Reminder
Click the blue font above to pay attention to China power micro signal
At present, the basic network of electric power and the information communication network are increasingly closely connected. A large number of data acquisition equipment, computing equipment and electrical equipment are interconnected through the two entity networks of power grid and communication network. The power system has the basic characteristics of the information physical system (CPS), which has become the information physical power system (CPPS). In CPPs environment, the composition and interaction form of source network load in the power physical layer tend to be more complex than before, and the disturbance and fault in the power communication network layer will also affect the power physical layer to varying degrees. Therefore, while using the information layer of CPPs to provide support for more accurate and effective optimization control of the whole network, it is necessary to consider reducing or avoiding the security risks introduced when transmitting information in the communication network layer. In order to show the research results in the field of information physical power system, China Power invited Professor Tang Yi of Southeast University and Dr. Ni Ming of NARI Group as its special editor, organized the column of "information physical power system (CPPS)", selected 10 representative papers, and published them. The main research directions include the analysis and evaluation of CPPs, disaster warning and prevention, information security, etc 。 Now we will introduce the abstract of the article for the readers.
1. Overview of research on physical system analysis and control of distribution information
Li Xiao, Li Manli, Ni Ming
Abstract: with the wide application of advanced information and communication technology in the distribution network, the integration degree of information side and physical side of the distribution network is increasing, which has the typical characteristics of information physical system. From the point of view of analysis and self-healing control, this paper analyzes and looks forward to the current research situation of CPDS. In the aspect of analysis, it summarizes and analyzes the reliability assessment and risk assessment of CPDS. In the aspect of self-healing control, the traditional physical side self-healing control and CPDs self-healing control coordinated by information physics are reviewed, pointing out the progress and shortage of research. Finally, the future research direction of CPDS is prospected from four aspects: CPDs architecture under the background of two networks integration, CPDs security analysis considering distributed power supply, CPDs hybrid communication network self-healing mechanism and CPDs self-healing control system.
2. Power system transient stability prediction based on data enhancement and deep residual network
Zhou Yanzhen, Zha Xianyu, LAN Jian, Guo Qinglai, sun Hongbin, Xue Feng, Wang Shengming
Abstract: in the traditional data-driven power system transient stability analysis method, the influence of input data noise and lack of information on the performance of the prediction model is less considered. A transient stability prediction method based on data enhancement and deep residual network is proposed. First of all, the original training data is expanded considering the situation of noise and information loss; then, the dynamic data of generator disturbed is used as the input feature; considering that the input high-dimensional time series data has the characteristics of image, a special convolution neural network - depth residual network in image processing is used to build a deep model for transient stability evaluation. The example analysis shows that the proposed method can improve the generalization ability of the model and has better robustness in the case of noise and lack of some generator information.
3. Risk assessment and prevention of extreme events in power system
Ding Maosheng, sun Weijia, Cai Xingpu, Wang Qi
Abstract: traditional power system risk assessment methods take common risk as the research object, which can not reflect the destructive impact of extreme events on system operation. Based on the theory of risk value in financial field, an extreme risk assessment and quantification method of power system is proposed to supplement and improve the traditional risk assessment method. The extreme value risk assessment method is used to investigate the impact of extreme events on the stable operation of the system, which reflects the serious consequences of low probability extreme events on the stability of the system. Based on the power system network attack event and extreme disaster event, the extreme risk assessment is carried out, and the rationality of the proposed extreme risk assessment method is verified by comparing the results of conventional risk assessment and extreme risk assessment. At the same time, based on the characteristics of power system vulnerability and anti vulnerability, the key difficulty of extreme risk prevention is how to coordinate ordinary risk and extreme risk.
4. CPS security evaluation of distribution network considering information physical combination predictive fault screening
Zhou Xia, Yang Zhou, Ni Ming, bu Chengjie, Lin Husheng, Li Manli
[Abstract] with the rapid development of ICT technology and automation control technology, the distribution network has gradually developed into a highly coupled distribution network information physical system (CPS) with information system and physical system. The information system supports the stable operation of the physical system, but also brings certain security risks to the CPS of the distribution network. Therefore, it is necessary to consider the information in the original single physical fault assessment Interest effect. This paper presents a method of CPS security analysis and evaluation for distribution network considering information physical combination predictive fault screening. Firstly, based on the structure of distribution network CPS, a CPS correlation matrix model which can reflect the coupling characteristics of power and information is established; secondly, the correlation between information fault and physical fault is analyzed, and the initial information physical combination predictive fault set is constructed according to the topological correlation and business correlation; secondly, the key information physical combination predictive fault is selected according to the order of fault recovery rate; Finally, aiming at the selected combined predictive fault, the CPS safety evaluation index of distribution network is used for safety evaluation, which is verified in the example. The result of the example shows that the proposed method is reasonable and effective.
5. Allocation strategy of power grid disaster prevention resources considering the sequence of disaster events
Zhao Yanjun, Liang kunjie, Long Fei, Tang Jingxing, Wang Hongtao
[Abstract] the allocation and construction of disaster prevention resources such as wind prevention and reinforcement of power grid in typhoon prone areas in southern China is one of the key issues in the security defense of large power grid. It is usually modeled as a dynamic game model of defender attacker defender (D-A-D) for solution, but the existing D-A-D model does not consider the dynamic process of attacker attack sequence and operation mode adjustment. This paper proposes a multi-stage dynamic game model considering the attack sequence of attackers and the dynamic process of power flow adjustment, and transforms it into an improved D-A-D model solved by the available column and constraint generation algorithm, and obtains the optimal defense resource allocation result. The simulation results show that the proposed model is more in line with the actual situation of the power grid and the algorithm is more efficient.
6. On-line warning of time and space intensity of transmission tower failure caused by rainfall and landslide
Tang Yi, Xu Xiangxiang, Chen Bin, Yi Chen
[Abstract] in view of the complex environment of transmission lines in mountainous and hilly areas, which may lead to transmission tower failure and affect the safe and stable operation of the power system, a method of transmission tower failure early warning under the rainfall landslide disaster is proposed, which can realize the whole process of the disaster from three aspects of landslide risk area, landslide activity time and tower damage Quantify online early warning. Firstly, the rainfall forecast area and the power grid area are overlapped to obtain the target risk area by using GIS technology, and the grid based landslide disaster index is proposed to determine the prone landslide area; secondly, the daily change of rainfall infiltrating into the soil is analyzed by combining with the meteorological data, and the dynamic updated cumulative effective rainfall model is constructed, which can be used to predict the time of landslide activity Finally, taking the impact energy of slope as the disaster intensity and the bending deformation energy corresponding to the deflection limit of transmission tower as the disaster resistance performance, the quantitative function relationship reflecting the damage degree of transmission tower in the landslide prone area is obtained from the perspective of work energy. The results show that the proposed method is feasible and reasonable.
7. Cable channel emergency monitoring system based on mobile wireless sensor network
Xu Xinqian, Xu Rhei, Chen Jie, Liu Yang, Zhang chaopeng, he Jiahong, Gao BINGTUAN
[Abstract] in order to meet the needs of environmental information monitoring in case of loss, failure or attack of cable channel communication network, a low-cost mobile wireless sensor network monitoring system based on WiFi and ultra wide band (UWB) communication is developed. The system includes a monitoring host computer and a wireless monitoring network composed of spherical robot mobile nodes. The mobile node is mainly controlled by STM32 series single-chip microcomputer, integrated with camera module and temperature and humidity module to realize on-site information collection. The integrated WiFi communication module adopts ad-hoc on-demand distance vector routing, AODV) constructs wireless sensor network to realize information transmission, integrates three UWB communication modules to realize the ranging and queue following of mobile nodes. The experimental results show that the system can realize the stable real-time monitoring of the cable channel environment. The following error of the mobile sensor queue is less than 20 cm, and the average transmission delay of the monitoring video is less than 150 ms.
8. Safety prediction of power transformer based on multi prediction model fusion
Li dianyang, Zhang Yujie, Wang Shanyuan, Feng Jian, Wang Hongzhe, Qin Ling
[Abstract] in order to solve the problem that traditional analysis measures fail to include multi-source and heterogeneous data information into the fault symptom analysis of power grid equipment and it is difficult to better deal with the diagnosis of multiple fault types of equipment based on a single algorithm with small samples, a unified and standardized method of data related to different types of power grid equipment is constructed. Chi square distribution algorithm is used to select the symptom set of specific fault type through data correlation mining, so as to avoid the interference of artificial experience in the analysis methods of fault symptom set such as AHP and Greenland verification. A multi algorithm fusion decision method is constructed to avoid the disadvantages of single algorithm decision. The results show that it is better to find fault symptom set for single fault type of power equipment than to select fault symptom set for equipment. The example also shows that the proposed fusion algorithm is better than the single algorithm.
9. Research and verification of network attack type of embedded terminal in Intelligent Substation
He Jindong, Wang Yu, Zhao Zhichao, Li June, Xie Xinzhi, Zhang Ruiwen, Liu Yuqing, Qiu Dexi
[Abstract] in order to meet the needs of environmental information monitoring in case of loss, failure or attack of cable channel communication network, a low-cost mobile wireless sensor network monitoring system based on WiFi and ultra wide band (UWB) communication is developed. The system includes a monitoring host computer and a wireless monitoring network composed of spherical robot mobile nodes. The mobile node is mainly controlled by STM32 series single-chip microcomputer, integrated with camera module and temperature and humidity module to realize on-site information collection. The integrated WiFi communication module adopts ad-hoc on-demand distance vector routing, AODV) constructs wireless sensor network to realize information transmission, integrates three UWB communication modules to realize the ranging and queue following of mobile nodes. The experimental results show that the system can realize the stable real-time monitoring of the cable channel environment. The following error of the mobile sensor queue is less than 20 cm, and the average transmission delay of the monitoring video is less than 150 ms.
10. Early warning method of cross space cascading fault of power grid CPS station level considering attack profit and loss
Wang Yufei, Qiu Jian, Li June
[Abstract] in order to accurately detect all kinds of cross space interlocking faults caused by network attacks, an early warning method of cross space interlocking faults at station level is proposed, which considers the attack profit and loss. First of all, according to the evolution mechanism of cross space interlocking fault