NNLS

AcronymDefinition
NNLSNon-Negative Least Squares
NNLSNovell Nterprise Linux Services
NNLSNew North London Synagogue (London, England, UK)
NNLSNorfolk and Norwich Law Society (Norfolk, England, UK)
NNLSNational Numeracy and Literacy Strategy
References in periodicals archive ?
However, a scaling problem arises in the respective NNLS problem when t is too big.
Let us formulate the NNLS problem (12) corresponding to this system of inequalities:
The problem (26) is transformed to the NNLS problem (12) and solved using the algorithm QPLS.
Three finite algorithms for solving the NNLS problem are described, based on QR decomposition of the coefficients matrix.
Redundant inequalities, theorems of alternative, and linear and quadratic programming problems are studied using the NNLS problem.
Comparison of Average Ramping DeltaQ Results to Actual Leakage for 28 Lower Envelope Leakage Tests Analysis Method DeltaQ efm [L/s] Actual efm [L/s] Supply Pressure scanning 56 [26] 66 [31] Pressure fitting, 57 [27] 66 [31] no corrections Pressure fitting, 54 [25] 66 [31] with corrections NNLS 69 [32] 66 [31] Return Pressure scanning 13 [6] 12 [6] Pressure fitting, 18 [8] 12 [6] no corrections Pressure fitting, 14 [7] 12 [6] with corrections NNLS 25 [12] 12 [6]
Summary of RMS Errors for Alternative DeltaQ Analysis Techniques Analysis Method For all 46 Tests For 28 Lower (cfm) [L/s] Envelope Leakage Tets (cfm) [L/s] Pressure scanning 22 [10] 23 [11] Supply Pressure fitting, 41 [19] 15 [7] no corrections Pressure fitting, 22 [10] 17 [8] with corrections NNLS n/a 12 [6] Return Pressure scanning 24 [11] 10 [5] Pressure fitting, 52 [10] 11 [5] no corrections Pressure fitting, 29 [14] 9 [4] with corrections NNLS n/a 15 [7]
Summary of Mean, RMS and Average Absolute (AA) Errors in Total Leakage for Alternative DeltaQ Analysis Techniques For all 46 Tests (cfm) For 28 Lower Envelope Leakage [L/s] (Mean Leakage Tests (cfm) [L/s] (Mean Leakage Flow = 102 cfm [48 L/s]) Flow = 78 cfm [37 L/s]) Analysis Mean RMS AA Mean RMS AA method Pressure 2 [1] 41 [19] 27 [13] -22 [-10] 30 [14] 24 [11] scanning Pressure 23 [11] 83 [39] 48 [23] -17 [-8] 26 [12] 20 [9] scanning With 1 [1] 43 [20] 32 [15] -22 [11] 32 [15] 25 [12] corrections NNLS n/a n/a n/a 0[0] 25 [12] 18 [8]
At sites 1 and 2, the automated software was used to take the ramping data and analyze it using the NNLS technique.
The test results for site 1 show that the Ramping test, analyzed with the NNLS method, resulted in higher leakage flows than the ASTM E1554 style tests.
The higher bias for NNLS is because it intrinsically cannot have negative results that would tend to offset a positive bias.
Again, the higher bias for NNLS was because it could not have negative results.