{"id":2651,"date":"2022-03-28T11:35:48","date_gmt":"2022-03-28T06:05:48","guid":{"rendered":"https:\/\/smarthelio.com\/?p=2651"},"modified":"2023-09-12T16:43:19","modified_gmt":"2023-09-12T11:13:19","slug":"clean-millions-of-data-points-within-secs-with-data-automation-tool","status":"publish","type":"post","link":"https:\/\/smarthelio.com\/clean-millions-of-data-points-within-secs-with-data-automation-tool\/","title":{"rendered":"Clean millions of data points within seconds with AI-based data automation for solar plants"},"content":{"rendered":"
[vc_row][vc_column][vc_column_text uncode_shortcode_id=”101726″]<\/p>\n
<\/span><\/p>\n Missing data and poor data quality can have serious impact on your solar analytics and overall performance of your solar PV plant. Read to know how you can solve this issue using AI-based Data Automation technology.<\/em><\/p>\n As solar companies are shifting towards data-driven decisions, it becomes critical to have good quality data to ensure<\/span> analysis is correct and actionable.<\/span><\/p>\n What is the role of data in solar PV plant’s performance?<\/strong><\/p>\n Data not only helps you understand how your solar PV plant is performing in general but also helps to compare and analyze how different sections of the solar plant are performing in comparison to each other using performance indicators like PR, SY, soiling, degradation, etc. For example, data can tell you precisely if a panel is degrading faster than the other panel installed at the same site due to a shadow from the nearby tree. Data can derive intelligence for multiple stakeholders: Sales, engineering, procurement, management and investors. In short, data intelligence empowers a solar company to take informed decisions and plan their solar O&M in a better way.\u00a0<\/span><\/p>\n The curious case of missing solar data points\u00a0<\/strong><\/p>\n Generally, a MW size of a solar PV plant produces 100,000 data points in a month. However, as reported by our customers in multiple geographies, 16-20% data points are usually missing<\/strong> in this data due to network or connection failure and\/or faulty readings by the data loggers. Because of the huge quantity of data produced otherwise, missing data<\/a> is usually ignored (an easy way out), thereby creating a void that leads to incorrect conclusions.\u00a0<\/span><\/p>\n For example, missing data during the sunshine hours can effectively lead to low PR value in days of high energy production, or measurements during night time hours can give a high CUF value which is misleading.\u00a0<\/span><\/p>\n Also, it is quite common to receive outliers\/ bad quality measurements from data loggers, pyranometers due to soiling, bad weather conditions, calibration issues, etc. Generally, as a part of data cleaning, this data is removed from the dataset leading to data loss.\u00a0<\/span><\/p>\n AI-based Data Automation tool for any kind of solar PV plant<\/strong><\/p>\n