[SMM PV Flash] Sungrow Patches iSolarCloud Authentication Vulnerability

Spanish energy company Cox has signed a US$2.9 billion investment agreement with the government of El Salvador covering solar, battery storage, natural gas and water infrastructure. The solar component, to be developed with the Lempa River Hydroelectric Executive Commission (CEL), includes two projects with an initial combined capacity of around 200MW and potential expansion to 400MW. Grid-connected battery storage would initially reach around 440MWh to 500MWh and could later expand to 1GWh. Around US$500 million is planned for the solar and storage components, while Cox intends to use project finance involving international financial institutions, multilateral lenders and export credit agencies. The programme combines large-scale PV with battery storage to support renewable energy integration and grid flexibility.

Dutch startup Rable has launched its Rable4Roofs self-supporting PV mounting system for older and structurally limited rooftops. The galvanized steel lattice structure distributes module and mounting loads to roof purlins rather than concentrating them on the roof deck and can be installed on roofs with slopes of up to around 10°. According to the company, the design can reduce the number of anchoring points by about 95%, with system loads ranging from approximately 6kg/m² to 15kg/m². The solution could expand rooftop PV deployment on commercial and industrial buildings that cannot accommodate conventional mounting systems because of structural load limitations.

Universiti Teknologi Malaysia (UTM), together with research institutions in Canada, Belgium, Norway and India, has developed an open image dataset for AI-based navigation of PV cleaning robots. The dataset contains 6,200 annotated module images covering clean surfaces, different levels of simulated dust and real laboratory contamination, with structural reference features such as cell gaps and module frames labelled for robotic navigation. The research focuses on whether computer vision systems can reliably identify navigation references under conditions including dust, water stains, bird droppings and glare. The dataset, published in Data in Brief, could support the development and validation of autonomous PV cleaning systems for more complex field environments.