7230.1926.60
AI controlSep 200.6926.8624.41
AI controlSep 210.6323.2020.68
AI controlSep 220.6121.0419.38
AI controlSep 230.6220.4218.89
AI controlSep 240.6423.1122.05
AI controlSep 250.6526.6323.95
AI controlSep 260.6625.6323.42
AI controlSep 270.7227.6224.71
AI controlSep 280.7225.8024.58
AI controlSep 290.7327.4024.55
AI controlSep 300.7124.1522.09
AI controlOct 10.6723.6219.32
AI controlOct 20.5922.9517.60
AI controlOct 30.6222.6318.51
AI controlOct 40.6623.5418.88
AI controlOct 50.6122.2616.79
AI controlOct 60.6021.4516.93
AI controlOct 70.6119.5516.60

Run the same measurement discipline on your plant

We can review BMS points, operating history, equipment sizes, and metering coverage to determine whether a similar AI supervisory layer is practical.

Book a demoBack to cases

Downloadable resource

Download the full PDF case study

Use the PDF version for internal review, project discussions, and sharing the full measured plant case file.

PDF case study, I-P units

Download PDF