Temporal Velocity Gap Detection: A Statistical Framework and Benchmark for Cross-Domain Research Transfer in Smart Grids

We propose TVGD (Temporal Velocity Gap Detection), a statistical framework for detecting cross-domain research transfer opportunities via velocity differentials on dynamic bipartite graphs. TVGD formalizes the detection of concept-methods that are actively researched in a source domain but underrepresented in a target domain, using stratified permutation tests with Benjamini-Hochberg false discovery rate correction. We construct the first benchmark for evaluating transfer detection targeting smart grids, with automated citation-path ground truth. This pre-registration specifies the framework, metrics, comparisons, and analysis plan before ground truth construction. This is Step 1 of a two-step registration protocol. Step 1 registers the framework before ground truth construction. Step 2 will amend with realized values of P, N, and computed numeric thresholds, filed after ground truth construction but before any test-set evaluation.