Overview

The objFuncs package is designed to provide users with an intuitive I/O interface to the EPICS system, specifically tailored for the Facility for Rare Isotope Beams (FRIB) driver LINAC and REA. It streamlines the construction of objective functions by scalarizing multi-objective optimization using heuristics that incorporate tolerances for each objective. Additionally, it includes wrapper functions for evaluating these objective functions on the machine.

Key Features

1. Easy I/O Interface:

   • objFuncs/construct_machineIO.py
   • Simplifies interaction with the FRIB machine.
   • Provides streamlined methods for data input and output.

2. Objective Function Construction:

   • Facilitates quick and easy creation of objective functions for the EPICS system.
   • Uses scalarization techniques to handle multi-objective optimization based on simple heuristics regarding tolerance levels for different objectives.

3. Wrapper Functions:

   • Includes wrapper functions to facilitate the evaluation of objective functions directly on the EPICS system to be used in general optimization routine.

Example Usage

import objFuncs
from objFuncs import objFuncGoals

objFuncs._global_machineIO._test = False
objFuncs._global_machineIO._fetch_data_time_span = timespan_for_average
objFuncs._global_machineIO._ensure_set_timewait_after_ramp = additional_wait_after_powersupply_ramp

obj = objFuncGoals(
    decision_CSETs=['FE_LEBT:PSC2_D0001:I_CSET','FE_LEBT:PSC1_D001:I_CSET',
    decision_min = [-2,-2],
    decision_max = [ 2, 2],
    decision_tols = [0.1, 0.1],
    objective_goal   = {'FE_LEBT:FC_D0002:PKAVG_RD': {'more than': sourceFC}},
    objective_weight = {'FE_LEBT:FC_D0002:PKAVG_RD': 1},
    objective_norm   = {'FE_LEBT:FC_D0002:PKAVG_RD': 0.2*sourceFC},
)

# set corrector to zero and read objective
x = [0, 0]
obj_val = obj(x)