Daily total energy by operating-day sequence
Baseline controlAI control
Cooling plant case
Government office cooling plant
AI supervisory control optimized chilled-water pump frequency, condenser-water pump frequency, and chiller leaving-water temperature on a compact office plant.
- Energy savings result
- 16.81%
- Average daily savings
- 239.77 kWh
- Total measured savings
- 2,397.71 kWh
Average result across the alternating operating window.
Reported average energy reduction during compared operation.
Aggregate savings for the evaluated operating window.
Plant profile
Two-chiller office plant with variable-speed pumps
The central cooling plant included two chillers, two chilled-water pumps, two condenser-water pumps, and two cooling towers. Normal operation kept one magnetic-bearing chiller online with one variable-speed chilled-water pump, one variable-speed condenser-water pump, and two fixed-speed cooling towers.
Equipment
Operating equipment
The active equipment set was small enough for direct supervisory setpoint optimization while still covering chiller, chilled-water, condenser-water, and cooling-tower operation.
| Equipment | Quantity | Specification | Status |
|---|---|---|---|
| Magnetic-bearing chiller | 1 active / 2 installed | Vendor recorded: Haier / 1010 kW cooling capacity / Rated efficiency 0.57 kW/ton | Running |
| Chilled-water pump | 1 active / 2 installed | Vendor recorded: Oriental Pump / 22 kW / 32 m head / 160 m³/h flow / Variable-speed capable | Running |
| Condenser-water pump | 1 active / 2 installed | Vendor recorded: Tucker / 18.5 kW / 20 m head / 200 m³/h flow / Variable-speed capable | Running |
| Cooling towers | 2 active / 2 installed | Vendor recorded: Shangfeng / 7.5 kW each / 350 m³/h flow / Fixed-speed operation | Normally open |
Control scope
Control scope
- Deployment
- Edge compute device plus AI reinforcement-learning control.
- Variables optimized
- Chilled-water pump frequency, condenser-water pump frequency, and chiller leaving-water temperature.
- Evaluation period
- Alternating operation from Jul 24 to Aug 18.
Calculation detail
Energy savings and operating-period calculation
- Comparison design
- Alternating AI and baseline operation.
- Baseline records
- 10 days; total energy 14,503.49 kWh; average 1,450.35 kWh/day.
- AI records
- 10 days; total energy 12,105.78 kWh; average 1,210.58 kWh/day.
- Average savings rate
- 16.81%
- Average savings
- 239.77 kWh
- Total savings
- 2,397.71 kWh
- 1. Baseline average daily energy = 14,503.49 kWh / 10 days = 1,450.35 kWh/day.
- 2. AI average daily energy = 12,105.78 kWh / 10 days = 1,210.58 kWh/day.
- 3. Average daily savings = 1,450.35 - 1,210.58 = 239.77 kWh/day.
- 4. Average savings rate = 239.77 / 1,450.35 x 100% = 16.81%.
- 5. Total measured savings = 239.77 kWh/day x 10 days = 2,397.71 kWh.
Case result
Average savings rate
=Baseline total energybaseline days-AI total energyAI daysBaseline total energybaseline days*100%
=14,503.4910-12,105.781014,503.4910*100%
=16.81%
Daily operating data
Daily operating data
The alternating verification record shows daily equipment energy by mode. The scatter plot uses operating-day sequence on the x-axis and total daily energy on the y-axis because the daily record does not include weather variables.
| Mode | Date | Chiller kWh | Chilled-water pump kWh | Condenser-water pump kWh | Cooling tower kWh | Total energy kWh |
|---|---|---|---|---|---|---|
| AI control | Jul 24 | 725.80 | 187.93 | 110.72 | 161.10 | 1186 |
| Baseline control | Jul 25 | 879.30 | 307.27 | 97.45 | 162.82 | 1447 |
| AI control | Jul 26 | 593.60 | 209.60 | 97.93 | 162.70 | 1064 |
| Baseline control | Jul 27 | 843.10 | 302.02 | 95.81 | 156.02 | 1397 |
| AI control | Jul 28 | 618.30 | 194.95 | 119.23 | 159.10 | 1092 |
| Baseline control | Jul 31 | 901.80 | 307.47 | 97.53 | 161.52 | 1468 |
| AI control | Aug 1 | 635.18 | 181.30 | 114.90 | 155.73 | 1087 |
| Baseline control | Aug 2 | 931.66 | 289.23 | 89.15 | 148.39 | 1458 |
| AI control | Aug 3 | 876.75 | 194.14 | 107.77 | 150.47 | 1329 |
| Baseline control | Aug 4 | 937.90 | 284.71 | 90.18 | 148.78 | 1462 |
| AI control | Aug 7 | 857.20 | 195.81 | 100.29 | 150.90 | 1304 |
| Baseline control | Aug 8 | 965.90 | 293.23 | 94.34 | 158.64 | 1512 |
| AI control | Aug 9 | 922.20 | 224.33 | 100.56 | 159.58 | 1407 |
| Baseline control | Aug 10 | 898.20 | 293.23 | 94.89 | 159.06 | 1445 |
| AI control | Aug 11 | 741.50 | 204.59 | 120.42 | 159.78 | 1226 |
| Baseline control | Aug 14 | 1083.21 | 297.22 | 95.98 | 161.87 | 1638 |
| AI control | Aug 15 | 820.80 | 201.35 | 101.51 | 156.58 | 1280 |
| Baseline control | Aug 16 | 933.00 | 292.50 | 92.19 | 150.04 | 1468 |
| AI control | Aug 17 | 708.00 | 188.80 | 92.72 | 141.66 | 1131 |
| Baseline control | Aug 18 | 702.50 | 276.90 | 87.32 | 141.16 | 1208 |
Compare your plant against these operating patterns.
Bring plant topology, BMS trend logs, and utility data. We will map what can be measured before proposing an optimization scope.