Soft measurement method for key indexes in complex industrial process

Abstract

The invention discloses a soft measurement method for key indexes in a complex industrial process, and the method comprises the following steps: 1, acquiring visual data and process parameter data, and obtaining values of the corresponding key indexes through later laboratory tests; 2, constructing a video data space-time sequence feature extraction network model on video data by adopting a cross-frame fusion convolutional neural network (DCFNN) provided by the invention, and establishing a video data space-time sequence feature extraction network model based on the video data space-time sequence feature extraction network model based on the DCFNN; 3, constructing a time sequence feature extraction model on the industrial parameter data by adopting a GRU network; 4, inputting the collecteddata and process parameters, carrying out dual-channel concurrent extraction on feature data with a time sequence, and carrying out key index soft measurement by adopting a full-connection network ofan attention mechanism, and carrying out reverse model training according to the real key index value obtained by assay; 5, carrying out real-time estimation on key indexes which are difficult to monitor on line by adopting the trained soft measurement model.

Description

technical field [0001] The invention belongs to the field of online monitoring of key indicators of complex industrial processes, and in particular relates to a soft measurement method for key indicators of complex industrial processes based on dual-channel fusion of machine vision and engineering parameters. Background technique [0002] In modern industrial processes, real-time monitoring of key indicators is of great significance to ensure process safety and production quality. However, in many cases, the key indicators of complex industrial processes are difficult to detect online due to problems such as long process flow, unclear internal mechanism, and many influencing factors. [0003] In recent years, soft sensing technology has been widely used in online monitoring of key indicators in complex industrial processes due to its advantages of fast response, low maintenance cost, and accurate prediction results. In complex industrial processes, machine vision has been w...

AI Technical Summary

Problems solved by technology

[0006] From the above analysis, it can be seen that due to the complexity of complex industrial processes and many influencing factors, the existing manual monitoring cannot accurately monitor the quality of its products, and the key indicators are often difficult to detect online, which leads to low yield rates of industrial products and low utilization of raw materials. Low efficiency, high resource consumption, etc.

Images