Overview of big data

　　Speaking of big data, it is estimated that everyone has only heard of the concept. In the big data era written by Viktor Mayer Schoenberg and Kenneth kukyer, four characteristics of big data are mentioned:

　　Large quantity: only when the data volume reaches the Pb level or above can it be called big data. 1PB equals 1024tb, 1TB equals 1024g, then 1PB equals 1024 * 1024 g data.

　　Great value: if you have more than 1PB of online data of all 20-35 young people in China, it will naturally have commercial value. For example, by analyzing these data, we can know their hobbies and guide the development direction of products. If we have the data of millions of patients across the country, we can predict the occurrence of diseases according to the analysis of these data. These are the value of big data.

　　Diversity: if there is only a single data, these data will have no value. For example, there is only a single personal data, or a single user submits data. These data can not be called big data, so big data also needs to be diverse. For example, among the current online users, everyone has different characteristics, such as age, education, hobbies, personality, etc, This is the diversity of big data. Of course, if it is extended to the whole country, the diversity of data will be stronger. There will be a variety of data diversity in each region and time period.

　　Fast speed: the logical processing speed of the data through the algorithm is very fast. The one second law can quickly obtain high-value information from various types of data, which is essentially different from the traditional data mining technology.

　　With the continuous improvement of people's requirements for living standards, big data has been widely used in all walks of life. In contrast, big data in the water industry has just started in recent years, and it is far from enough for data quality and mining. At this stage, water big data is more used to ensure the safety and production of water supply and drainage, which is still far from the deep-seated mining of big data in other industries. I think the reasons are as follows:

　　1. The deployment and construction coverage of infrastructure, especially sensing equipment, is still low. At present, it is only divided into blocks without forming a network;

　　2. It is difficult to guarantee the accuracy and stability of the perceived data, resulting in frequent deviations between the calculated results and the actual results;

　　3. The connection between data is not effectively opened, and there is no linkage between business layer and data layer;

　　4. Although there is a large amount of data, there is a lack of unified data requirements and standards, which makes it difficult to classify and integrate the data.

　　The above problems are also the urgent problems to be solved by water enterprises in the process of smart water construction. In such a large amount of data, there is a kind of data that can take the lead in meeting the requirements of big data analysis, and the data is closely related to people's livelihood, operation and production. This kind of data is user water volume.

　　Development of intelligent water meter

　　In the smart city, in addition to geographic information, the data of each user node is particularly important, which can reflect the dynamic data of each node user, that is, the data of three meters of water, electricity and gas, in which water is a necessity of life. Therefore, the water volume data is particularly important in the perception layer of the smart city.

　　As early as 2000, some domestic water enterprises began to try intelligent meter reading. With the birth of demand, the intelligent meter industry has developed vigorously, such as wired meter reading, IC card prepayment, photoelectric direct reading, camera direct reading, etc. the communication methods include GPRS, ZigBee, PDA, Nb IOT, Lora, etc. various technologies emerge in endlessly and are all popular for a time. With the development of technology, in recent years, the electronization of large-diameter water meters has gradually become a trend, including electromagnetic and ultrasonic water meters have gradually become the mainstream of the market. Therefore, no matter the consumption or the degree of electronic update, water volume data is undoubtedly the field with the most potential to take the lead in realizing big data analysis.

　　Multidimensional data application

　　With the acceleration of the construction of smart city and the gradual transformation of water enterprises, water volume data has evolved from a single reading demand to key data with multiple application significance.

　　Leakage analysis

　　In recent years, in order to better control leakage, most domestic water enterprises began to implement block management, and realized network management in some areas, that is, calculate the leakage between nodes according to the water volume of each node, so as to realize the management benefit of rapid positioning leakage. Moreover, based on the water volume of each water node, multi-level water volume verification can be realized to effectively control the measurement loss error.

　　Economic prosperity

　　In most countries, non resident water use is often a wind vane of urban economic prosperity. During the global financial crisis in 2008, the water consumption of cities, especially in coastal areas, became a good portrayal of the prosperity. At present, the electronic meter for non resident users will become a trend in the future

　　Livelihood analysis

　　In early 2020, there was no parallel in history. With the improvement of the epidemic situation, a large number of enterprises began to return to work. At this time, the water use data is a barometer for restoring order. Through the dynamic statistics of water volume, we can calculate the return to work rate of each city, count the return to work time, calculate the recovery degree of industry and commerce, and ensure the water quality safety of users who have been out of service for a long time. In addition, for residential water use, water volume data can help us understand the return rate of community residents, the real leakage level of community pipe network, the safety of the elderly, etc. it can be seen that water volume data can not only be used as an important reference for the safe production of water enterprises, but also directly go deep into the field of people's livelihood applications and be extended as a service form for the external output of water enterprises. This may be the development direction of smart water and even smart city.

　　The future of smart measurement

　　Of course, there are still many above application mining, but the closer the application is to people's livelihood, the higher the requirements for perceived data, which is undoubtedly a greater challenge for the measurement industry. We can not simply understand the intellectualization of measurement as solving the accuracy and transmission problems of a single measurement point, but should take the measurement nodes of the whole city as a system. In this system, we need to evaluate the data quality, data density and data dimension of the transmission system. Only in this way can we say that intelligent measurement has come.