Penicillin fermentation process quality related fault detection method

Abstract

The invention relates to a penicillin fermentation process quality related fault detection method. The method comprises the steps of collecting multiple batches of data in a normal operation condition, dividing the data into a process data set and a quality data set, and performing expansion and standardization to obtain new process data set and quality data set of normal operation; performing multi-subspace typical variable analysis on the process data and the quality data of the normal operation to obtain five subspaces and five statistics variables, and calculating a threshold of monitoring statistics; collecting real-time running data, and performing standardization to obtain new process data set and quality data set; and executing the multi-subspace typical variable analysis on the real-time running data, and by analyzing the condition that the monitoring statistics of the five subspaces exceed the threshold, obtaining a fault detection result. According to the method, based on typical variable analysis, and by further considering variance information and performing five-subspace mapping on original data, fine fault detection can be realized, so that whether the quality is influenced by process faults or not is judged.

Description

technical field [0001] The invention belongs to the technical field of industrial process fault detection and relates to a fault detection method for intermittent industrial processes, in particular to a quality-related fault detection method for a penicillin fermentation process. Background technique [0002] In order to adapt to the ever-changing market demand in modern society, modern industrial production is gradually relying on the intermittent industrial production process of producing multiple varieties, small batches and high added value. Compared with the continuous process, the mechanism of the batch industrial production process is complex and the operation complexity is high, and the product quality is easily affected by uncertain factors such as environmental conditions, raw material quality, and equipment status. Typical batch industrial production processes include pharmaceuticals, fermentation, and semiconductor processing, and products produced by batch indu...

AI Technical Summary

Problems solved by technology

[0005] The present invention aims at the shortcomings of being unable to distinguish whether the fault causes abnormal product quality and low fault detection effect in fault detection of intermittent industrial processes, and provides a simple, accurate and effective detection method for quality-related faults in the penicillin fermentation process, which can judge whether the process fault is Leading to abnormal product quality, it can be finely monitored and the fault detection effect can be improved

Images