The growing need for renewable energy sources is increasing every day as the current fossil fuels are becoming scarce, which is why renewable energy sources are becoming crucial as studies are being conducted to prepare for when nonrenewable fuels are fully utilized. Luckily, scientists have been using simulation models such as the HOMER, and using simulation is a great way to investigate what needs to be done, but this means the user must do multiple trials and errors.  Deep reinforcement learning can be used instead and stems from machine learning, which is considered artificial intelligence, to maximize the overall reward of a system. It consists of an agent which is considered the “brain” as well as an environment which is everything around or the surface of a system. It is trained to make the best decision using an agent that consists of “actors” or several types of agents, called a “critic”. Within a photovoltaic system, the photovoltaics are treated as the generators and the electrical batteries the storage unit. Having astute energy handling is the most important part to be able to incorporate deep reinforcement learning into the electrical grid.  For the research of this paper, deep reinforcement learning is used to enhance the output of photovoltaic systems within new solar power builds. The simulation was used with Matlab Simulink and coding, but instead of using a regular proportional-integral-derivative (PID) controller system and having to run the program countless times, a reinforcement learning agent was used instead to generate observations, calculate the rewards, and stop the simulation when need be. With the use of reinforcement learning, instead of using a tracking system to track the sun’s position or having to change daily based on predicted weather, the agent would provide a way to learn how to acquire the most energy by accounting for multiple factors at once, solving the difficult issues of not only the now but what the system should do in the future.