Application of poramboku_optimizat​ion for Economic Dispatch

This program solves Economic Dispatch problem (EDP) by poramboku_optimization algorithm considering valve point loading effect.

This algorith is inspired by the A Traditional Indian Board Game played in Tamil Nadu

Real Meaning of Poramboku according to tamil culture is a fun and engaging traditional board game played in Tamil Nadu, steeped in strategy and careful planning. Below is a more detailed breakdown of the game, its rules, strategies, and variations:

Game Overview:

Number of Players: 2-4

Equipment:

• Poramboku board: A rectangular board with a grid of squares.
• Pieces: Each player has 7-10 pieces of their own color.
• Dice: One dice is used to determine the movement.

Pseudocode: Poramboku Optimization Algorithm with Velocity Constraints and Advanced Features

1. Initialize Parameters:

• Set bounds VARmin, VARmax, Vmin, Vmax.
• Define the number of dimensions dim, population size pop_size, and iterations num_iter.
• Initialize randomness factor alpha, momentum coefficient momentum, diversity threshold, and restart interval.

1. Initialize Population and Velocities:

• Generate random positions within [VARmin, VARmax].
• Generate random velocities within [Vmin, Vmax].

1. Evaluate Initial Fitness:

• Calculate the fitness of all individuals using the objective function.
• Identify the best solution and its fitness.

1. Main Optimization Loop:

• For each iteration iter from 2 to num_iter:

1. Update Velocities:

• Use momentum and adaptive randomness:velocity = momentum * velocity + alpha * random_noise
• Apply velocity constraints:velocity = clip(velocity, Vmin, Vmax)

1. Update Positions:

• Update positions based on the new velocities.population = population + velocity
• Apply position constraints:population = clip(population, VARmin, VARmax)

1. Dynamic Constraints:

• Every 10 iterations, shrink VARmin and VARmax slightly to encourage fine exploration.

1. Evaluate Fitness:

• Compute fitness for all individuals using the objective function.
• If a better fitness is found, update the global best solution.

1. Elitism:

• Replace the worst-performing individual with the global best solution.

1. Diversity Maintenance:

• If population diversity drops below a threshold, reinitialize part of the population randomly.

1. Adaptive Randomness:

• Gradually reduce the randomness factor alpha to encourage convergence.

1. Output Results:

• Store the best solution and its fitness.
• Track the fitness evolution for convergence analysis.

1. End of Algorithm:

• Return the global best solution, its fitness, and the fitness evolution.

% i dont think any body will claim for citications

fine tune the algorithm parameters, i have implemented a 13 unit system with a system demand of 1800MW and by doing parameter tuning i hope this might suit for complex optimization problem.

%source Chatgpt