Search

The short-term prediction of the electricity consumption of the distribution network during the peak summer period is of great significance to the safe and stable operation of the distribution network and to reduce the burden on the grid staff. This paper obtains and organizes the operating data of distribution transformers in a city in southern China, takes the active load rate as the electricity consumption index, and uses ARIMA and PCR models to make short-term electricity consumption prediction. The results show that ARIMA prediction results are significantly better than PCR. The research results of this paper have certain guiding significance for the actual work of distribution network.

This paper presents a coordination scheme of superconducting magnetic energy storage (SMES) and distributed generation (DG) system for regulating the DC bus voltage in the low voltage direct current (LVDC) distribution systems under symmetrical alternative current (AC) grid fault. The scheme can regulate DC bus voltage in milliseconds by taking advantage of SMES device including high power injection and fast response. On the other hand, the active control of DG system is introduced to increase the fully compensation time of the SMES under serious faults. By evaluating the reason of DC bus voltage sag, the control of SMES is proposed in cooperative with the DG system. A LVDC distribution system is modelled in PSCAD/EMTDC. Simulation results have demonstrated the correctness and feasibility of the proposed scheme.

With the increase of power load and the access of distributed generation, AC/DC hybrid distribution network will become the development trend of urban distribution network in the future. The core of AC/DC distribution and distributed generation are voltage source converters based on full control switches. This paper simply analyzes the operation principle and control mode of voltage source converters, a power flow calculation method of AC/DC hybrid distribution system is proposed, and the optimization model of AC/DC hybrid distribution network with minimum network loss and minimum voltage deviation of DC network is established. The particle swarm swarm (PSO) algorithm is used to solve the calculation. Considering the load disturbance, the droop control is set for each VSC, at the same time this paper compares the advantages and disadvantages of the whole time optimal power calculation and droop control combined with the optimal power calculation method. Then the control mode of the distribution network is proposed. Finally, an example of PG&E 69-node distribution network in the United States is given to verify the correctness of the proposed model and algorithm.

The premise of voltage sag mitigation for sensitive industrial processes is to reasonably assess the economic losses caused by voltage sag. And the process immunity time is the key information in the process of economic loss assessment. Existing research usually obtains process immunity time through monitoring systems or overall testing. However, in the actual situation, due to various constraints, the traditional method of obtaining the process immunity time is inconvenient. To get PIT more conveniently, a sensitive process immunity evaluation method based on response information flow is proposed in this paper. This method regards sensitive industrial processes as a composite system. According to the function of each part of the system and the response characteristics of voltage sag, it can be divided into three modular combinations, which are delayed structure, competitive structure, and conductive structure. According to the delay time and module combination characteristics of module response to the voltage sag, the propagation path of response information flow caused by sag can be determined, and the process immunity time can be evaluated.Finally, a case study of the painting process of a car factory in southern China shows that the proposed method can be applied to the sensitive process immune time assessment.

When transformers are energized under no-load or lightly loaded condition, the inrush current may occur on the primary circuit of the transformers. The magnetizing inrush current has an adverse effect on both the transformer itself (especially on the high-voltage built-in high-impedance transformer) and the power quality of the power system. The deprivation of effective magnetizing inrush suppression technology could lead to transformer trip and equipment damage. Since the inrush current of high-voltage built-in high-impedance transformer is large and decays slowly, the study proposes a technique combing Convolutional Neural Network (CNN) and inrush suppression technology. This new technique can adaptively correct the proper transformer switching angle by taking into account various facts, like temperature, operating voltage, oil pressure and times of control action. Based on PSCAD software, it simulates the theoretical effects of magnetizing inrush suppression technology, and verifies its practical benefit through building the actual system.