The concept of digital employees originated from RPA, as a product of highly integrated artificial intelligence and business requirements, its functional areas and application scenarios are becoming increasingly widespread. For enterprises, how to break the boundary between humans and machines and create a human-machine coupling operation mode that suits their own needs and characteristics has been seen as the key to entering the next stage of industrial digital and intelligent transformation.
In traditional production environments, enterprises inevitably encounter repetitive and cumbersome operational steps. As a digital workforce, digital employees can help enterprises automatically perform repetitive and regular tasks. They can liberate people from business processes and enable them to invest in work that generates more benefits and value. It can be seen that the essence of the emergence of digital employees is not a "replacement", but the ultimate goal is to help enterprise employees complete their work more efficiently and create new value for the enterprise. So, where is the figure of digital employees in the clean energy field that emphasizes efficient innovation?
Jinfeng Energy Asset Management has implemented digital employee application practices in various business scenarios.
Hydrogen fueled unmanned aerial vehicle - automatic inspection for longer endurance
The application of drones in clean energy asset management scenarios has become very common, mainly concentrated in multiple aspects such as line inspection, photovoltaic inspection, and wind turbine blade inspection. Compared to traditional drone inspection operations, Goldwind Energy Asset Management has further upgraded and innovated its technology to address challenges such as automatic inspection, automatic fault identification, and improved endurance. For unmanned aerial vehicle inspection schemes in photovoltaic areas with large power station capacity, large number of components, and difficult to detect damage, incorporating core technologies such as data intelligent processing, automatic inspection and AI recognition, defect GPS positioning, and automatic report generation can greatly improve the inspection effect and defect elimination efficiency in photovoltaic areas. Combined with the centralized control operation of the booster station, on-site operators can be liberated to carry out defect elimination and power generation improvement work in photovoltaic areas. At present, the photovoltaic power stations managed by the North Operations Center of Goldwind Energy Asset Management have adopted a 230MW unmanned aerial vehicle intelligent inspection and diagnosis system. It is planned to complete the full coverage application of Goldwind Energy's 1000MW photovoltaic management by 2023.
Long endurance hydrogen fueled drones are used in the inspection of power transmission lines and wind turbines, while mesh wireless networks are deployed in the task execution area. Real time data transmission is carried out based on the mesh wireless network of the entire wind farm and photovoltaic power plant. During the automatic inspection process, drones complete tasks such as fine line inspection, non-stop inspection of wind turbine blades, infrared imaging inspection of photovoltaic modules, mountain fire prevention, and sea transportation of wind turbine spare parts, Realize intelligent inspection and automatic operation, reduce operation and maintenance labor costs. Using hydrogen fueled drones, compared to lithium-ion batteries, the endurance has been improved from 40 minutes to 2 hours, and a single flight can complete longer inspection tasks.
Signal Tower Digital Intelligent Monitoring Panel - Fault Messages Faster and Easier
The "Beacon Tower" digital intelligent monitoring system is a wind farm auxiliary monitoring tool that realizes automatic identification of fault messages and tracks the status of the entire process link for fault handling, effectively solving the problem of low efficiency in manual processing of massive messages. The system currently has 7 warning modules and 23 fault type recognition and push functions. It can classify and identify the collected fault messages based on the fault message recognition module after a fault occurs, and then use the notification platform module to push and distribute the fault messages. The "Beacon Tower" integrates multiple business systems, introduces artificial intelligence and big data models, and realizes automatic real-time duty of personnel monitoring for some work, laying a technical foundation for the unmanned monitoring mode. In general, it takes about 3 minutes for a single person to confirm the current status of a single fan fault, which is now 1 minute. Taking the tracking status of 150 fan faults per day as an example, it can save 5 hours per day.