This paper introduces a generalized averaging method for power conversion circuits, particularly resonant type converters, which extends the applicability of state-space averaging beyond the "small ripple" condition. The method is based on a time-dependent Fourier series representation of a waveform over a sliding window, allowing for the analysis of arbitrary waveform behaviors. Key aspects include: 1. **State-Space Averaging Limitations**: Traditional state-space averaging is limited to circuits satisfying the "small ripple" condition, which requires linear ripple approximation and commutativity of certain vector fields. 2. **Generalized Averaging Scheme**: The proposed method uses a Fourier series representation of a waveform over a sliding window, capturing both sinusoidal and slowly varying behaviors. This approach is more flexible and can handle a broader range of circuits, including resonant converters. 3. **Application to Resonant Converters**: The method is applied to resonant converters, which exhibit oscillatory behavior, by retaining appropriate Fourier coefficients. Examples include a series resonant converter with a voltage source load and a DC-DC series resonant converter with a capacitor load. 4. **Refinement with Higher Order Terms**: The refined model includes higher-order terms in the Fourier series expansion, improving the accuracy of the averaged model, especially in the presence of significant ripple. 5. **Conclusion**: The generalized averaging method offers a more robust framework for analyzing and designing power electronic circuits, particularly those with resonant behavior, and can be useful in both simulation and control design. The paper also discusses the handling of time-varying frequencies and provides detailed mathematical derivations and examples to illustrate the effectiveness of the proposed method.This paper introduces a generalized averaging method for power conversion circuits, particularly resonant type converters, which extends the applicability of state-space averaging beyond the "small ripple" condition. The method is based on a time-dependent Fourier series representation of a waveform over a sliding window, allowing for the analysis of arbitrary waveform behaviors. Key aspects include: 1. **State-Space Averaging Limitations**: Traditional state-space averaging is limited to circuits satisfying the "small ripple" condition, which requires linear ripple approximation and commutativity of certain vector fields. 2. **Generalized Averaging Scheme**: The proposed method uses a Fourier series representation of a waveform over a sliding window, capturing both sinusoidal and slowly varying behaviors. This approach is more flexible and can handle a broader range of circuits, including resonant converters. 3. **Application to Resonant Converters**: The method is applied to resonant converters, which exhibit oscillatory behavior, by retaining appropriate Fourier coefficients. Examples include a series resonant converter with a voltage source load and a DC-DC series resonant converter with a capacitor load. 4. **Refinement with Higher Order Terms**: The refined model includes higher-order terms in the Fourier series expansion, improving the accuracy of the averaged model, especially in the presence of significant ripple. 5. **Conclusion**: The generalized averaging method offers a more robust framework for analyzing and designing power electronic circuits, particularly those with resonant behavior, and can be useful in both simulation and control design. The paper also discusses the handling of time-varying frequencies and provides detailed mathematical derivations and examples to illustrate the effectiveness of the proposed method.