With the accelerated construction of the smart grid, new energy sources such as photovoltaic and wind power are connected to the grid. In addition to power frequency, the current signal of power grid also includes several DC signals, as well as medium-high and high-frequency transient signals. Traditional current sensors for power grids are bulky, have a narrow measurement range, and cannot measure both AC and DC at the same time. Therefore, this paper designs a non-intrusive, AC-DC wide-bandwidth current sensor based on the composite measurement principle. The proposed composite current detection scheme combines two different isolation detection technologies, namely tunneling reluctance and the Rogowski coil. These two current sensing techniques are complementary (tunneling magnetoresistive sensors have good low-frequency characteristics and Rogowski coils have good high-frequency characteristics, allowing for a wide detection bandwidth). Through theoretical and simulation analysis, the feasibility of the composite measurement scheme was verified. The prototype of composite current sensor was developed. The DC and AC transmission characteristics of the sensor prototype were measured, and the sensitivity and linearity were 11.96 mV/A, 1.14%, respectively. Finally, the sweep current method and pulse current method experiments prove that the designed composite current sensor can realize the current measurement from DC to 17 MHz.

According to the requirements of weak current measurement in power grid, a weak current sensor with anti-low frequency interference ability is developed. The sensor adopts the principle of fluxgate detection and adds a magnetic ring on the original basis. The structure of the magnetic ring is simulated using comsol to further improve detection sensitivity. In order to solve the problem that the electromagnetic current sensor is vulnerable to the interference of geomagnetic field and power frequency magnetic field in weak current measurement, a magnetic shielding method with low cost is selected, and the shielding shell structure is designed using a finite element analysis method. The experimental results show that the minimum measurable current is 1 mA, the measurement range is 1 mA–1 A, and the bandwidth is DC-16 kHz. The designed magnetic shielding shell can effectively reduce 97.3% of the DC magnetic field interference and 95.7% of the power frequency magnetic field interference. The sensor can realize accurate measurement of weak current in power grid.

No abstract

In order to effectively detect the waveform characteristics of the closing inrush current of distribution transformers and distinguish between the magnetizing inrush current and the fault current when closing, this paper proposes a new method based on neighborhood preserving embedding (NPE) and principal component analysis (PCA) transformer closing inrush current detection method. This method can detect and process the global and local feature information of the data. First, the NPE-PCA algorithm is used to reduce the current data to two-dimensional space, and then the fitting error σ is obtained by fitting the two-dimensional space data. The relationship between σ and a given threshold is used to identify the magnetizing inrush current when closing. Finally, a model is built on the ATP/EMTP platform to test the proposed method for detecting the waveform characteristics of the closing inrush current. The simulation results show that the NPE-PCA inrush current waveform detection algorithm proposed in this paper can effectively identify the waveform characteristics of the transformer closing inrush current, which is consistent with the second harmonic wave. Algorithm comparison analysis shows that the performance of this algorithm is better.

In today's rapid economic development, industrial and civil electricity consumption is growing year by year, and how to guarantee stability of power system operation has become the focus of attention of the power sector in each country. Power load forecasting has been closely associated with the modernization of power system management and is a vital guarantee for the safe and stable operation and economic efficiency of the power system. In this article, the authors propose a recurrent neural network (RNN) decision fusion forecasting framework based on the wavelet transform to address the power load forecasting problem. The framework firstly performs the wavelet transform on the power load data and uses Daubechies wavelets to extract the high-frequency and low-frequency parts of the data; then the data with different frequencies are combined with the original data and fed into the RNN model separately, and the decision fusion is performed in the output layer; finally, the prediction results are obtained by superposition of two RNN networks. The results showed that the error of the predicted data in the last nine years decreased by 50%, compared with the traditional method of feeding the data into the RNN model for training, which provides a new idea for future power load forecasting.