Abstract: In order to diagnose the single soft fault in analog circuit quantitatively,a method based on particle swarm optimization (PSO) for analog circuit with parameter tolerance is proposed. Firstly,node-voltage incremental equations based on the sensitivity analysis are built as constraints of linear programming (LP) equation. Then,through inducing the penalty function,the theory of PSO is introduced into the solution of the LP equation. In the end,after evaluating the parameters deviations,it enables us to state whether the actual parameters are within tolerance ranges or someone is faulty. Simulation results on examples illustrated that the proposed method can locate the faulty element and present its parameter deviation. The simulation results also show the effectiveness of the method in soft-fault diagnosis with tolerance in analog circuit.