Constraints for objective function in Pymoo

Question

I have a dataset when i used lstm as a metamodel in pymoo using nsga2. I have 4 objective functions where some are negative values and some are positive values. The goal is to bring the 4 objective functions as close to 0 as possible. However I'm not sure how what to change for pymoo. My code is below. Thank you

...

class MyProblem(Problem):

def __init__(self):
    super().__init__(n_var=6,
                     n_obj=4,
                     xl=np.array([xl0[i],xl1[i],xl2[i],np.min(xl3),xl4[i],xl5[i],np.min(xl6)]),
                     xu=np.array([xl0[i]+0.01,xl1[i]+0.01,xl2[i]+0.01,np.max(xl3),xl4[i]+0.01,xl5[i]+0.01,np.max(xl6)]))

def _evaluate(self, X, out, *args, **kwargs):
f1 = model_o1.predict(X.reshape(X.shape[0], 1, X.shape[1]))
f2 = model_o2.predict(X.reshape(X.shape[0], 1, X.shape[1]))
f3 = model_o3.predict(X.reshape(X.shape[0], 1, X.shape[1]))
f4 = model_o4.predict(X.reshape(X.shape[0], 1, X.shape[1]))

    out["F"] = np.column_stack([f1, f2, f3,f4])

vectorized_problem = MyProblem()

res_interval = []

algorithm = NSGA2(
        pop_size=200,
        n_offsprings=200,
        sampling=get_sampling("real_random"),
        crossover=get_crossover("real_sbx", prob=0.8, eta=15),
        mutation=get_mutation("real_pm", eta=20),
        eliminate_duplicates=True
    )

termination1 = MultiObjectiveDefaultTermination(
        x_tol=1e-8,
        cv_tol=1e-6,
        f_tol=0.0025,
        nth_gen=5,
        n_last=30,
        n_max_gen=500,
        n_max_evals=100000)

res = minimize(vectorized_problem,
           algorithm,
           termination1,
           seed=1,
           save_history=False,
           verbose=False)

...