Table 4 lists the optimal solution vectors for the VEP-NLMM-L of different inflow dividing levels (L = 1–5). The SSQs calculated by the VEP-NLMM-L of different L have obvious reductions, at least 88.23% (L = 1) lower than the value obtained by the LMM-L. For VEP-NLMM-L with L = 5, the optimal inflow dividing values are ,
,
,
.
Table 4
Optimal parameter vectors by estimating different Muskingum models for River Wyre flood, October 1982
| Model | K | w | α | m/mi (i = 1,…,L) | SSQ | |
|---|---|---|---|---|---|---|
| LMM-L (O'Donnell 1985) | 5.5400 | 0.1370 | 2.6200 | – | 468.840 | |
| VEP-NLMM-L (this study) | L = 1 | 5.1485 | 0.2282 | 2.5285 | 1.0000 | 55.193 |
| L = 2 | 4.9473 | 0.2524 | 2.5212 | 1.0120, 1.0059 | 28.462 | |
| L = 3 | 4.7858 | 0.2518 | 2.5225 | 1.0188, 1.0102, 1.0134 | 26.239 | |
| L = 4 | 4.2020 | 0.2730 | 2.5142 | 1.0135, 1.0498, 1.0458, 1.0398 | 20.490 | |
| L = 5 | 3.7303 | 0.2856 | 2.5166 | 1.0477, 1.0841, 1.0772, 1.0722, 1.0658 | 17.574 | |
| Model | K | w | α | m/mi (i = 1,…,L) | SSQ | |
|---|---|---|---|---|---|---|
| LMM-L (O'Donnell 1985) | 5.5400 | 0.1370 | 2.6200 | – | 468.840 | |
| VEP-NLMM-L (this study) | L = 1 | 5.1485 | 0.2282 | 2.5285 | 1.0000 | 55.193 |
| L = 2 | 4.9473 | 0.2524 | 2.5212 | 1.0120, 1.0059 | 28.462 | |
| L = 3 | 4.7858 | 0.2518 | 2.5225 | 1.0188, 1.0102, 1.0134 | 26.239 | |
| L = 4 | 4.2020 | 0.2730 | 2.5142 | 1.0135, 1.0498, 1.0458, 1.0398 | 20.490 | |
| L = 5 | 3.7303 | 0.2856 | 2.5166 | 1.0477, 1.0841, 1.0772, 1.0722, 1.0658 | 17.574 | |