Narrow Results

Renewable energy sources connected in distribution systems utilizing power electronics devices to interface lead to various power quality problems. This chapter presents a review on power quality issues associated with the grid-connected renewable energy systems and mitigation techniques. To mitigate the power quality issues, an effective role is played by power electronic devices and custom power devices such as active power filters (APFs) and flexible AC transmission systems (FACTs). This chapter also discusses IEC and IEEE standards for grid-connected renewable energy system.

Power quality disturbance identification and classification are important for power quality management, for which a new method is proposed to deal with power quality disturbance identification and classification based on GA-ELM (Genetic Algorithm-Extreme Learning Machine). Decompose the disturbance signals with wavelet for ten layers, it is proposed that using standard deviation and mean of the energy difference, and the ratio of them as the feature vectors. In addition, the root mean of disturbance signals and normal signals is calculated as a supplement in order to reduce the dimension of the importing vectors. The Input Layer and the Hidden Layer Connection Weights and Hidden Layer Neuron Threshold which are set randomly in ELM can influence classification accuracy. So it is proposed that ELM training error is used as the fitness function of GA to optimize Input Layer and the Hidden Layer Connection Weights and Hidden Layer Neuron Threshold to enhance the accuracy of classification and also maintain high classification speed. Simulation results demonstrate that this method can accurately and effectively identify seven common disturbance types.

There has been a significant change to the domestic loads in recent years. With power electronic technique being more and more advanced and popular, an increasing number of domestic electrical appliances adopting power electronic devices are used and connected to low voltage distribution networks, which raises the harmonic level, resulting to a negative impact on the power quality. In this paper, domestic loads are classified into different categories according to simplified circuit topology. Furthermore, suggestions on how to improve power quality in networks are put forward.

To overcome the deviation caused by the effect of subjective factors in power quality evaluation, and make the evaluation more accurate and objective, on the basis of power quality comprehensive evaluation criterion, the projection data are built with projection pursuit method. Two hundred power quality comprehensive evaluation samples are randomly generated within the range of each class. The projection pursuit model is built by unitary processing according to these samples, which converts the evaluation into non-linear multi-constrained optimization problems. The optimal projection direction vector and the weight coefficient of each evaluation index are obtain with composite simplex method. Meanwhile the corresponding relationship between region of variation of projection eigenvalue and classification grade is established. The precision of sample evaluation reached 100% and the evaluation of the measured data achieved good effect. Thus, the scientific evaluation of power quality is realized by the projection pursuit model.

To improve accuracy and speed of a power quality monitoring device, the monitoring device with field-programmable gate array (FPGA) as the core of processing and controlling is designed. The hardware of the device includes signal acquisition unit, communication unit, power supply unit, etc; the software includes an NIOS II embedded core, optimizing the flow of data processed parallel, and Phase-locked frequency multiplication module and FFT harmonic analysis module implementing the synchronous sampling and analysis of power quality data are built. Test results show that the device has advantages of fast response, high accuracy and satisfactory real-time performance, and conforms to the national standards of power quality monitoring.

In this paper, a set of visual evaluation method based on the visualization technology is presented to evaluate power quality of power network in coal mine. Firstly, according to the characteristics of 10kV network, 14 indexes of voltage quality, current quality and power factor are selected to establish the power quality evaluation system of power network in coal mine. Then, parallel coordinates is used to delete two indexes of the evaluation results in order to reduce the follow-up calculation. Finally, using radar plot to evaluate power quality of power grid in coal mine, a combination weighting method based on sensitivity analysis and IOWA operator is put forward to mix many kinds of subjective and objective weight together in order to improve the accuracy of evaluation results. Through the actual evaluation for power quality of power grid in four coal mines, the effectiveness of the method is verified by the evaluation index system and the evaluation method.