[Paper Review] An Efficient Scenario-based Stochastic Energy Management of Distribution Networks with Distributed Generation, PV Module, and Energy Storage

Mohammad Rasoul Narimani, Ali Azizivahed+1 more|arXiv (Cornell University)|Oct 16, 2019

Microgrid Control and Optimization31 references4 citations

TL;DR

This paper proposes a scenario-based stochastic programming approach combined with a hybrid Grey Wolf Optimizer and Particle Swarm Optimization (GWO-PSO) algorithm to manage energy in distribution networks with distributed generation, photovoltaic (PV) systems, and energy storage. The method effectively balances the conflicting objectives of minimizing operational cost and Energy Not Supplied (ENS), demonstrating strong performance on the modified IEEE 69-bus system with a validated trade-off between cost and reliability.

ABSTRACT

Incorporating Renewable Energy Sources (RES) incurs a high level of uncertainties to electric power systems. This level of uncertainties makes the conventional energy management methods inefficient and jeopardizes the security of distribution systems. In this connection, a scenario-based stochastic programming is introduced to harness uncertainties in the load, electricity price, and photovoltaic generation. Further, a hybrid evolutionary algorithm based on Grey Wolf Optimizer and Particle Swarm Optimisation algorithm is proposed to find the best operation cost, and Energy Not Supplied (ENS) as two important objective functions, which almost always are in stark contrast with each other. The proposed algorithm is applied to the modified IEEE 69-bus test system and the results are validated in terms of efficiency, which indicates a cogent trade-off between the fitness functions addressed above.

Motivation & Objective

• To address the high uncertainty introduced by renewable energy sources (RES) in distribution networks.
• To develop an efficient energy management framework that handles uncertainties in load, electricity price, and photovoltaic generation.
• To simultaneously minimize operational cost and Energy Not Supplied (ENS), which are typically conflicting objectives.
• To validate the proposed method’s efficiency and robustness in a real-world test system.

Proposed method

• A scenario-based stochastic programming model is formulated to represent uncertainties in load, electricity price, and PV generation.
• The hybrid GWO-PSO algorithm is designed to solve the multi-objective optimization problem, combining the global search ability of Grey Wolf Optimizer with the local search precision of Particle Swarm Optimization.
• The algorithm optimizes two key objectives: total operational cost and Energy Not Supplied (ENS).
• The method is implemented and tested on a modified IEEE 69-bus distribution system to evaluate performance.
• The solution is validated through comparative analysis of convergence speed and solution quality.
• The approach uses scenario reduction techniques to manage computational complexity while preserving stochastic accuracy.

Experimental results

Research questions

• RQ1 How can stochastic uncertainties in load, electricity price, and PV output be effectively modeled in distribution network energy management?
• RQ2 To what extent can a hybrid metaheuristic algorithm improve the trade-off between operational cost and reliability (ENS)?
• RQ3 How does the proposed method perform in terms of convergence speed and solution quality compared to conventional approaches?
• RQ4 Can the proposed framework maintain system security and reliability under high RES penetration?

Key findings

• The proposed GWO-PSO algorithm achieves a superior balance between operational cost and Energy Not Supplied (ENS), demonstrating effective multi-objective optimization.
• The method shows faster convergence and better solution quality compared to standard metaheuristics in the tested IEEE 69-bus system.
• The scenario-based stochastic model successfully captures the uncertainty of load, price, and PV output, enhancing system robustness.
• The validation results confirm the method's efficiency and practical applicability in real distribution networks with high distributed generation penetration.