optivgi.scm.pulp_numerical_algorithm¶

Provides an SCM algorithm implementation using the PuLP linear programming library.

This algorithm formulates the charging schedule optimization as a linear program and uses a solver (like CBC, GLPK, CPLEX, etc.) interfaced through PuLP to find an optimal solution based on the defined objective and constraints.

class optivgi.scm.pulp_numerical_algorithm.PulpNumericalAlgorithm(evs, peak_power_demand, now)¶

Bases: Algorithm

SCM Algorithm using PuLP for Linear Programming Optimization.

This algorithm aims to maximize the percentage of requested energy delivered across all EVs, while respecting individual EV power limits, arrival/departure times, and aggregate peak power constraints for the group. It also includes terms to encourage utilizing available peak power and minimizing power fluctuations.

Parameters:

evs (list[EV])
peak_power_demand (list[float])
now (datetime)

calculate()¶

Formulates and solves the charging optimization problem using PuLP.

Steps: :rtype: None

Create an LpProblem instance.
Define decision variables: - ev_vars: Power allocated to each EV at each time step. - ev_vars_diff: Absolute difference in power between consecutive time steps for each EV. - percentage: Minimum percentage of energy charged across all EVs (to be maximized).
Define the objective function: Maximize percentage, scaled, plus a term for peak power utilization, minus a penalty for power fluctuations (ev_vars_diff).
Define constraints: - Power difference calculation (using absolute value formulation). - Energy charged for each EV must be >= percentage * requested energy. - Power limits (min/max) for each EV during its connected time. - Zero power before arrival and after departure for each EV. - Aggregate power demand constraint for each time step.
Solve the linear programming problem using the configured solver (default CBC).
Extract the results (optimal power values) from ev_vars and store them in the ev.power list for each EV object.
Log summary information.