Abstract: The present invention provides a high-precision identification method and system for substations, including building a Mask RCNN objection recognition network model based on convolutional neural networks; inputting acquired image information of a object into the Mask RCNN object recognition network model for preliminary recognition and outputting a recognition result of the object; using an information entropy to create a semantic decision tree and correcting the recognition result of the object according to a principle of relative correlation between different objects and outputting a final recognition decision result; reading the recognition decision result to obtain a true type of the object to be recognized. The present invention greatly improves the accuracy of image recognition of substations, and has a positive role in the research and development of automatic inspection equipment for inspection robots.

Abstract: The present disclosure provide a method for identifying power equipment targets based on human-level concept learning, including: creating a dataset of power equipment images, and annotating power equipment in power equipment images; training neural network and Bayesian network with the annotated dataset and respectively acquire identification results and conditional probabilities; calculating probabilities of unions with the conditional probabilities; and filtering the identification result corresponding to the highest probability of the union as identification result of the dataset of the power equipment images and complete the identification of the power equipment. The present disclosure combines Mask R-CNN and probabilistic graphical model. The bottom layer uses Mask R-CNN, and the top layer uses Bayesian network to train in identifying power equipment images, so that a small amount of data samples can achieve good recognition, which improved the performance of Mask R-CNN model.

Abstract: The present invention provides a high-precision identification method and system for substations, including building a Mask RCNN objection recognition network model based on convolutional neural networks; inputting acquired image information of a object into the Mask RCNN object recognition network model for preliminary recognition and outputting a recognition result of the object; using an information entropy to create a semantic decision tree and correcting the recognition result of the object according to a principle of relative correlation between different objects and outputting a final recognition decision result; reading the recognition decision result to obtain a true type of the object to be recognized.

Abstract: A heterogeneous image registration method includes: performing edge detection on collected images, in combination with a curvature scale space strategy to extract contour curved segments in an edge image. Implementing a feature point detection strategy based on global and local curvature detecting feature points in the contour curved segments, and obtaining the nearest minimum local curvature of the feature points pointing to starting and end points of the contour, respectively. Calculating the number of neighborhood sampling points and neighborhood auxiliary feature points of neighborhoods on both edges of each of the feature points according to the nearest minimum local curvature. Using neighborhood auxiliary feature points and feature points to form a feature triangle, calculating an angle bisector vector and a main direction corresponding to the feature point in the feature triangle.

Abstract: The present disclosure provide a method for identifying power equipment targets based on human-level concept learning, including: creating a dataset of power equipment images, and annotating power equipment in power equipment images; training neural network and Bayesian network with the annotated dataset and respectively acquire identification results and conditional probabilities; calculating probabilities of unions with the conditional probabilities; and filtering the identification result corresponding to the highest probability of the union as identification result of the dataset of the power equipment images and complete the identification of the power equipment. The present disclosure combines Mask R-CNN and probabilistic graphical model. The bottom layer uses Mask R-CNN, and the top layer uses Bayesian network to train in identifying power equipment images, so that a small amount of data samples can achieve good recognition, which improved the performance of Mask R-CNN model.

Abstract: Power distribution network early failure detection method comprising: a waveform is decomposed into an approximation part and detail parts by wavelet transform, wherein the approximation part is referred to as an approximate shape primitive, and the detail parts are referred to as distortion primitives; the distortion primitives are divided into three primitives including harmonics, pulses and other distortions based on extreme points; characteristics of the primitives and time relationships between the primitives are extracted; probability distribution of the waveform is constructed according to the characteristics of the primitives and the time relationships between the primitives; and a judgment result of the waveform is obtained according to probability distribution of waveforms of different types.

Abstract: Power distribution network early failure detection method comprising: a waveform is decomposed into an approximation part and detail parts by wavelet transform, wherein the approximation part is referred to as an approximate shape primitive, and the detail parts are referred to as distortion primitives; the distortion primitives are divided into three primitives including harmonics, pulses and other distortions based on extreme points; characteristics of the primitives and time relationships between the primitives are extracted; probability distribution of the waveform is constructed according to the characteristics of the primitives and the time relationships between the primitives; and a judgment result of the waveform is obtained according to probability distribution of waveforms of different types.

Abstract: Nonlinear control method for the micro-grid inverter with anti-disturbance. By generating reference currents that satisfy specific active and reactive power command under various working conditions, and introducing a nonlinear control method based on Lyapunov function to control the inverter, fast and accurate tracking of the generated reference signals is realized. The method realizes effective decoupling control of active power and reactive power. The system has high dynamic response and good robustness. Besides, the control structure of the method is simple and easy to implement, and the synchronous control link and the additional voltage and current regulator are omitted. The method realizes fast and accurate power exchange and stable power transmission between the inverter and the grid in the micro-grid under various working conditions, and provides a guarantee for improving the energy management efficiency within the micro-grid.

Abstract: Nonlinear control method for the micro-grid inverter with anti-disturbance. By generating reference currents that satisfy specific active and reactive power command under various working conditions, and introducing a nonlinear control method based on Lyapunov function to control the inverter, fast and accurate tracking of the generated reference signals is realized. The method realizes effective decoupling control of active power and reactive power. The system has high dynamic response and good robustness. Besides, the control structure of the method is simple and easy to implement, and the synchronous control link and the additional voltage and current regulator are omitted. The method realizes fast and accurate power exchange and stable power transmission between the inverter and the grid in the micro-grid under various working conditions, and provides a guarantee for improving the energy management efficiency within the micro-grid.

Abstract: A parallel connection method and a device for multi-channel partial discharge (PD) signals have input sources of pre-processed and amplified signals received by multiple PD sensors. Then isolation and processing are provided to the signals to ensure that each signal is able to be self-triggered and unidirectional transmitted. Finally, processed signals of each channel are outputted. For ultra-high frequency (UHF) PD signal, the delay transmission is realized by a surface acoustic wave (SAW) delay line method. The present invention makes the amplified multi-channel PD signals self-trigger and unidirectional transmit though isolation processing and then realizes parallel connection without interference. The method also decreases signal transmission loss. Meanwhile, all the detected PD signals transmit on one signal bus through parallel connection method, so as to use only one set of acquisition unit to monitor and acquire the signals received by multiple sensors, which obviously lowers the system cost.

Abstract: A maximum power output circuit for EHC and its design method are presented. The circuit is comprised of a magnetic core, that is, a primary coil and a secondary coil, with a load resistor and a capacitor parallel connected at the two ends of the secondary coil. The circuit enables the EHC to be always working at the maximum power output, thus realizing maximum power output of the energy harvesting power source.

Abstract: A method for detecting and locating faulty line on the distribution network circuit based on full waveform information, which uses the current data on the fault occurrence and whole-process operation of the compensation device to effectively solve such common problems as low fault current, poor reliability and low sensitivity in case of single-phase grounding fault to the low-current system through precise GPS synchronization. It improves the sensitivity and reliability of the grounding fault detection, which does not interfere with the system and is applicable to high-resistance grounding.

Abstract: A feedforward voltage series compensator based on complimentary use of wind, solar, and electric power comprising a controller, a rectifier unit, an H bridge inverter, a series transformer, a wind-power DC voltage sensor, a wind-power DC current sensor, an AC voltage transducer, a DC boost unit, a PV DC voltage sensor, a PV DC current sensor, and a grid-connected inverter. The compensator makes use of wind-electric and photovoltaic-electric complimentary interactions to solve the traditional energy issue for series compensator, and as the grid-connected inverter is feedforward, there is extra capacity for the series transformer and the series complimentary inverter unit, and hence it enjoys the feedforward and quick voltage complimentary characteristics of the wind and solar power generation.

Abstract: A method for designing cascaded multi-level inverters with minimization of large-scale voltage distortion, based on KKT (Karush-Kuhn-Tucker) conditions and with simplified computation of conduction angles, simplifies the computation process, and is conducive to on-line calculation. Meanwhile, its fundamental voltage is adaptive, minimization of total harmonic is realized for cascaded multi-level inverters at high-voltage, and voltage power quality at grid connected nodes is improved.

Abstract: A maximum power output circuit for EHC and its design method are presented. The circuit is comprised of a magnetic core, that is, a primary coil and a secondary coil, with a load resistor and a capacitor parallel connected at the two ends of the secondary coil. The circuit enables the EHC to be always working at the maximum power output, s thus realizing maximum power output of the energy harvesting power source.

Abstract: A method for identifying a lightning fault of an overhead transmission line (OTL) comprising determining a polarity of a travelling wave of each of three ABC phases subsequent to a single phase outage of the OTL. If the polarity of each phase is the same, the outage is a lightning fault and its type a back flashover; if not the same, proceed to the second step to determine a current change rate R of the fault phase. If R is larger than a threshold value, the outage is a lightning fault and its type a shielding failure; otherwise, the outage is a single-phase grounding fault. A system for identifying the lightning fault which comprises successively at least a group of fault detectors, a wireless communication module, a remote monitoring master station which adopts the above method to determine the fault type of the overhead transmission line.

Abstract: A method for detecting and locating faulty line on the distribution network circuit based on full waveform information, which uses the current data on the fault occurrence and whole-process operation of the compensation device to effectively solve such common problems as low fault current, poor reliability and low sensitivity in case of single-phase grounding fault to the low-current system through precise GPS synchronization. It improves the sensitivity and reliability of the grounding fault detection, which does not interfere with the system and is applicable to high-resistance grounding.

Abstract: A method for designing cascaded multi-level inverters with minimization of large-scale voltage distortion, based on KKT (Karush-Kuhn-Tucker) conditions and with simplified computation of conduction angles, simplifies the computation process, and is conducive to on-line calculation. Meanwhile, its fundamental voltage is adaptive, minimization of total harmonic is realized for cascaded multi-level inverters at high-voltage, and voltage power quality at grid connected nodes is improved.

Abstract: A parallel connection method and a device for multi-channel partial discharge (PD) signals have input sources of pre-processed and amplified signals received by multiple PD sensors. Then isolation and processing are provided to the signals to ensure that each signal is able to be self-triggered and unidirectional transmitted. Finally, processed signals of each channel are outputted. For ultra-high frequency (UHF) PD signal, the delay transmission is realized by a surface acoustic wave (SAW) delay line method. The present invention makes the amplified multi-channel PD signals self-trigger and unidirectional transmit though isolation processing and then realizes parallel connection without interference. The method also decreases signal transmission loss. Meanwhile, all the detected PD signals transmit on one signal bus through parallel connection method, so as to use only one set of acquisition unit to monitor and acquire the signals received by multiple sensors, which obviously lowers the system cost.

Abstract: A feedforward voltage series compensator based on complimentary use of wind, solar, and electric power comprising a controller, a rectifier unit, an H bridge inverter, a series transformer, a wind-power DC voltage sensor, a wind-power DC current sensor, an AC voltage transducer, a DC boost unit, a PV DC voltage sensor, a PV DC current sensor, and a grid-connected inverter. The compensator makes use of wind-electric and photovoltaic-electric complimentary interactions to solve the traditional energy issue for series compensator, and as the grid-connected inverter is feedforward, there is extra capacity for the series transformer and the series complimentary inverter unit, and hence it enjoys the feedforward and quick voltage complimentary characteristics of the wind and solar power generation.