Magnetically Suspended Control and Sensitive Gyro (MSCSG) is an attitude measurement and attitude control instrument. In order to realize high precision measurement of attitude angle velocity of spacecraft under the condition of output torque, a method of magnetically suspended rotor tilt modulation is proposed based on MSCSG. In the paper, the attitude angular velocity measurement principle of MSCSG is analyzed, and a conclusion is drawn that there is a contradiction between the output torque and the measurement accuracy. According to the working characteristics of MSCSG, the tilt modulation method of magnetically suspended rotor is proposed. By actively controlling the angular momentum tilt of the rotor, the method can measure the attitude angular velocity under the condition of output torque. By analyzing the measurement error, the method can effectively compensate the constant error and improve the measurement accuracy of attitude angular velocity. The experimental results show that tilt modulation can improve the measurement accuracy of attitude angle rate by 10 times compared with no tilt modulation. Simulation and experiment verify the correctness and superiority of the proposed method. The proposed method provides theoretical support for the integration of MSCSG attitude measurement and control.

Recently, magnetic bearings have been widely used in inertial actuators of spacecraft. Aiming at the problem of unbalanced vibration in magnetically suspended system, the research status of active vibration control method is described in detail. Firstly, the structure and working principle of magnetically suspended system are introduced, and the causes of unbalanced vibration are expounded. Then, the research status of active vibration control methods at home and abroad is introduced from two aspects of synchronous vibration and harmonic vibration, and the advantages and disadvantages of each control method are summarized. Finally, the future development trend of active vibration control technology for magnetically suspended system is prospected.

To effectively recognize the incipient potential faults caused by degradation of multiple components in DC-DC converters, a fault diagnosis method that involves multi-dimensional feature fusion and sensitive feature extraction is proposed. Firstly, the time-domain statistical characteristics of fault and normal samples are extracted. The KL divergence and normalized kurtosis of intrinsic mode functions (IMFs) between them are calculated by empirical mode decomposition (EMD). In order to further improve the feature discrimination, a sensitive feature extraction method based on Mahalanobis distance (SFMD) is designed to screen out the key features. Finally, the sensitive features are used to construct the SA-LSSVM (Simulated annealing-Least squares support vector machine) model to realize the fault diagnosis. The accuracy of fault diagnosis in simulation and hardware experiment are 99.61% and 97.93% respectively. Compared with other fault diagnosis and feature selection methods, the proposed method still has higher accuracy and better engineering practicability.

In order to eliminate the influence of synchronous harmonic current caused by mass imbalance of magnetic bearing on attitude angular velocity measurement accuracy, a notch filter which can adjust the notch depth and center frequency is proposed. Firstly, the influence of synchronous harmonic current on the measurement accuracy of magnetically suspended control and sensitive gyroscope(MSCSG) is analyzed, and then a notch filter is designed which can adjust the notch depth and center frequency, and the measurement accuracy and stability of the system after eliminating synchronous harmonic current are analyzed. Simulation results show that the proposed method can effectively suppress the unbalance vibration torque and eliminate synchronous harmonic current. Finally, the correctness and effectiveness of the proposed method are verified by experiments. Experimental results show that the proposed method can reduce the vibration torque by 50% and improve the measurement accuracy of attitude angular velocity significantly.

DC-DC converter is the vital part of the power system, and its fault can cause the failure of complex electronic equipment. Therefore, the timely and accurate fault diagnosis of DC-DC converter is particularly important. This paper proposes a DC-DC converter fault diagnosis method based on improved SE-ResNet algorithm. First, in order to extract all information about the signals, we transform the one-dimensional signal into Gramian angular differential field (GADF) images. Then, the Squeeze-and-Excitation (SE) network is added to ResNet-50, combined with h-swish function and label smoothing cross entropy loss function (LSCE) to improve the model performance. In addition, simplify the redundancy layer of the network and achieve lightweight to improve the training efficiency. In the end, the logsoftmax is used for evaluating the effectiveness of the proposed method. The simulation experimental accuracy is up to 99.3220%,which is higher than other five classical deep learning algorithms. The hardware experiment also denotes the engineering practicability of the proposed method.

In order to solve the problem of poor precision in traditional modeling of DC-DC converter, an adaptive update method of digital twin based on mechanism-data drive is proposed in this paper to build a high-fidelity digital twin model. Firstly, a digital twin modeling method based on hybrid drive of mechanism-data driven model is proposed. The data-driven model based on Gate Recurrent Unit (GRU) is constructed to compensate the mechanism model in parallel. The virtual-real error can be reduced to 0.11%, which effectively improves the precision of digital twin modeling. Secondly, in order to solve the problem of model tracking accuracy reduction caused by component degradation, parameter identification of digital twin model based on Genetic Algorithm (GA) is proposed. The relative error of the identification results is close to 5%, which can realize the adaptive updating of the model. Finally, the fault prediction model of CNN-LSTM (Convolutional Neural Networks-Long short-term memory) based on high fidelity digital twin model is constructed to realize degradation trend prediction and fault warning of electrolytic capacitors.