Search results for "DTW"

showing 5 items of 5 documents

On the Non-Intrusive Load Monitoring in dwellings: a feasibility perspective

2021

The oncoming modernization process of the power grids, driven above all by decarbonisation objectives and the continuous improvement of digital technologies, is encouraging active participation in the electricity market by consumers through the Demand-Response mechanism. From this perspective, the introduction of smart meters and energy consumption monitoring devices plays a fundamental role, being able to give benefits to consumers, suppliers and the electricity grid itself. This paper proposes a supervised method of non-intrusive load monitoring (NILM) based on the recognition of patterns in the time domain with the Dynamic Time Warping algorithm which is suitable for low-cost smart meter…

Dynamic time warpingComputer scienceProcess (engineering)Energy consumptionHeuristic algorithmsPower system dynamicsEuropeLow-carbon economyElectricity supply industrySmart meterssmart gridnon-intrusive load monitoring (NILM)dynamic time warping (DTW)Electricity marketMetering modeTime domainEnergy consumptionLow-carbon economyIndustrial engineeringPower (physics)2021 IEEE International Conference on Environment and Electrical Engineering and 2021 IEEE Industrial and Commercial Power Systems Europe (EEEIC / I&CPS Europe)
researchProduct

Reāllaika laikrindu analīze prognozēšanai un anomāliju detektēšanai

2021

Šajā darbā tiek aprakstīts laikrindu anomāliju noteikšanas modeļa izstrādes process un tā realizācija. Darbs tiek balstīts uz temperatūras mērījumu sensoru datiem. Anomāliju noteikšanas modeļa izstrādes ietvaros tiek apskatītas sekojošas tēmas - simulāciju veidošanda, laikrindu analīze, laikrindu priekšapstrāde, laikrindu klasterēšana, laikrindu prognozējošo modeļu izveide, anomāliju noteikšana un modeļu ansambļa izveide. Darba mērķis ir apskatīt dažāda tipa modeļus, metodes un to apvienojumus, lai izveidotu robustu anomāliju noteikšanas modeļu ansambli. Darba rezultātā tika izveidots laikrindu anomāliju noteikšanas modeļu ansamblis, kura pamatā ir četri modeļi - LightGBM, LSTM, Holt-Winter…

Matemātikaanomāliju noteikšana laikrindāsDTWLSTMHolt-WintersLightGBM
researchProduct

Human Activity Signatures Captured under Different Directions Using SISO and MIMO Radar Systems

2022

In this paper, we highlight and resolve the shortcomings of single-input single-output (SISO) millimeter wave (mm-Wave) radar systems for human activity recognition (HAR). A 2×2 distributed multiple-input multiple-output (MIMO) radar framework is presented to capture human activity signatures under realistic conditions in indoor environments. We propose to distribute the two pairs of collocated transmitter–receiver antennas in order to illuminate the indoor environment from different perspectives. For the proposed MIMO system, we measure the time-variant (TV) radial velocity distribution and TV mean radial velocity to observe the signatures of human activities. We deploy the Anc…

direction-independent human activity recognition; fall detection; distributed MIMO; FMCW radar; micro-Doppler signatures; aspect angle; multistatic radar systems; passive step counter; DTW; velocity estimationFluid Flow and Transfer ProcessesProcess Chemistry and TechnologyGeneral EngineeringVDP::Medisinske Fag: 700General Materials ScienceVDP::Matematikk og Naturvitenskap: 400InstrumentationComputer Science::Information TheoryComputer Science ApplicationsApplied Sciences
researchProduct

An EEMD Aided Comparison of Time Histories and Its Application in Vehicle Safety

2017

In the context of signal processing, the comparison of time histories is required for different purposes, especially for the model validation of vehicle safety. Most of the existing metrics focus on the mathematical value only. Therefore, they suffer the measuring errors, disturbance, and uncertainties and can hardly achieve a stable result with a clear physical interpretation. This paper proposes a novel scheme of time histories comparison to be used in vehicle safety analysis. More specifically, each signal for comparison is decomposed into a trend signal and several intrinsic mode functions (IMFs) by ensemble empirical mode decomposition. The trend signals reflect the general variation a…

model validationDynamic time warpingGeneral Computer ScienceComputer science02 engineering and technologyHilbert–Huang transformEngineering (all)0203 mechanical engineeringVehicle safety0202 electrical engineering electronic engineering information engineeringIn vehicledynamic time warping (DTW)General Materials Sciencevehicle crashSimulationSignal processingdynamic time warping (DTW); Ensemble Empirical Mode Decomposition (EEMD); model validation; Time-history; vehicle crash; Computer Science (all); Materials Science (all); Engineering (all)Computer Science (all)General Engineering020302 automobile design & engineeringEnsemble Empirical Mode Decomposition (EEMD)Measurement uncertainty020201 artificial intelligence & image processingMaterials Science (all)lcsh:Electrical engineering. Electronics. Nuclear engineeringlcsh:TK1-9971AlgorithmTime-historyShape analysis (digital geometry)Motor vehicle crash
researchProduct

Odtwarzanie pamięci przeżywanej

2017

mythisationfictionalisationmemory reconstructionrelived memoryfabularyzacjapamięć przeżywanamuzealizacja historiiodtwarzanie pamięcimusealisation of historymityzacja
researchProduct