How to find a solution for a non-linear equation in python using ifelse defined function

I am encountering a problem solving a nonlinea equation using nsolve from the python simpy package, this problem was easily solved in R, below the code I use in R and what I am trying in python:

R :

library(dplyr)
library(plyr)
sol = adply(df,1, summarize,
             solution_1 = uniroot.all(function(x)(eval(parse(text=as.character(FX),df))),lower = -10000, upper = 10000, tol = 0.00001))

In Python:

from sympy import Symbol, solve, nsolve
from sympy.parsing.sympy_parser import parse_expr
import pandas as pd

def ifelse(cond,yes,no):
    if (cond):
        return(yes)
    else: 
        return (no)   

    data = {'A': [10,20,30],
        'B': [20,10,40],
        'FX': ["ifelse(A+B+x<0,0,A+B-x)","ifelse(A+B+x<0,0,A*B-x)","ifelse(A+B+x,A*B-x,A+B-x)"]}
df = pd.DataFrame(data)

x = Symbol("x", real=True)
cols= df.columns
print(cols)
for _, row in df.iterrows():
    print(nsolve(parse_expr(row["FX"],local_dict=dict({c:row[c] for c in cols}, **{'x':x,"ifelse":ifelse}) ),x,0))

The expected solution is: [30,200,30]

But the following error was returned:

TypeError: cannot determine truth value of Relational

Because the code in python can't evaluate A+B-x, I don't know how to change the code so that it evaluates the expression with the initial value of x.

Any ideas on how to work around this problem?