Modeling and Simulation of Predictive Cooling Optimization for Hyperscale Data Centers
Implementation plan:
—————————
Step 1: Initially, we collect and load data from the NREL RSF Measured Dataset.
Step 2: Then, we preprocess the dataset by cleaning missing values, normalizing IT load, temperature, humidity, and cooling-power features.
Step 3: Next, we build a regression model using Linear Regression and Random Forest to predict cooling energy consumption.
Step 4: Next, we train the data using deep learning models such as LSTM and GRU for short-horizon thermal and cooling demand prediction.
Step 5: Next, we apply Reinforcement Learning DQN to control cooling setpoints based on predicted thermal conditions.
Step 6: Finally, we analyze energy-saving percentage, and thermal-stability scores based on loaded data.
Step 7: Finally, we plot performance metrics for the following
7.1: Number of epochs vs. Prediction Accuracy (%)
7.2: Number of epochs vs. Training Loss (%)
7.3: Number of epochs vs. RMSE (kW)
7.4: Time vs. Cooling Energy Consumption (kW)
Software Requirements:
—————————–
1) Development Tool: Python 3.11.4 or above
2)Operating System: Windows 10 (64-bit) or above
Dataset:
———-
Link: https://data.openei.org/submissions/358
Note:
——-
1) If the proposed plan does not fully align with your requirements, please provide all necessary details—including steps, parameters, models, and expected outcomes—in advance. Kindly ensure that any missing configurations or specifications are clearly outlined in the plan before confirming.
2) If there’s no built-in solution for what the project needs, we can always turn to reference models, customize our own, different math models or write the code ourselves to fulfil the process.
3) If the plan satisfies your requirement, Please confirm with us.
4) Project based on Simulation only.
5) If you have any dataset to change, kindly provide us before implementing it.
