data asset management system project
Mininglamp Technology has built for customer banks a next-generation data asset management system tailored to financial institutions. Based on the business-oriented data asset catalog and full-graph and global lineage and supported by the knowledge graph, machine learning, and natural language processing (NLP) technology, the system allows enterprise-level data to be managed and shared in customer banks.
Maintenance planning based on personnel's workload capacity helps the State Grid substation maintenance teams improve management precision;
Our solution helps to solve the problems such as unbalanced work planning, excessive reliance on manual calculation, high effort input, and high fault tolerance; realize a better assignment of tasks to staff to eliminate the reliance on manual effort in resource allocation; and quantize the staff workload to provide a basis for performance evaluation.
Fault knowledge bases built for substations increases the incident recognition rate and reduces maintenance costs
With the substation fault knowledge bases, the reliance on the monitoring alarm window is reduced to avoid the huge pressure on the monitoring screen in case of frequent faults. The problems such as insufficient recognition of alarm signals, inaccurate analysis, and non-standard operations have been solved, with the efficiency of processing alarm signals in real time increased greatly.
Intelligent fault diagnosis is realized at the State Grid substation maintenance center to eliminate the need for manual interference.
The time for searching for fault information is shortened from a few minutes to less than 1 minute. Fault forms are generated with just one click to reduce the workload of information collection by 85% for a single fault.
O&M prediction and risk evaluation are conducted through intelligent data analysis to reduce the O&M cost of steel manufacturing;
When building a PHM model, as opposed to the conventional method, our solution allows the equipment to obtain offline data through intelligent data analysis, without the need to train the model offline before putting it into use. This improves the model updating and iteration efficiency, and makes the response to equipment updates faster.