Machine learning driven global optimisation framework for analog circuit design

We propose a machine learning-driven optimisation framework for analog circuit design in this paper. Machine learning based global offline surrogate models, with the circuit design parameters as the input, are built in the design space for the analog circuits under study and are used to guide the optimisation algorithm towards an optimal circuit design, resulting in faster convergence and reduced number of spice simulations. Multi-layer perceptron and random forest regressors are employed to predict the required design specifications of the analog circuit. Multi-layer perceptron classifiers are used to predict the saturation condition of each transistor in the circuit. We validate the proposed framework using three circuit topologies—a bandgap reference, a folded cascode operational amplifier, and a two-stage operational amplifier. The simulation results show better optimum values and lower standard deviations for fitness functions after convergence, with a reduction in spice calls by 56%, 59%, and 83% when compared with standard approaches in the three test cases considered in the study.