This dissertation focuses on the control, optimization, and mechanism design of standalone and networked energy systems with heterogeneous distributed energy resources (DER) compositions and consumption preferences. By formulating an inclusive Net Energy Metering (NEM) X tariff design that captures features of existing NEM policies, we establish a generalized theoretical framework to analyze prosumer decisions, DER adoption dynamics, and optimal pricing by regulators and DER aggregators. After introducing the NEM X tariff design and its properties, the first part of this dissertation provides the first analytical characterization of optimal prosumer decisions under NEM X with differentiated retail and sell rates. When considering flexible loads and stochastic renewable distributed generation, we show that the optimal consumption policy for a prosumer has a two-threshold structure that schedules the consumption based on the available renewable generation. Both thresholds are obtained in closed-form and computed apriori. When a battery energy storage system is incorporated, a stochastic dynamic program is formulated to co-optimize the random DER. To overcome the computational complexity of the otherwise intractable program, a greedy co-optimization algorithm that even does not require the underlying probability distributions of the renewable generation is proposed. We show that the performance loss of the proposed greedy algorithm is quite small and provide a sufficient condition for the algorithm to be optimal. We delineate the structural properties and intuitions of the optimal prosumer decisions and consider the impacts of exogenous parameters such as the level of renewable distributed generation and NEM X parameter on the endogenous quantities of flexible consumption, storage output, and prosumer surplus. In the second part of this dissertation, we leverage the analytical framework for the optimal prosumer decisions under NEM X to analyze the regulator’s rate-setting process that determines NEM X parameters. In particular, a nonlinear feedback system model is used to capture the long-run dynamics of the rate-setting process via solving a stochastic Boiteux-Ramsey pricing problem. As a result, we evaluate the performance of several NEM X variants using real and synthetic data to illuminate the impacts of NEM policy designs on social welfare, cross-subsidies of prosumers by consumers, and DER adoption dynamics in both short and long run. The third part of this dissertation focuses on mechanism design for energy communities and DER aggregation that are cognizant of network constraints. In the energy community work, a Stackelberg bi-level optimization is formulated to capture the interaction between the community operator who optimizes its pricing rule to maximize community welfare, and its members who maximize individual consumption benefits given the operator’s pricing rule. An ex-ante market mechanism, referred to as Dynamic NEM (D-NEM), is proposed, which dynamically sets thecommunity NEM price based on aggregated community DER. D-NEM achieves the community's maximum welfare in a decentralized way by incentivizing community members through pricing. D-NEM also stabilizes the coalition by achieving group rationality and efficiency. We show how D-NEM can be generalized to internalize network constraints. Lastly, a competitive profit-maximizing DER aggregator model is proposed with the goal of aggregating prosumers' DER and bidding it in the wholesale electricity market. This work develops the first profit-maximizing competitive DER aggregation solution for a DER aggregator to participate in the wholesale market. A constrained optimization that maximizes DER aggregator profits while ensuring the individual rationality of its members is proposed as a DER aggregation approach that competes with the conventional DSO.