Research on industrial Internet of Things and power grid technology application based on knowledge graph and data asset relationship model

Xue Wang; Le Ma; Zhenwei Yang

Dettagli rivista e numero

Applied Mathematics and Nonlinear Sciences

Dettagli della rivista

Formato: Rivista
eISSN: 2444-8656
Prima pubblicazione: 01 Jan 2016
Frequenza di pubblicazione: 2 volte all'anno
Lingue: Inglese

With the development of the digital economy and digital technology, more and more scholars pay attention to data asset research [1–4]. The discussion on grid data capitalisation still focuses on the theoretical level, including the discussion and definition of related concepts, the value expression method of data assets, the conceptual framework, mechanism and challenges of data capitalisation [5, 6]. In the related research on the value expression of data assets, the possible impact of a mining data value on enterprises is mainly discussed [7, 8].

A knowledge graph is a graph knowledge base that can organise scattered knowledge very well [9, 10]. The nodes of the knowledge graph in a certain domain represent the entities in the domain, and the lines represent the connections between entities in the domain, organising scattered domain knowledge [11]. Since knowledge graphs support semantic search, they are mainly divided into open domain and vertical domain knowledge graphs [12]. With the in-depth application of knowledge graphs in specific fields, many knowledge bases or knowledge graph platforms have emerged in the field of power grids.

In recent years, the Internet of Things has been widely used in industrial production optimisation, management improvement, service improvement, energy conservation and emission reduction [13, 14]. The application prospect of the Industrial Internet of Things is very broad. The Internet of Things builds smart factories with adaptability and resource efficiency by improving the computerization, digitisation and intelligence of the manufacturing industry [15, 16]. The series of policy documents for the ‘Internet + Advanced Manufacturing’ to develop the Industrial Internet propose to vigorously develop intelligent manufacturing and accelerate the in-depth integration of information and manufacturing technologies [17, 18]. At present, China’s industrial data acquisition technology and application are still in their infancy, and there are serious problems such as insufficient sensor deployment, a limited amount of collected data, low accuracy, and low efficiency [19, 20]. In addition, practical problems such as the wide variety of industrial communication protocols, poor compatibility and interoperability are prominent. It cannot support real-time industrial data collection and real-time analysis, intelligent optimisation and scientific decision-making and other business requirements [21, 22]. The following are the problems that arise in the Industrial Internet of Things:

The networking rate of industrial equipment is low.

Industrial communication protocol standards are numerous and incompatible with each other.

Industrial data collection has potential data security risks. It involves a large amount of important industrial data and user privacy information. There are certain data security risks during transmission and storage, and there is also the risk of hackers stealing data and attacking enterprise production systems.

The power grid technical standard system planning is the smart grid standard planning, which plays a basic supporting role in the construction of smart grids [23]. With the changes in internal and external situations, the original standard system has been unable to meet the needs of smart grid standardisation in the new era, and the smart grid technical standard system has been revised [24].

The global energy structure and pattern have undergone profound changes. The large-scale development and wide application of clean energy have become the mainstream of world energy development.

Electricity marketisation has brought new changes. The electricity market reform characterised by the approval of transmission and distribution prices and the liberalisation of the electricity sales market is gradually deepening.

Common supporting technologies improve the level of power grid intelligence. The wide application of common technologies such as cloud-based technology, superconducting materials, and power meteorology enables smart grids to be used in large-scale power grid security early warning and active distribution network fault reconstruction.

There have been major adjustments to achieve some strategic goals in the implementation of smart power grids. The technical structure, business model and management method of the strong smart grid will also undergo great changes.

The reform of standardisation work is deepened. Comprehensive construction mode actively plays the supporting role of standardised services, and carries out the construction of national technology standard innovation base (smart grid).

In summary, this paper focuses on the research of the industrial Internet of Things and power grid technology based on the knowledge graph and the data asset relationship model. First, the existing literature in this paper is sorted out to define the concept and characteristics of data assets. Then, a training model based on the knowledge graph and data asset relationship model is proposed. At the same time, the key technologies of the Industrial Internet of Things and the theoretical model framework of power grid technology are proposed, including device access, protocol conversion, edge data processing, connectivity, the first entry of data and the rapid growth of data volume. Finally, experiments are carried out to verify the framework proposed in this paper. Experiments show that the research on industrial Internet of Things and power grid technology based on the knowledge graph and data asset relationship model is of great significance for enterprises to exert data value and promote the development of the digital economy.

Based on the knowledge graph and data asset relationship model

This paper first defines the concept and characteristics of data assets. Then, it introduces the way in which the value of data assets is reflected in the existing research and the related discussions on the path of data assetisation.

Definition of the concept and characteristics of data assets: Clarifying the concept of data assets is an important basis for enterprises to realise data assetisation, which is helpful for enterprises to sort out and take stock of data assets.

The way to reflect the value of data assets: The decision support system based on data analysis is the main way for data to exert its value. Reliable data evidence, rather than intuition and experience, is more accurate in forecasting, and a data-driven approach can significantly improve business performance and competitiveness.

Theoretical and practical exploration of data assetisation: It has become a general consensus that data has become a key asset of an enterprise. Lack of corresponding regulations and standards for data assetisation, such as data ownership definition, privacy and data security protection, and data asset value evaluation standards prevail. Part of it is difficult to effectively develop and utilise relevant data.

2.1

Data asset relationship model based on convolutional neural network (CNN) + bidirectional long short-term memory (BiLSTM) + CRF

In this paper, a knowledge graph model is constructed based on CNN and BiLSTM network. Based on BiLSTM and CNN, CRF is introduced, and the specific structure is shown in Figure 1.

Based on CNN + BiLSTM + CRF data asset relationship model. BiLSTM, bidirectional long short-term memory; CNN, convolutional neural network

First, the input needs to extract the data asset text input sequence. To calculate the convolution equation: 1 $c_{i} = R e L U (ω w_{i : i + h - 1}) + b$ where the text sequence length is n and the window length is n-h + 1. Calculated as follows: 2 $C = {C_{1}, \dots, C_{n - h + 1}}$

The contextual semantic content of the word window convolution by the CNN model can better represent the local features of the text sequence.

Then, long short-term memory (LSTM) is a special kind of recurrent neural network (RNN), which consists of one cell unit with three gates. The cell unit is the core computing power and records the current computing state. Forget gates, input gates, and output gates regulate the flow of information to and from memory cells. The forget gate clears the memory cells of useless information. The input gate selects the input information of the current memory cell. The output gate determines the final output of the information so that the storage unit can efficiently store the semantic information of a longer sequence. The calculation process is as follows: 3 $i_{i} = σ (W_{i} e (w_{i - 1}) + U_{i} h_{i - 1} + V_{i} c_{i - 1} + b_{i})$ 4 $f_{i} = σ (W_{f} e (w_{i - 1}) + U_{f} h_{i - 1} + V_{f} c_{i - 1} + b_{f})$ 5 $o_{i} = σ (W_{o} e (w_{i - 1}) + U_{o} h_{i - 1} + V_{o} c_{i - 1} + b_{o})$

Where x_t is input text vector. σ is the sigmoid function; tanh is the activation function; respectively i_t, o_t, f_t are the input gate, the output gate and the forgetting gate. The unit of storage information at time t is c_t. where the input gate and forget gate are not used to adjust the information unit.

6 $c_{t} = \tanh (W_{c} x_{t} + U_{c} h_{t - 1} + b_{c})$

At the same time, c_t stores the unit information of the forget gate and the input gate at the last moment.

7 $c_{t} = f_{i} ⊙ c_{t - 1} + i_{i} ⊙ \tilde{c_{i}}$ 8 $h_{i} = o_{i} ⊙ \tanh (c_{i})$

Among them, the hidden layer h_t is calculated by the output gate and the storage gate c_t, and the BiLSTM network layer is used and constructed to splice the forward and backward outputs of the LSTM unit at time t.

9 $h_{i} = [\vec{h_{t}}, \overset{\leftarrow}{h_{t}}] \in R^{n}$

Finally, CRF can obtain an optimal prediction sequence through the relationship between adjacent labels. It can make up for the shortcomings of BiLSTM. Calculated as follows: 10 $S (X, Y) = \sum_{i = 0}^{n} A_{y_{i}, y_{i + 1}} + \sum_{i = 1}^{n} P_{i, y_{i}}$ 11 $P (Y | X) = \frac{e^{s (X, Y)}}{\sum y \in y_{x}} s (X, Y)$ 12 $\ln (p (Y | X)) = s (X, Y) - \ln (\sum_{y \in y_{x}} s (X, Y))$ 13 $Y^{*} = \arg max s (X, Y)_{y \in y_{x}}$

Industrial Internet of Things and power grid technology

Industrial data collection is the use of ubiquitous sensing technology to efficiently collect and aggregate information on multi-source equipment, heterogeneous systems, operating environment, human and other elements in real time and in the cloud. The scope of industrial data collection in this paper mainly refers to the data collection of industrial field equipment and the data collection of intelligent products/equipment outside the factory, as shown in Figure 2.

Power grid industrial data acquisition technology architecture

3.1

Key technologies of the industrial Internet of Things

The Industrial Internet of Things belongs to the Internet of Things technology in the industrial field, and its core essence is to establish a comprehensive interconnection between people and various elements of the industrial system. Through data flow and analysis, an intelligent transformation of industrial systems is formed. It can integrate various types of acquisition and sensing devices, control and execution devices with sensing and control capabilities, as well as wireless transmission networks, big data analysis platforms and other technologies into all aspects of the industrial production process to improve production efficiency, reduce production costs, and finally realise the transformation and upgrading of traditional industries to intelligent production processes. Therefore, industrial Internet of Things technology research is an interdisciplinary project involving sensing technology, data interaction technology and information processing technology.

The Industrial Internet of Things needs to solve the technical problem of ubiquitous perception. Industrial systems involve a huge number of devices and the generalisation of information, which requires a high number of sensors and easy maintenance. Sensors with low prices, small sizes, free long-term maintenance, and high intelligence will become the cornerstone of industrial IoT applications. Sensing elements should be miniaturised to reduce resource consumption. The sensing device should be intelligent, with self-calibration, self-diagnosis, self-learning, self-adaptive and self-decision-making capabilities. At the same time, it has the energy capture ability to meet long-term power supply and maintenance requirements.

The Industrial Internet of Things needs to solve the technical problem of data interaction. The data generated by on-site generalised perception needs to be transmitted to regional edge computing nodes and fusion collection stations or ground data centrer for processing to generate corresponding value. Perception data must have an identification mark, which can be identified. A standard interaction protocol is established so that the generalised perception data can be effectively received. For traditional sensing devices in use, corresponding protocol gateways need to be developed to facilitate compatibility.

The Industrial Internet of Things needs to solve the technical problem of data transmission. The network is the main component of the Industrial Internet of Things, through which data can be transferred between different levels of the system. At present, the overall technology of wired broadband networks is relatively mature, and it is mainly used in backbone transmission and data centres. However, for industrial production sites, due to a large number of sensors, to reduce wiring costs and facilitate maintenance, it is necessary to build a wireless sensor network suitable for industrial sites for achieving latency and low power consumption. At the same time, to meet the dynamic expansion of sensors, industrial wireless sensor networks are required to have ad-hoc network capabilities.

The Industrial Internet of Things needs to solve the technical problems of effective processing and sharing of data. The amount of data in industrial systems is relatively large and is called giant systems. It is imperative to develop the knowledge to effectively analyze, process and store relevant data, and extract instructive results for industrial production through fusion and mining analysis. At the same time, the application of industrial data is an ecology, and it is necessary to ensure that the collected data can be applied by all links and objects of use, to realise the organic sharing of data.

3.2

Grid technology

Power grid technology is the technical foundation supporting the overall technology of each main business. The power grid technology direction is mainly divided into eight technical fields: power chips, new electrical materials, superconducting power, artificial intelligence, power meteorology, power storage, power quality, and power electronic devices.

Power chips can be embedded in all kinds of primary and secondary power equipment and devices, and are widely used in all aspects of smart grids. They are smart grid systems and equipment.

With the rapid development of new electrical material technology, it is necessary to develop electrical insulating materials, electrical magnetic materials, electrical conductor materials, environmental protection materials, new energy storage materials and intelligent sensing materials.

Superconducting power standards mainly focus on the preparation, construction, testing, online monitoring and system protection of superconducting power core components including superconducting cables, superconducting current limiters, and superconducting transformers.

Artificial intelligence has broad application prospects in the fields of smart grid construction, operation control, maintenance and equipment management. Standardisation work needs to be carried out in power data labelling, power intelligent perception, power intelligent cognition, power intelligent behaviour, artificial intelligence platform and artificial intelligence application maturity evaluation.

Global climate change leads to changes in the spatial and temporal distribution characteristics of wind and solar resources and a significant increase in extreme weather events. Standards for basic observation data of power meteorology, numerical weather forecast and analysis of power meteorology, forecast and early warning of power meteorological disasters, and production and application of power meteorology are formulated.

As the proportion of clean energy increases, local consumption, plug-and-play, and flexible mobility of electric energy storage will become a development trend. Standardisation work on electrochemical energy storage, physical energy storage, hydrogen energy storage planning and design, equipment manufacturing, grid connection and testing are carried out.

In the fields of DC power transmission, new power grid connection, railway power supply, industrial smelting and other fields, the large-scale use of power electronic devices leads to serious power quality problems such as voltage flicker, harmonic distortion and reactive power shock in the power grid. There is a need to formulate and improve the basic standards for power quality, monitor and evaluate, and employ technical standards for power quality control.

The core of new energy grid connection lies in flexible AC and DC transmission, and microgrid operation control. To guide and regulate the development of power electronic devices, it is necessary to carry out standardisation work on silicon power devices and new power devices.

3.3

Industrial Internet of Things and power grid technology framework based on the knowledge graph and data asset relationship model

Industrial Internet of Things and power grid technology data collection access different devices, systems and products through various communication means, collect large-scale, deep-level industrial data, protocol conversion and edge processing of heterogeneous data, and build the data foundation of the industrial Internet platform. The technical architecture of industrial data acquisition includes three layers of equipment access, protocol conversion, and edge data processing, connecting equipment or smart products downwards, and connecting with the industrial Internet platform/industrial application system upwards.

Equipment access: Through various wired and wireless communication technologies such as industrial Ethernet, industrial optical fibre network, industrial bus, 3G/4G, NB-IoT, etc., access to various industrial field equipment, such as smart products/equipment, through the collection of power grids technical data.

Protocol conversion: Protocol analysis and conversion, middleware and other technologies are compatible with various industrial communication protocols such as ModBus, CAN, Profinet, etc., to realise data format conversion and unification. HTTP, MQTT and other methods are used to transmit the collected data to the cloud data application analysis system or data aggregation platform.

Edge data processing: Based on technical support such as high-performance computing, real-time operating systems, and edge analysis algorithms, data preprocessing is performed on the network edge side close to the device or data source. Storage and intelligent analysis applications improve operational response sensitivity, eliminate network congestion, and form a synergy with cloud data analysis.

Connectivity: Connectivity is the foundation of industrial data collection with a variety of connected physical objects and a variety of application scenarios. Industrial data acquisition has rich connection functions, such as various network interfaces, network protocols, network topology, network deployment and configuration, network management and maintenance.

The first entrance of data: As a bridge from the physical world to the digital world, industrial data acquisition is the first entrance of data, with a large amount of real-time and complete data. As the first entrance to data, industrial data collection also faces challenges such as real-time data, certainty and diversity.

Rapid growth of data volume: As industrial systems extend from physical space to information space and from the visible world to the invisible world, the scope of industrial data collection continues to expand. The amount of data generated, collected and processed by industrial equipment, including equipment status parameters, operating conditions, loads, and operating environment, is growing explosively, far greater than the data generated by computers and humans in enterprises with the development of smart manufacturing and IoT technology.

Experimental results and analysis

4.1

Experimental data

After the text data is preprocessed, it is vectorised and then input into the industrial Internet of Things and power grid technology framework model based on the knowledge graph and data asset relationship model for the next step. In this paper, the cleaned power grid corpus data is used for word vector training. The corpus data size is about 5.3 GB, and the Word2Vec tool is used for large-scale text training to convert the text into a low-density vector space. This article selects the Word2Vec tool Skip-gram model to train word vectors. The Skip-gram model takes all the words in the corpus as the central word and predicts the lexical information of its context through the conditional probability distribution of the correspondence between the central word and the context.

4.2

Experimental evaluation criteria

In this paper, the cross entropy is used as the loss function, and the operation effect of the model is measured by the accuracy rate, recall rate and F value. Taking Chinese word segmentation as an example, the calculation equation of the indicator is as follows: 14 $R = \frac{T P}{T P + F N} \times 100 %$ 15 $P = \frac{T P}{T P + F P} \times 100 %$ 16 $F = \frac{2 P R}{P + R} \times 100 %$

4.3

Analysis of results

In this paper, CNN, CNN-CRF, LSTM, LSTM-CRF, BiLSTM, BiLSTM + CRF and CNN + BiLSTM + CRF are used for performance analysis. The experimental results are shown in Table 1 and Figure 3. Comparing the experimental results of the CNN model and the LSTM model, it can be seen that LSTM is better than CNN in extracting entities from this dataset. Comparing the experimental results of the CNN and LSTM models and the CNN-CRF and LSTM-CRF models, it can be seen that the F value is improved after the CRF module is added. This is because CRF can make full use of the association of adjacent entity labels. Comparing the experimental results of the LSTM-CRF model and the BiLSTM-CRF model, it can be seen that BiLSTM performs better than LSTM because LSTM can only use the above information and cannot use the following information. From the experimental results of the CNN-BiLSTM-CRF model, it can be seen that the F value has been significantly improved because CNN can deeply extract the semantic information of entity texts and fully characterise polysemy.

Table 1

Model effect comparison (/%)

Model	P	R	F
CNN	80.83	79.52	79.67
LSTM	81.59	82.15	81.37
CNN-CRF	82.14	81.87	81.15
LSTM-CRF	82.68	82.62	82.34
BiLSTM	83.48	83.42	83.02
BiLSTM + CRF	85.87	85.45	85.09
CNN + BiLSTM + CRF	93.63	91.46	92.41

BiLSTM, bidirectional long short-term memory; CNN, convolutional neural network; LSTM, long short-term memory

Conclusion

This paper focuses on the research of industrial Internet of Things and power grid technology based on the knowledge graph and data asset relationship model. The key technology of the Industrial Internet of Things and the theoretical model framework of power grid technology proposed in this paper can be widely used in the remote monitoring and intelligent maintenance application scenarios of various industrial equipment or intelligent products of process type or discrete type. Through the collection and analysis of the operation data, such as the location, status, sensor, alarm and performance of the equipment/smart products, the functions of remote monitoring of equipment, fault alarm, preventive maintenance and performance optimisation analysis are realised, and the service efficiency and effectiveness are improved. However, the scheme proposed in this paper still has defects. If user information is not considered, the accuracy of entity extraction is not high. In future work, we will experiment with different models for in-depth research.

Figure e tabelle

Model effect comparison (/%)

Model	P	R	F
CNN	80.83	79.52	79.67
LSTM	81.59	82.15	81.37
CNN-CRF	82.14	81.87	81.15
LSTM-CRF	82.68	82.62	82.34
BiLSTM	83.48	83.42	83.02
BiLSTM + CRF	85.87	85.45	85.09
CNN + BiLSTM + CRF	93.63	91.46	92.41

Riferimenti

[1] Wu, Yao, Xie, et al. Value co-creation between firms and customers: The role of big data-based cooperative assets[J]. Information & Management, 2016. Wu, Yao, Xie, Value co-creation between firms and customers: The role of big data-based cooperative assets[J]. Information & Management, 2016. Search in Google Scholar

[2] Zhang W, Wang W. Cost modelling in maintenance strategy optimisation for infrastructure assets with limited data[J]. Reliability Engineering & System Safety, 2014, 130(oct.):33-41. Zhang W, Wang W. Cost modelling in maintenance strategy optimisation for infrastructure assets with limited data[J]. Reliability Engineering & System Safety, 2014, 130(oct.):33-41.10.1016/j.ress.2014.04.025 Search in Google Scholar

[3] Vasilevskaya M, Nadjm-Tehrani S. Quantifying Risks to Data Assets Using Formal Metrics in Embedded System Design[C]//34th International Conference on Computer Safety, Reliability, and Security (SAFECOMP). Springer, Cham, 2014. Vasilevskaya M, Nadjm-Tehrani S. Quantifying Risks to Data Assets Using Formal Metrics in Embedded System Design[C]//34th International Conference on Computer Safety, Reliability, and Security (SAFECOMP). Springer, Cham, 2014.10.1007/978-3-319-24255-2_25 Search in Google Scholar

[4] Bao L S, Zha Y Y, Jin-Ling H E, et al. Exploration and Practice of Data Assets Management and Control Mode of Provincial Grid Company[J]. Electric Power Information & Communication Technology, 2018. Bao L S, Zha Y Y, Jin-Ling H E, Exploration and Practice of Data Assets Management and Control Mode of Provincial Grid Company[J]. Electric Power Information & Communication Technology, 2018. Search in Google Scholar

[5] Zhang Y, Wang H, Hong Z, et al. The design and application research of visualization system on data assets in power grid system based on 3D GIS platform[J]. Electrical Measurement & Instrumentation, 2018. Zhang Y, Wang H, Hong Z, The design and application research of visualization system on data assets in power grid system based on 3D GIS platform[J]. Electrical Measurement & Instrumentation, 2018. Search in Google Scholar

[6] Bo, Chai, Siyan, et al. Fault Range Analysis of Information and Communications Assets in Power Grid: A Graph Data Perspective[C]//2017 International Conference on Applied Mathematics, Modelling and Statistics Application (AMMSA 2017). 0. Bo, Chai, Siyan, Fault Range Analysis of Information and Communications Assets in Power Grid: A Graph Data Perspective[C]//2017 International Conference on Applied Mathematics, Modelling and Statistics Application (AMMSA 2017). 0.10.2991/ammsa-17.2017.34 Search in Google Scholar

[7] Coppolino L, D’Antonio S, Elia I A, et al. Security Analysis of Smart Grid Data Collection Technologies[C]//International Conference on Computer Safety, Reliability, and Security. Springer, Berlin, Heidelberg, 2011. Coppolino L, D’Antonio S, Elia I A, Security Analysis of Smart Grid Data Collection Technologies[C]//International Conference on Computer Safety, Reliability, and Security. Springer, Berlin, Heidelberg, 2011.10.1007/978-3-642-24270-0_11 Search in Google Scholar

[8] Zhang S, Zheng G Q, Liu L, et al. A Machine Learning based Knowledge Graph Framework for Heterogeneous Power Grid Systems[C]//2021 IEEE International Conference on Consumer Electronics and Computer Engineering (ICCECE). IEEE, 2021. Zhang S, Zheng G Q, Liu L, A Machine Learning based Knowledge Graph Framework for Heterogeneous Power Grid Systems[C]//2021 IEEE International Conference on Consumer Electronics and Computer Engineering (ICCECE). IEEE, 2021.10.1109/ICCECE51280.2021.9342047 Search in Google Scholar

[9] Hu J, Zhao S, Nie Q. Research on Modeling of Power Grid Information System Based on Knowledge Graph[C]//2021 IEEE International Conference on Power Electronics, Computer Applications (ICPECA). IEEE, 2021. Hu J, Zhao S, Nie Q. Research on Modeling of Power Grid Information System Based on Knowledge Graph[C]//2021 IEEE International Conference on Power Electronics, Computer Applications (ICPECA). IEEE, 2021.10.1109/ICPECA51329.2021.9362503 Search in Google Scholar

[10] Hu J, Zhao S, Nie Q. Research on Modeling of Power Grid Information System Based on Knowledge Graph[C]//2021 IEEE International Conference on Power Electronics, Computer Applications (ICPECA). IEEE, 2021. Hu J, Zhao S, Nie Q. Research on Modeling of Power Grid Information System Based on Knowledge Graph[C]//2021 IEEE International Conference on Power Electronics, Computer Applications (ICPECA). IEEE, 2021.10.1109/ICPECA51329.2021.9362503 Search in Google Scholar

[11] Huang H, Hong Z, Zhou H, et al. Knowledge Graph Construction and Application of Power Grid Equipment[J]. Mathematical Problems in Engineering, 2020, 2020(6):1-10. Huang H, Hong Z, Zhou H, Knowledge Graph Construction and Application of Power Grid Equipment[J]. Mathematical Problems in Engineering, 2020, 2020(6):1-10.10.1155/2020/8269082 Search in Google Scholar

[12] Dan L, Xin C, Huang C, et al. Intelligent Agriculture Greenhouse Environment Monitoring System Based on IOT Technology[C]//2015 International Conference on Intelligent Transportation, Big Data & Smart City (ICITBS). IEEE, 2016. Dan L, Xin C, Huang C, Intelligent Agriculture Greenhouse Environment Monitoring System Based on IOT Technology[C]//2015 International Conference on Intelligent Transportation, Big Data & Smart City (ICITBS). IEEE, 2016.10.1109/ICITBS.2015.126 Search in Google Scholar

[13] Zhang Y, Qian T, Tang W. Buildings-to-distribution-network integration considering power transformer loading capability and distribution network reconfiguration[J]. Energy, 2022, 244. Zhang Y, Qian T, Tang W. Buildings-to-distribution-network integration considering power transformer loading capability and distribution network reconfiguration[J]. Energy, 2022, 244.10.1016/j.energy.2022.123104 Search in Google Scholar

[14] T. Qian, Xingyu Chen, Yanli Xin, W. H. Tang*, Lixiao Wang. Resilient Decentralized Optimization of Chance Constrained Electricity-gas Systems over Lossy Communication Networks [J]. Energy, 2022, 239, 122158. Qian T., Chen Xingyu, Xin Yanli, Tang* W. H., Wang Lixiao. Resilient Decentralized Optimization of Chance Constrained Electricity-gas Systems over Lossy Communication Networks [J]. Energy, 2022, 239, 122158.10.1016/j.energy.2021.122158 Search in Google Scholar

[15] T. Qian, Y. Liu, W. H Zhang, W. H. Tang*, M. Shahidehpour. Event-Triggered Updating Method in Centralized and Distributed Secondary Controls for Islanded Microgrid Restoration[J]. IEEE Transactions on Smart Gird, 2020, 11(2): 1387-1395. Qian T., Liu Y., Zhang W. H, Tang* W. H., Shahidehpour M.. Event-Triggered Updating Method in Centralized and Distributed Secondary Controls for Islanded Microgrid Restoration[J]. IEEE Transactions on Smart Gird, 2020, 11(2): 1387-1395.10.1109/TSG.2019.2937366 Search in Google Scholar

[16] Zhang B, Lin H, Zuo S, et al. Research on Intelligent Robot Engine of Electric Power Online Customer Services Based on Knowledge Graph[C]//ICIAI 2020: 2020 the 4th International Conference on Innovation in Artificial Intelligence. 2020. Zhang B, Lin H, Zuo S, Research on Intelligent Robot Engine of Electric Power Online Customer Services Based on Knowledge Graph[C]//ICIAI 2020: 2020 the 4th International Conference on Innovation in Artificial Intelligence. 2020.10.1145/3390557.3394312 Search in Google Scholar

[17] Wei X G, Yue-Fa O U, Liu K M, et al. Research and practice of “Internet + advanced manufacturing” training base construction[J]. Heilongjiang Science, 2019. Wei X G, Yue-Fa O U, Liu K M, Research and practice of “Internet + advanced manufacturing” training base construction[J]. Heilongjiang Science, 2019. Search in Google Scholar

[18] Luo M, Yang X. Research on “Industrial Internet” to promote the development of Advanced Manufacturing Industry in Guangdong, China[J]. Journal of Physics Conference Series, 2020, 1601:032052. Luo M, Yang X. Research on “Industrial Internet” to promote the development of Advanced Manufacturing Industry in Guangdong, China[J]. Journal of Physics Conference Series, 2020, 1601:032052.10.1088/1742-6596/1601/3/032052 Search in Google Scholar

[19] G Quirós, Cao D, Caedo A. Dispersed automation for industrial Internet of Things – An enabler for advanced manufacturing[C]//2017 13th IEEE Conference on Automation Science and Engineering (CASE). IEEE, 2018. Quirós G, Cao D, Caedo A. Dispersed automation for industrial Internet of Things – An enabler for advanced manufacturing[C]//2017 13th IEEE Conference on Automation Science and Engineering (CASE). IEEE, 2018.10.1109/COASE.2017.8256113 Search in Google Scholar

[20] Wang Y, Liu W, Zhou J, et al. Discussion on Industrial Internet Platform Construction Based on Power Internet of Things[J]. Journal of Physics Conference Series, 2021, 1939(1):012077. Wang Y, Liu W, Zhou J, Discussion on Industrial Internet Platform Construction Based on Power Internet of Things[J]. Journal of Physics Conference Series, 2021, 1939(1):012077.10.1088/1742-6596/1939/1/012077 Search in Google Scholar

[21] Almarboui K, Almahdawi E. Oman rural grid extension employing renewable energy technology[J]. IOP Conference Series: Earth and Environmental Science, 2021. Almarboui K, Almahdawi E. Oman rural grid extension employing renewable energy technology[J]. IOP Conference Series: Earth and Environmental Science, 2021.10.1088/1755-1315/779/1/012035 Search in Google Scholar

[22] Xiong A, Tian H, He W, et al. A Distributed Security SDN Cluster Architecture for Smart Grid Based on Blockchain Technology[J]. Security and Communication Networks, 2021. Xiong A, Tian H, He W, A Distributed Security SDN Cluster Architecture for Smart Grid Based on Blockchain Technology[J]. Security and Communication Networks, 2021.10.1155/2021/9495093 Search in Google Scholar

[23] Wang X, Huang S, Aidong X U, et al. A Similarity Matching Based Data Privacy-Protection Scheme for Edge Terminal in Smart Grid[J]. Southern Power System Technology, 2020. Wang X, Huang S, Aidong X U, A Similarity Matching Based Data Privacy-Protection Scheme for Edge Terminal in Smart Grid[J]. Southern Power System Technology, 2020. Search in Google Scholar

[24] Kryukov A V, Cherepanov A V, Lyubchenko I A. Complex use of smart grid technology in railway traction networks[J]. IOP Conference Series: Materials Science and Engineering, 2021, 1151(1):012003 (10pp). Kryukov A V, Cherepanov A V, Lyubchenko I A. Complex use of smart grid technology in railway traction networks[J]. IOP Conference Series: Materials Science and Engineering, 2021, 1151(1):012003 (10pp).10.1088/1757-899X/1151/1/012003 Search in Google Scholar

Applied Mathematics and Nonlinear Sciences

Research on industrial Internet of Things and power grid technology application based on knowledge graph and data asset relationship model

Xue Wang,
Xue Wang,
Le Ma e
Le Ma e
Zhenwei Yang
Zhenwei Yang

Pubblicato online: 23 Dec 2022

Volume & Edizione: AHEAD OF PRINT

Pagine: -

Ricevuto: 08 Jul 2022

Accettato: 07 Sep 2022

Dettagli rivista e numero

Applied Mathematics and Nonlinear Sciences

Anteprima PDF

Articolo

Fig. 1

Fig. 2

Fig. 3

Figure e tabelle

Fig. 1

Fig. 2

Fig. 3

Model effect comparison (/%)

Riferimenti

Articoli recenti

Pianifica la tua conferenza remota con Sciendo

Applied Mathematics and Nonlinear Sciences

Research on industrial Internet of Things and power grid technology application based on knowledge graph and data asset relationship model

Xue Wang,Xue Wang,Le Ma eLe Ma eZhenwei YangZhenwei Yang

Pubblicato online: 23 Dec 2022

Volume & Edizione: AHEAD OF PRINT

Pagine: -

Ricevuto: 08 Jul 2022

Accettato: 07 Sep 2022

Dettagli rivista e numero

Applied Mathematics and Nonlinear Sciences

Anteprima PDF

Articolo

Fig. 1

Fig. 2

Fig. 3

Figure e tabelle

Fig. 1

Fig. 2

Fig. 3

Model effect comparison (/%)

Riferimenti

Articoli recenti

Pianifica la tua conferenza remota con Sciendo

Xue Wang,
Xue Wang,
Le Ma e
Le Ma e
Zhenwei Yang
Zhenwei Yang