[Retracted] Prediction Model of Rotor Yarn Quality Based on CNN-LSTM

Table 3

Comparison table of predicted values of rotor yarn strength by three algorithms. Unit: cN/tex.

Serial number	Actual value	CNN-LSTM	BP	Regression algorithm
1	16.4	16.4	16.0	17.0
2	16.1	16.1	15.9	16.0
3	15.4	15.5	15.0	16.1
4	14.9	14.9	14.0	15.5
5	14.8	14.7	14.6	15.1
6	14.8	14.9	14.8	15.5
7	15.6	15.4	15.0	15.0
8	15.2	15.2	14.9	14.5
9	14.7	14.8	14.3	14.0
10	14.5	14.6	14.2	15.0
11	14.8	14.7	14.2	15.2
12	14.8	14.7	14.0	14.0
13	15.2	15.2	14.8	14.5
14	14.8	14.7	15.1	15.3
15	14.9	15.0	15.3	15.4
16	15.4	15.2	15.8	16.0
17	15.7	15.8	16.0	16.2
18	15.7	15.7	16.0	16.2
Mean absolute difference		0.080	0.400	0.556
Average relative error		0.008	0.026	0.037
Mean square error (MSE)		0.010	0.203	0.336
Root-mean-square error		0.110	0.451	0.579
square		0.990	0.870	0.560