•A fast assessment scheme of power angle and voltage is proposed: Firstly, the critical clearing time (CCT) was solved by the variable step size dichotomy method, and the safety boundary of power angle & voltage was constructed from the time dimension. Secondly, the results of system stability and the corresponding margin are given by the boundary of model fitting. The scheme is no need to know the fault clearing time, and only three sampling points are needed to achieve advance assessment.•A new MRSE-CNN model for assessment is proposed: The model integrates power angle and voltage through a multi task framework. At the same time, multi-scale convolution, residual module and squeeze excitation (SE) mechanism are used to improve the basic CNN, so as to maximize the assessment accuracy of the model.•An improved Huber loss function is proposed to solve the problem of different costs of being wrong in online assessment: The improved Huber loss function uses dynamic weight coefficient, so that the model can reduce the number of missing judgments and avoid the increase of misjudgments, further improving the stability of the model.•Solves the problem of accuracy degradation when the model faces unlearned scenarios, and significantly reduces the update cycle: Transfer learning is introduced into the integrated assessment of power angle and voltage for the first time. On this basis, an updated method active transfer learning (ATL) which integrating active learning and transfer learning is proposed. The adaptive update of the model in three dimensions of load, energy and topology is realized by using ATL, which greatly shortens the empty window period caused by the decline of model accuracy. Transient voltage and transient power angle are in the same time scale, both of which are the basis of safety operation of the power system. However, there are few studies on the integration of power angle and voltage at present. How to quickly realize the integrated assessment of voltage & power angle and how to shorten the update period when the model accuracy drops are the problems that need to be solved urgently. Therefore, this paper proposes a fast assessment scheme that considers both transient voltage and transient power angle. In this study, the stable boundary between the transient power angle and the transient voltage is first constructed from the time dimension by the variable step dichotomy. And then, a convolutional neural network with multiscale residual squeeze excitation (MRSE-CNN) is proposed, which can assess the voltage and power angle without post-fault-clearance data. It only takes three sampling points to accurately learn the mapping between the input feature and the stable boundary. At the same time, the results of whether the transient voltage and the transient power angle are stable, and the corresponding margin are output. By introducing the improved Huber loss function to dynamically adjust the cost of misjudgment and missing judgment, the reliability of the model is further enhanced. In the online application, a pool-based active transfer learning is proposed for the unlearned scenarios under load, topology, and renewable energy, which greatly reduces the adaptive update time of the model in unlearned scenarios. The model is verified in the improved IEEE 39 bus system and provincial interconnection system. It shows that the proposed method can quickly and accurately realize the integrated adaptive assessment of transient power angle and voltage in any scenarios.