A Reproducible Blockchain-Audited Workflow for Multi-Source Power Data Governance

This article presents a reproducible data governance framework for multi-source heterogeneous data in new-type power systems. The proposed protocol integrates a lifecycle-oriented governance matrix, configurable stream-and-batch pipelines, a hierarchical quality rule repository, and permissioned blockchain auditing. To demonstrate the workflow, diverse datasets—comprising SCADA, smart-meter, renewable-generation, weather, and phasor measurement unit (PMU) data—are processed and systematically evaluated. Comparative analyses against traditional extract-transform-load (ETL), rule-based cleaning, centralized governance, and blockchain-only approaches reveal significant performance improvements. Specifically, the framework achieves 96.8% schema-mapping accuracy, 94.1% temporal-alignment accuracy, and 95.3% cross-source consistency. Record-level anomaly detection attains an accuracy of 0.956, while cross-source integrity recall reaches 0.950, culminating in an overall quality score of 0.912. Beyond data quality, the configuration-driven architecture enhances engineering efficiency by reducing new-source integration time to 4.3 hours, yielding a 91.8% configuration reuse rate, and ensuring a 98.4% cross-environment reproduction success rate. Furthermore, the integrated blockchain mechanism guarantees a 99.1% tampering detection rate with acceptable computational and storage overhead. Ultimately, this methodology offers an accurate, efficient, and auditable solution to secure data trustworthiness in increasingly data-intensive smart grids.