Abstract
Understanding and acknowledging data analytics’s roles in modern industrial systems is good. There is also development in information and communication technology. There is an added layer for the conventional electricity consumption and distribution network, especially for data collection, analysis, and storage. Currently, there is a lot of comprehensive installation of sensors and smart meters. The bests known application of big data and data analysis has its application on smart grids (Das et al., 2018). Big data and smart grids have a wide range of data collection. They are a prelude in discussing and illustrating possible merits and motivations associated with implementing advanced data analytics. This paper will discuss and explain the advanced application of varied data analytics in the smart grid. The electricity network is known to deal with a mammoth amount of data. Big data brings benefits in electricity consumption, and a lot of benefits have also been brought to help the existing power system (Das et al., 2018). Big data and analytics have also brought customer service and social welfare improvement. In the advanced application of data analytics and big data when it comes to real smart grids, issues like awareness, techniques, and synergies, among others, need to be overcome.
Introduction
Cloud computing and digital technologies are fast developing. Much data is being processed daily and produced via digital equipment and sensors. Computers and smartphones are examples of advanced measuring infrastructure. The company activities can be improved by the existing rational and efficient data analysis where huge values and benefits can be enjoyed (Das et al., 2018). Big data is not a new term and has essential concepts meant for discovering valuable information from the large collected data, especially in commercial operations, to help with the knowledge to make business decisions. It is a tool for business intelligence. A lot has been achieved through the internet, like a reduction of time and cost in data acquisition (“Clustering of electricity consumption behavior dynamics towards big data applications,” 2018). Human society needs to be electrified, and it has been made possible by a smart and super grid which are reliable sources of renewable energy replacing the traditional fossil fuels, which are being depleted and facing challenges of de-carbonation. The smart grid is replacing the traditional electricity meters in the distribution systems, which generally had a small amount of data that would have been collected manually and analyzed for billing motives. Data analytics is needed as the smart grid needs a lot of data being mined and extracting valuable information for billing purposes and the state of the electricity network. The collected or mined data can be used to analyze the customer’s behaviors, optimize energy generation, and forecast demand. It is through data analytics that predictive maintenance and detection of faults can be noted (“Clustering of electricity consumption behavior dynamics towards big data applications,” 2018). The advanced metering infrastructure is essential to the power system’s security. An embedded information layer in the energy network helps produce a mammoth volume of data. It includes the measurement and control of the instructions in the grid for collection, transmission, analysis, and storage purposes. There are many opportunities that the smart grid brings on, especially in the data analysis platform.
The smart grid and big data for electricity consumption
The smart grid’s big data is uniformdespite lacking a standard definition. The big data in smart grid is an emerging technical challenge caused by data sets of mammoth volume .there are varied categories and complex structures that need a framework and some critical techniques to be able to excavate essential information effectively (“Clustering of electricity consumption behavior dynamics towards big data applications,” 2018). The big data definition is determined by the ability of data mining algorithms and the following hardware equipment to deal with large data sets.ICT technologies and the sizes of data, the smart grid is a string system embedded with a more uniform layer that permits two-way communication between the local actuators, central controllers, and the logistic units to correspond digitally for physical elements or urgent situations. According to the European Union, the smart grid is defined as the electricity network that can intelligently integrate the actions of all users connected to the electricity(Das et al., 2018). It combines the consumers, generators, and the two to efficiently deliver a more sustainable, secure, and economical electricity supply. According to the US definition of the smart grid for the future in a way that incorporates digital technology to improve the reliability, security, and efficiency concerning the electric system through information exchange, it touches on the distributed generation and solid resources for a complete automated power delivery. There is a call for green energy that is efficient, reliable and causes no harm to the environment, and it is replacing the traditional power system (Divina et al., 2021). Distributed generators are more widespread due to their ability to give clean energy as a resource. The distributed generators are also causing tremors and shaking up the hegemony position, especially in a large centralized Power plant. It causes the conventional centralized control strategy to be less efficient due to the unidirectional flow of lower.
Small-scale power generation connection, common to the public, distribution of the grid needs a two-directional operation and the distribution grid control (Das et al., 2018). Small-scale power generation has challenges with increased complexity in control and protection strategies. The challenge causes a need to enhance the conventional electro-mechanism electric grid with the assistance of innovations present in the ICT, which helps overcome the cost incurred in the case of a power outage or quality disturbance. With the SGAM, which represents the Smart grid architecture model in place, it is possible to assess the smart grid. The model is a three-dimensional framework combining the domains, layers, and zines. The conventional structure of a power system can easily be noted in the transmission, generation, distribution, distributed energy resources, and customer premises. Zones representing the layers of powers system management include the station, the field, the operation, the market, the process, and the enterprise.
The 5Vs in big data and smart grid
Big data is always broad and revolves around collecting, analyzing, processing, and storing data. Features or characteristics of big data is also experienced in smart grid. The features or characteristics of big data in electricity consumption under a smart grid follow the commonly known 5 Vs. in the big data model. They include veracity, volume, value, velocity, and variety.
Volume is a term about the large amount of data generated and makes the set of data too big to be stored. The analysis of it is via traditional database technologies. The result of a possible solution to the challenge is the way the system is distributed to store the data in separate locations (Misra & Bera, 2018). The data from varied locations are connected via the network, where they are brought together courtesy of the software. From the smart grid and the electricity consumption, it is true that the smart meter application and the advanced sensor technologies can harvest or bring on large amounts of data.
Velocity-velocity in a smart grid refers to the speed at which data is being generated and the speed at which data is being moved around. Curtsy of h technology in place, the real tie exchange of data is tremendously increasing (“secure data learning scheme for big data applications in the smart grid,” 2018). Studies show that the current sampling rate stands at four times an hour. A million installed sat meters in the smart grid will possibly result in about 35.04 billion records, which equates to 2090 tele bytes when quantified.
Variety-in smart grid, the variety in big data also manifests in the form of variety; it shows the types of data ready for usage. Traditionally we used to depend on structured data that would easily fit in tables and rational databases (“secure data learning scheme for big data applications in the smart grid,” 2018). . The current technologies relating to begging data call for the ability to handle varied types of data that are unstructured. Most data come from social media, messages, digital images, voice recordings, and sensor data (“secure data learning scheme for big data applications in the smart grid,” 2018). The technology can bring in board various data together and combine them with structured data. Data comes in different formats and dimensions as the data are diverse in structure.
Veracity –veracity is characteristic is big data used to show the trustworthiness or the extent of mess present in data .large amount of data, in most cases, can not be trusted, and technically it is true today that the accuracy and the quality are less trusted (Misra & Bera, 2018). The outcomes of the mined data can be doubted since smart grids can mine a lot of data simultaneously and may come in different formats. There are a lot of imperfections in the devices used, and there are also a lot of mistakes in the way data is retransmitted in the smart grid, and it is true to say that the errors are always present in the smart grid (Misra & Bera, 2018). There is a need to have secure and more efficient power system operations and to depend on the assessment of data and the way they are estimated.
Value-value is a characteristic of big data, and it is not an exemption in the smart grid for electricity consumption. It refers to the ability and capability to extract valuable information from a mammoth amount of data and get a clear understanding of the value it brings. Mammoth data sets have a lower or decreased density of valuable information (“secure data learning scheme for big data applications in the smart grid,” 2018). The intelligent devices are improving, leading to the adoption of the smart grid, translating to an increase in data value in analytics from the many applicants.
Sources of data
The smart grid is part of artificial intelligence. A smart grid is an important information source covering electricity generation, consumption, distribution, and transmission data. Their data sources include the electrical data from the distribution station, electricity meters, distribution switch station, markets, and regional economic data, , which comprise the non-electrical information (Das et al., 2018). Analysts of the collected data help the power plants schedule, subsystems operations, vital power equipment maintenance, and understanding the market behaviors. The data sources mentioned can effectively be sorted into three essential classes: business data, measurement data, and external data. Installed sensors and smart meters are the ones that measure most of the operation parameters, and they are the ones to show both historical and current system status (Das et al., 2018). Smart meters cannot measure the weather conditions and events conducted socially, but the two can impact the operation planning in the power system. Marketing strategies and behaviors of a rival are the business data.
Data mining techniques
The smart grid is known o use a variety of collection techniques. The smart meter can collect and transmit data. The data is usually energy-related and gives information to customers and utility companies. For residential customers’ energy consumption, the smart meter reading is likely to go up .there are also emerging components that are consuming electricity in eh market. They include electric vehicles and plug-in hybrid EVs which are becoming more popular with the movement toward electrification in the transport sector and artificial intelligence (“Data mining techniques for segmentation,” 2019). The normal operation status was initially controlled at the distribution system, which the DSO used to depend on primary substation measurements. The typical smart meters have their measurements in node voltage, power factor, feeder current, total harmonic distortion, and reactive power, among others. There are techniques used in data communication in smart grids, including the HAN, NAN, and WAN as a network. Some essential communication technologies can also be used, and they are wireless and wired infrastructures. The wireless technologies allow the data center to gather information on measurements from the smart meter at a relatively low cost and simplify the connection, which may face electromagnetic challenges (“Data mining techniques for segmentation,” 2019). Power line communication is an example of wired communication technology ad has modulated carrier signal that the power cables implemented in the power system
Conclusion
The big data in electricity consumption is something interesting in the market. There are smart meters that are vital in measuring consumption and the status of electricity usage. The mechanism used in data mining also goes hand in hand with the 5vs as characteristics of big data. Increased connection and the emergence of electric vehicles are also increasing energy consumption. A smart grid can be a tool used to mine data and serve as a reliable source of information.
References
Clustering of electricity consumption behavior dynamics towards big data applications. (2018). International Journal of Recent Trends in Engineering and Research, 4(4), 102-106. https://doi.org/10.23883/ijrter.2018.4212.eavsd
Das, H., Barik, R. K., Dubey, H., & Roy, D. S. (2018). undefined. Springer.
Data mining techniques for segmentation. (2019). Data Mining Techniques in CRM, 65-132. https://doi.org/10.1002/9780470685815.ch3
Divina, F., F A., & García-Torres, M. (2021). Advanced optimization methods and big data applications in energy demand forecast. MDPI.
Misra, S., & Bera, S. (2018). Smart grid technology: A cloud computing and data management approach. Cambridge University Press.
A secure data learning scheme for big data applications in the smart grid. (2018). Smart Grid Communication Infrastructures, 187-203. https://doi.org/10.1002/9781119240136.ch9
write