Wavelet network current instantaneous value detection method applied to megawatt converter
A technology of current instantaneous value and detection method, which is applied in the direction of measurement using digital measurement technology, which can solve the problems of current loop dynamic performance deterioration, current phase lag, system instability, etc.
Active Publication Date: 2013-10-09
GUANGDONG ZHICHENG CHAMPION GROUP
View PDF2 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Compared with analog control, digital control has its advantages, but due to the limitation of sampling and A / D conversion time of digital control system and the influence of noise, it is difficult to meet real-time detection and feedback
In order to suppress the interference of high-frequency noise in the sampling circuit and at the same time make the sampling signal closer to the real value of the grid-side current, the current general-purpose sampling algorithm is mostly to add a Low-pass filter, but the low-pass filter will cause the detected current phase to lag behind the actual current waveform, the dynamic performance of the entire current loop will deteriorate, and even lead to system instability
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0030] The present invention adopts compact wavelet neural network [4] , let x=[x 1 ,x 2 ,...,x n ] T is the input vector, that is, the unfiltered current sampling value, y=[y 1 ,y 2 ,...,y m ] T is the output vector, that is, the filtered current output value, and the weight from the output layer to the hidden layer is w ij , the weight from the hidden layer to the input layer is w jk , the scaling parameter and translation parameter of the middle hidden layer are respectively a j and b j , if the number of neurons in the hidden layer is N, then i=1,2,...,m, j=1,2,...,N, k=1,2,...,n, the realization of the network equation can be expressed as:
[0031] y i ( t ) = f i ( t ) = f [ Σ ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a wavelet network current instantaneous value detection method applied to a megawatt converter. The method comprises the steps of using converter current sampling value as input vectors of a wavelet neural network, using current output value of a filtered converter as output vectors of the wavelet neural network, selecting Morlet wavelet functions, initializing the extension parameter and the translation parameter of a middle hidden layer of the wavelet neural network through a method of calculating the gravity center of samples, utilizing the extension parameter and the translation parameter to adjust network parameters in a gradient descent method, and obtaining the current output value of the filtered converter. The method does not need to sample three phase grid-side currents of the converter in real time so as to ensure the stability and the reliability of a converter system, improve the accuracy of current instantaneous value detection, and improve the dynamic performance of a current loop circuit. The method is convenient to realize, little in operation amount, good in stability and interference resistance, and ideal.
Description
technical field [0001] The invention relates to a power electronic intelligent detection technology, in particular to a wavelet network current instantaneous value detection method applied to a megawatt converter. Background technique [0002] In the field of power electronics applications, since the converter can conveniently realize the two-way flow of energy, and can obtain an approximately sinusoidal current waveform on the grid side, the power factor is close to 1, compared with the traditional rectification technology, the pollution to the grid is very small. It can realize "green power conversion", and has been more and more widely used in reactive power compensation, active power filtering, unified power flow control, UPS, and utilization of renewable energy such as solar energy and wind energy. [0003] The control algorithm based on the dq transformation of the rotating coordinate system is the most widely used control method for three-phase megawatt converters. I...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G01R19/25
Inventor 戴瑜兴郜克存郑崇伟全惠敏张正江
Owner GUANGDONG ZHICHENG CHAMPION GROUP