Weighted cost emission dispatch optimization using GA–APO hybridization under priority sensitive scheduling for thermal power systems

Why keeping the lights on is getting more complicated

Every time you flip a switch, power plant operators are juggling two competing demands: keep electricity affordable and keep the air clean. As more rules limit pollution and demand becomes less predictable, the old way of simply picking the cheapest power plants no longer works. This study explores a new planning method that helps grid operators schedule coal or gas units over a full day while balancing cost and emissions in a flexible, transparent way.

The tug-of-war between money and smoke

For decades, the main goal of power plant scheduling was to meet demand at the lowest fuel cost. That was manageable when demand patterns were steady and environmental rules were loose. Today, utilities must also respect strict caps on greenhouse gases and other pollutants. To make matters trickier, real power plants do not behave smoothly: small changes in output can suddenly change fuel use and emissions because of internal hardware limits and different fuels. On top of this, wind, solar power, and electric vehicle charging make demand and supply less predictable. All of these factors turn what was once a simple cost-minimizing task into a multi-objective puzzle with many constraints.

A two-part search for better daily schedules

The authors propose that no single search method is good at every stage of this puzzle. Early on, the algorithm needs to roam widely to avoid getting trapped in a poor solution; later, it must make gentle, precise tweaks near the edges of technical and environmental limits. Their answer is a two-step hybrid scheme. In the first step, a Genetic Algorithm creates and evolves many candidate 24-hour schedules for three thermal generators. Each candidate respects basic plant limits and is judged by a combined score that mixes cost and emissions. This broad exploration phase homes in on promising regions of the solution space without worrying too much about fine details.

Letting a "puffin" tidy up the details

Once the Genetic Algorithm has found its best schedule, the second step begins. Here the same schedule is handed to a newer technique called Arctic Puffin Optimization. Inspired loosely by the way seabirds forage and huddle, this method is designed to make short, careful moves around a good starting point. It first looks slightly beyond the original schedule, then dives into more precise adjustments, and finally lets the best candidates pull the rest into a stable group. This behaviour is especially helpful near tight limits, where a small change in generator output can reduce cost or emissions without breaking safety or environmental rules.

Dialing priorities without rebuilding the model

A key practical feature of the method is how it treats the trade-off between money and pollution. Instead of building separate models for “cheap”, “clean”, or “in-between” operation, the authors use a single combined score with two weights: one for cost, one for emissions. By simply shifting these weights, the same framework can mimic a cost-focused policy, an emissions-focused policy, or a balanced one. In all three modes, tested on a realistic 24-hour demand pattern, the hybrid approach kept generation closely matched to demand while respecting technical and environmental limits. Compared with using the Genetic Algorithm alone, the hybrid reduced total operating cost by up to about 1.9 percent in cost-priority mode and slightly improved emissions in the environmentally focused mode, all without creating unstable or infeasible schedules.

What this means for future power planning

In plain terms, this work offers grid operators a smarter “slider” between cheap and clean energy that does not require reprogramming the system each time priorities change. While the test case uses only three conventional generators, the results show that the hybrid method gives modest but reliable gains and keeps schedules stable when policies shift. With further development for larger grids and more renewables, approaches like this could help power systems move toward lower-carbon operation while still keeping electricity bills under control.

Citation: Srinivas, C., Reddy, M.R.P., Kumar, V. et al. Weighted cost emission dispatch optimization using GA–APO hybridization under priority sensitive scheduling for thermal power systems. Sci Rep 16, 12160 (2026). https://doi.org/10.1038/s41598-026-41270-3

Keywords: economic emission dispatch, power plant scheduling, optimization algorithms, electricity cost and pollution, thermal power systems