36 results about "Residual generator" patented technology

Embodiments of the present invention provide a pipeline ADC front-end sampling structure that provides a continuous time input signal to a flash comparator for sampling. By providing a continuous time input signal to the flash comparator, no delay is introduced from the need to transfer a DC charge representing the sampled input to the flash comparator. Matching sampling networks in the residual generator and the flash comparator are avoided due to the high bandwidth response requirements of the residual generator and the flash comparator when operating on high frequency input signals.

The invention discloses a robust detection method for intermittent sensor faults in the brake system of a train set, which belongs to the field of signal processing. The method includes: the steps of establishing a state space model of the brake system of a train set, and the steps of designing a robust residual generator , design fault detection strategy steps. The invention effectively guarantees the actual application requirements of fault detection of intermittent passenger sensors of EMUs, can detect the occurrence and disappearance of faults of intermittent passenger sensors in real time on-line in the case of random uncertainties in the system, and achieves automatic and intelligent detection.

The invention discloses an optimal gyroscope combination selection method based on residual error generators. The method comprises steps as follows: (1), four gyroscopic instruments of four standby gyroscopic assemblies including three upright gyroscopic assemblies and one oblique gyroscopic assembly are divided into four groups of gyroscope combinations; (2), the three-axis angular speed of each satellite is calculated according to measurement values of the four groups of gyroscopic combinations according to mounting matrixes of the standby gyroscopic assemblies, and three-axis angular speeds of the four satellites are obtained; (3), the three-axis attitude provided by a star sensor and the three-axis angular speeds, calculated by the four groups of gyroscopic combinations, of the satellites are used for forming the residual error generators, and four attitude residual errors are obtained totally; (4), an evaluation function of each attitude residual error is solved, the four evaluation functions are sequenced according to numerical magnitudes, and the smallest group is the optimal gyroscope combination. The gyroscope combination with the optimal performance is dynamically selected in orbit in real time, redundancy of the gyroscopic assemblies on the satellite is effectively used, angular speed information with higher accuracy can be provided for an attitude control system, and the method has certain fault-tolerant capacity for gyroscopic faults.

The invention relates to the technical field of equipment fault monitoring, and particularly discloses an equipment fault warning and state monitoring method. The equipment fault warning and state monitoring method is characterized by including two processes of model building and model operation; the process of model building includes the steps of acquiring training data, performing data preprocessing operation on the training data, then adopting a nonparametric learning algorithm to select a memory matrix, training a residual generator and acquiring a residual threshold of each parameter; the process of model operation includes the steps of acquiring real-time data, performing data preprocessing operation on the real-time data, then calculating all parameter residuals of the real-time data, analyzing the residuals to judge whether the equipment state is normal or not, and further positioning fault causes. The method has the advantages of data-driven method based universality, robustness and high adaptive capability, the shortcoming that warning results are difficult to analyze and explain is avoided, and additionally accuracy and efficiency of fault warning are both improved due to introduction of the nonparametric learning algorithm.

The invention discloses a distributed security state reconstruction method based on a double-layer dynamic switching observer, which includes the following steps: constructing a dynamic model of its sensing channel after being attacked according to a multi-agent system; combining specific multi-agent The body system model builds a double-layer observer for each multi-agent agent, constructs a suitable observation communication topology, and designs a corresponding residual generator; analyzes and generates dynamic information of the residual threshold for the observation model, and checks the relationship between each residual signal and The size between the thresholds dynamically switches the communication topology between observers according to the size of the comparison, and performs new data communication interactions; combines its own observation data and received neighbor observation information to iteratively update to generate new observation data, and Whether the residual signal is greater than the current threshold is used as a criterion for judging whether the corresponding communication channel is under attack. The invention can ensure correct identification of all transmission channels subjected to malicious attacks and safe reconstruction of the real state of the system.

The invention relates to a fault separability evaluation method of an unmanned aerial vehicle flight control system. In order to realize the optimal design of the UAV flight control system, it is necessary to evaluate the difficulty of system fault separation in the UAV design stage. The present invention establishes the UAV flight control system model according to the UAV flight control system principle and fault type; uses the equivalent space method to construct the fault diagnosis residual generator; uses the cosine distance of the residual vector caused by different faults to establish the fault separation condition ; Using the similarity of the residual vector to quantitatively evaluate the difficulty of fault separation, combining the distance similarity and direction similarity in pattern recognition, an improved quantitative evaluation index of fault separability is proposed. The evaluation method can comprehensively evaluate the difference in distance and direction of residual vector caused by different faults, and provide a reference for the design of UAV system and fault separation algorithm.

The invention discloses a propulsive power dynamic self-adaptive control method. The method comprises the steps that the main distribution board busbar switch state of a power management system, and the network state and the load factor of a diesel generator set are acquired and sent to a controller; and the controller synthesizes the information and the current output power of an inverter to carry out dynamic power limiting and fast power limiting on the inverter. In the dynamic power limiting process, the controller monitors the load factor of the diesel generator set in real time, and constantly adjusts the limit of the output power of the inverter to ensure safe operation of the diesel generator set. In a fast power limiting process, the controller monitors the load factor of the diesel generator set and the corresponding main switch state in real time. When the corresponding main switch is turned off due to the failed generator set, the controller can quickly limit the output power of the inverter to ensure safe operation of remaining generator sets. The propulsive power dynamic self-adaptive control method provided by the invention has the advantages of simple system structure and high automation degree, and can ensure safe and reliable sailing of an electric propulsion ship in all cases.

The invention discloses a fault detection method for a servo system based on an observer and relates to a fault detection method for a servo system, which belongs to the technical field of electromechanical failure detection. The method comprises the following steps of 1, modeling a dual-motor servo system as a linear model with an unknown input, an actuator fault and a sensor failure; 2, generating a residual signal by a Kalman filtering observer-based residual generator and conducting the fault detection and logic judgment; 3, through a norm-based residual evaluation function and a fixed threshold value, conducting the fault detection and logical judgment to detect a fault signal, and completing the fault detection of the dual-motor servo system. According to the technical scheme of the invention, the Kalman filtering observer-based residual generator reduces the influence of the external interference on the dual-motor servo system. Meanwhile, based on the norm-based residual evaluation function and the fixed threshold-value logical judgment, the fault signal of the system can be effectively detected.

The invention relates to a dynamic compensation control method and system for a microgrid based on a distributed residual error generator. The method comprises the steps of establishing a state spaceequation of the distributed residual error generator according to an obtained state space equation of the ith inverter when N inverters are connected in parallel, and obtaining a control model based on the distributed residual error generator; converting the control model based on the distributed residual error generator into a model taking disturbance di as input, calculating a dynamic compensation controller Q(s) to obtain a dynamic compensation control model based on the distributed residual error generator, and adding the dynamic compensation control model into droop control to generate acompensation control strategy so as to perform dynamic compensation control on the microgrid. The plug and play of the inverter and the load can be realized according to the invention.

The invention relates to the field of industrial process monitoring and fault diagnosis, and discloses a fault detection method and system of a non-Gaussian industrial process in order to online detect the non-Gaussian industrial process conveniently and practically. The method includes: step 1, selecting a certain quantity of normal historical data sets, performing mean value preprocessing on the data sets, and establishing a residual generator by employing a canonical correlation analysis method; step 2, determining a corresponding threshold of the corresponding residual generator by employing a Monte Carlo method; and step 3, performing real-time online detection on industrial process data according to the determined threshold of the corresponding residual generator.