A Calculation Method of Stack Delay in Special Process of Silk Making Based on Distributed Lag Model
A technology of distributed lag model and calculation method, which is applied in the field of stack delay calculation in special process of silk making, can solve the problems of lack of methodological foundation and whether the unstacked delay setting is reasonable and verified, and achieves the effect of improving accuracy and good recognition effect.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0028] 1. Select the process data of the whole batch feeding process of the same production line of a certain brand in a factory throughout the year, and the data collection frequency is 1 time in 6s. Select the liquid flow rate (Y t ) as the explained variable of the model, the feeding process flow rate (X) and its 10-period lag as the explanatory variable, that is, X=(X t ,X t-1 ,...,X t-10 ). The maximum lag period of the model is determined by the process layout of the interval between the electronic scale and the nozzle of the feeder.
[0029] Using the error of the ten-fold cross-validation model, each batch selects the variable with the largest coefficient according to the size of the coefficient of the lag item, such as figure 1 As shown, when λ=0.0508, the mean square error of the model is the smallest. The coefficient estimates of each explanatory variable in the model are shown in Table 1. Among the 11 explanatory variables in the addition process, only X t-7...
Embodiment 2
[0034] 1. Select the process data of the whole batch of flavoring process in the same production line of a certain brand in a certain factory throughout the year, and the data collection frequency is 1 time per 6s. Select Spice Flow (Y t ) as the explained variable of the model, and the flow rate (X) of the flavoring process (X) and its 10-period lag as the explanatory variable, that is, X=(X t ,X t-1 ,...,X t-10 ). The maximum lag period of the model is determined by the process layout of the interval between the electronic scale and the aroma machine nozzle. Using the error of the ten-fold cross-validation model, each batch selects the variable with the largest coefficient according to the size of the coefficient of the lag item, such as figure 2 As shown, when λ=0.00078, the mean square error of the model is the smallest.
[0035] The coefficient estimates of each explanatory variable in the model are shown in Table 2. Of the 11 explanatory variables for the perfumin...
Embodiment 3
[0040] 1. Select the process data of the whole batch of cut stem blending process in the same production line of a certain brand in a certain factory throughout the year, and the data collection frequency is 1 time in 6s. Select stem flow (Y t ) as the explained variable of the model, the blending main scale process flow (X) and its 10-period lag as the explanatory variable, that is, X=(X t ,X t-1 ,...,X t-10 ). The maximum lag period of the model is determined by the process layout of the blending master scale and the blending scale. Using the error of the ten-fold cross-validation model, each batch selects the variable with the largest coefficient according to the size of the coefficient of the lag item, such as image 3 As shown, when λ=0.0010, the mean square error of the model is the smallest.
[0041] The coefficient estimates of each explanatory variable in the model are shown in Table 3. Among the 11 explanatory variables in the blending process of cut stems, onl...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More