adding a visualization for datacenter simulation environment (aws#809)

Author: bbalaji-ucsd

reinforcement_learning/rl_hvac_coach_energyplus/images/datacenter_env.png

@@ -16,6 +16,12 @@
  "\n",
  "As training an RL algorithm in a real HVAC system can take time to converge as well as potentially lead to hazardous settings as the agent explores its state space, we turn to a simulator to train the agent. [EnergyPlus](https://energyplus.net/) is an open source, state of the art HVAC simulator from the US Department of Energy. We use a simple example with this simulator to showcase how we can train an RL model easily with Amazon SageMaker RL.\n",
  "\n",
+ "<br>\n",
+ "\n",
+ "<img width=\"85%\" src=\"images/datacenter_env.png\" />\n",
+ "\n",
+ "<br>\n",
+ "\n",
  "1. Objective: Control the data center HVAC system to reduce energy consumption while ensuring the room temperature stays within specified limits.\n",
  "2. Environment: We have a small single room datacenter that the HVAC system is cooling to ensure the compute equipment works properly. We will train our RL agent to control this HVAC system for one day subject to weather conditions in San Francisco. The agent takes actions every 5 minutes for a 24 hour period. Hence, the episode is a fixed 120 steps. \n",
  "3. State: The outdoor temperature, outdoor humidity and indoor room temperature.\n",