⚡ electrical engineering

Optimal Grid Connection Capacity of a Hybrid Renewable Microgrid: An Energy Management and Economic Optimisation Model

This paper presents an energy management and economic optimization model to determine the optimal grid connection capacity for a hybrid renewable microgrid, successfully reducing its levelized cost by 2.8% across various load scenarios through a priority-based algorithm that balances import, backup generation, and storage costs.

Original authors: Ihor Buratynskyi, Artur Zaporozhets, Andrii Bosak

Published 2026-09-23

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Ihor Buratynskyi, Artur Zaporozhets, Andrii Bosak

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a small, self-contained neighborhood that generates its own electricity using sunshine and wind, stores the extra power in a giant battery, and keeps a gas generator on standby for cloudy, calm days. This is a microgrid. In the past, engineers designed these systems by guessing how much power they might need to pull from the main city grid or how much they could send back. They often sized the connection to the main grid based on a worst-case guess, which meant paying for a large, expensive cable that sat mostly unused. But electricity markets are changing. Prices for buying and selling power fluctuate every hour, and the cost of keeping a connection open is a fixed bill that must be paid regardless of how much electricity actually flows through it. The question researchers are now asking is simple but difficult: what is the exact size of that connection to the main grid that keeps the whole system running the cheapest, without leaving anyone in the dark?

A team of researchers from Ukraine has built a computer model to answer this question for a specific type of hybrid microgrid. They did not just look at how much power the solar panels or wind turbines could make; they created a detailed economic simulation that weighs every possible cost against every possible benefit. Their system includes a sixty-kilowatt solar array, a twenty-kilowatt wind turbine, a forty-kilowatt backup gas generator, and a battery system capable of storing two hundred kilowatt-hours of energy. The model also accounts for a crucial reality: not all electricity users are the same. Some loads, like hospitals or security systems, are critical and must never lose power, while others, like streetlights or office heating, can be interrupted if necessary. The researchers assigned a financial penalty to any power shortage, making the cost of a blackout a real number in their equation, just like the price of fuel.

The core of their work is a new way of deciding where to get electricity when the sun isn't shining and the wind isn't blowing. Instead of blindly turning on the backup gas generator, the model compares the cost of the fuel needed to run it against the current price of buying electricity from the main grid. If the grid price is lower than the cost of the gas, the system buys power. If the gas is cheaper, it runs the generator. This decision happens every hour, reacting to the changing market. The model then runs thousands of simulations, testing different sizes for the connection to the main grid, ranging from a tiny wire to a massive cable capable of handling the maximum possible power the system could ever export. For each size, it calculates the total daily cost, which includes the price of fuel, the wear and tear on the battery, the fees for the connection itself, and the penalties for any power that couldn't be delivered to customers.

The results of these simulations reveal a clear pattern. As the total demand for electricity in the microgrid increases, the optimal size of the connection to the main grid also grows, but not in a straight line. For a small load of forty kilowatts, the most economical connection size is surprisingly small, just under eighteen kilowatts. As the load grows to two hundred kilowatts, the optimal connection size rises to one hundred and seventy kilowatts, which is the maximum power the system can physically export. However, the financial benefit of finding this perfect size diminishes as the system gets larger. For the smallest load scenario, optimizing the connection size reduced the total daily cost by nearly three percent. For larger loads, the savings became smaller, dropping to less than one percent, because the system was already forced to use a large connection to meet the high demand.

Crucially, the study found that if the load grows too large—specifically to two hundred and forty kilowatts—the system hits a hard wall. Even with the largest possible connection, the available solar, wind, battery, and generator power cannot meet the demand. In this scenario, the model shows that the system would inevitably fail to supply sixty-four kilowatt-hours of electricity per day, triggering the financial penalties for lost power. This indicates a limit to how much the system can grow without adding more generation or storage. The researchers also noted a limitation in their current model: the battery is managed by a simple rule that fills it when there is extra power and empties it when there is a shortage, without looking ahead to see if electricity prices will drop later that day. This means the battery might not be used as efficiently as it could be if the system could predict the future price of power.

Ultimately, this work provides a practical tool for engineers and planners. It demonstrates that treating the connection to the main grid as a flexible variable, rather than a fixed guess, can lead to genuine cost savings. By balancing the price of imported power against the cost of local generation and storage, and by accounting for the real cost of power outages, the model finds a "sweet spot" for the connection size. This approach ensures that the microgrid remains reliable for its most critical users while avoiding the waste of paying for a connection that is too large for its actual needs. The study confirms that while the savings might seem small in percentage terms, they represent a more efficient use of resources, proving that even in a complex system of wires, batteries, and generators, the right economic calculation can make a measurable difference.

Technical Summary: Optimal Grid Connection Capacity of a Hybrid Renewable Microgrid

Problem Statement
The paper addresses a specific gap in microgrid optimization: the determination of the optimal power capacity for connection to the external electrical grid. Existing literature typically treats grid connection capacity as a fixed, exogenous parameter or focuses on transmission-level expansion planning without considering the specific economic trade-offs of a microgrid with heterogeneous consumer loads. In a grid-connected microgrid with variable renewable energy (VRE), the connection capacity dictates the limits for both importing energy (to cover deficits) and exporting energy (to sell surpluses). If the connection is oversized, the microgrid owner incurs unnecessary fixed connection costs; if undersized, the system faces restrictions on import/export that may lead to energy undersupply (loss of load) or forced curtailment of renewable generation. The study aims to develop an optimization model that treats the grid connection power as a decision variable to minimize the Levelized Cost of the Microgrid (LCOM), balancing connection costs against operational costs, fuel costs, and penalties for unserved energy.

Methodology
The authors developed a daily optimization model implemented in Microsoft Excel using the Solver add-in. The model operates on an hourly time step over a 24-hour horizon and minimizes the LCOM, defined as the sum of daily costs divided by the daily energy served.

The objective function includes:

Production Costs: Daily costs for PV, Wind Power Plant (WPP), and a backup Reserve Power Generator (RES).
Storage Costs: Daily Levelized Cost of Storage (LCOS) for the Battery Energy Storage System (BESS).
Cash Flow: Costs of importing electricity, revenue from exporting electricity, and fuel costs for the backup generator.
Penalties: Costs associated with Energy Not Served (ENS), differentiated between critical and non-critical loads.
Connection Costs: Reduced daily costs of the grid connection, calculated based on a power fee, a linear component (line length), and a Capital Recovery Factor (CRF) derived from the Weighted Average Cost of Capital (WACC).

Key Operational Logic:

Priority Source Selection: The model employs an economic algorithm to determine the priority source for covering load deficits after BESS discharge. At each hour, the Day-Ahead Market (DAM) price is compared with the fuel-component-only production cost of the backup generator. The cheaper source is prioritized.
BESS Dispatch: The BESS is dispatched to cover deficits first. If a surplus exists, the BESS charges. If the BESS is full and surplus remains, the microgrid exports to the grid (limited by connection capacity). If the BESS is empty and a deficit exists, the priority source (import or generator) is activated.
Constraints: The model enforces hourly power balance, BESS state-of-charge dynamics (SOC limits and efficiency), and the technical limit of the grid connection power ( $P_{CONN}$ ), which serves as the upper bound for both import and export.
Load Classification: Consumers are divided into critical (20%) and non-critical (80%) loads, with differentiated penalty costs for unserved energy.

Case Study and Input Data
The model was applied to a hybrid microgrid with the following fixed assets:

Generation: 60 kW PV, 20 kW WPP, 40 kW Backup Gas Generator (RES).
Storage: 50 kW / 200 kWh BESS (4-hour duration, 90.3% round-trip efficiency).
Scenarios: Six scenarios were simulated with maximum consumer loads ranging from 40 kW to 240 kW (in 40 kW steps).
Data: Profiles for generation and load were based on typical July data from the Ukrainian Integrated Energy System (IES). DAM prices and fuel costs were used to drive the economic dispatch logic.

Results
The optimization determined the optimal grid connection power ( $P_{CONN}$ ) for each load scenario to minimize LCOM:

Optimal Capacity Trend: The optimal connection power increased almost proportionally with the maximum consumer load, rising from 17.9 kW (at 40 kW load) to a maximum of 170 kW (at 200 kW and 240 kW loads). The 170 kW limit corresponds to the sum of all generation and BESS discharge capabilities.
Economic Impact: Optimizing the connection power yielded a reduction in LCOM compared to a baseline where connection was sized for maximum theoretical export. The reduction was most significant at lower loads:
- 40 kW load: 2.8% reduction.
- 80 kW load: 1.1% reduction.
- 120 kW load: 0.5% reduction.
- 160 kW load: 0.2% reduction.
- At loads of 200 kW and 240 kW, the optimal connection reached the technical limit (170 kW), and further optimization of connection size was not possible; at 240 kW, the system experienced 64 kWh/day of unserved energy.
Dispatch Behavior: In lower load scenarios (e.g., 40 kW), the backup generator covered most deficits, with import used only during hours when DAM prices fell below the generator's fuel cost. In higher load scenarios (e.g., 80 kW), the generator hit its 40 kW limit, necessitating parallel import to cover the remaining deficit.

Significance and Contributions
The paper claims five primary contributions:

Unified Objective Function: It proposes the LCOM as a single objective function that integrates production, storage, cash flow, ENS penalties, and grid connection costs.
Decision Variable: It introduces grid connection power as an explicit decision variable, allowing the model to trade off connection costs against operational flexibility, unlike models where connection is fixed.
Economic Dispatch Algorithm: It proposes a specific algorithm for selecting the priority power source (import vs. generator) based on a real-time comparison of DAM prices and generator fuel costs.
Differentiated Reliability: It incorporates differentiated penalty rates for critical and non-critical loads, assessing reliability in monetary terms alongside equipment costs.
Quantified Optimization: It demonstrates that for the specific hybrid configuration studied, optimizing the connection power can reduce the microgrid's levelized cost by up to 2.8%, with the benefit diminishing as the load approaches the system's technical generation limits.

The authors note a limitation in the current BESS dispatch logic, which uses an hourly rule without reserving charge for anticipated low-price periods or guaranteeing a target state of charge for the next day. They identify the extension of the model to include predictive charge/discharge strategies as a direction for future research.

More like this