Batch Event Identification for Precise Execution Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Batch automation systems generate a mix of important and less relevant events during batch processes, making it difficult to align batch executions for meaningful analysis, as arbitrary events can obscure the progression of batch execution phases.
Innovation Solution
An automatic method identifies important batch events by determining the distance between time series of process variables across multiple batch executions, using algorithms like dynamic-time-warping (DTW) to select events with the smallest distance, which are then used for alignment, ensuring that only relevant events support alignment algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If all batch events generated by batch automation systems are used for alignment algorithms, then the quantity of events available for analysis increases, but the quality of alignment deteriorates due to inclusion of arbitrary or harmful events that do not reflect batch progression
Solution Approach 1:
The patent extracts and selects only the relevant batch events from the complete set of events generated by batch automation systems. The selection process identifies events that truly reflect batch progression (such as phase transitions and critical process points) while excluding arbitrary or harmful events, thereby improving alignment quality without losing essential information.
Solution Approach 2:
The patent applies different quality criteria to different events based on their local characteristics and relevance to batch progression. Instead of treating all events uniformly, the system evaluates each event's contribution to understanding batch phases and selects those with high informational value, creating a heterogeneous set of events with varying degrees of importance.
2Quantity of substance
If arbitrary batch events are included in alignment algorithms, then more data points are available for multivariate analysis, but the meaningfulness of analysis results deteriorates due to noise from irrelevant events
Solution Approach 1:
The patent converts the potentially harmful effect of including arbitrary events into a benefit by using them as a contrast reference. The selection process identifies meaningful events precisely by comparing them against the background of arbitrary events, using the noise to highlight the signal through statistical or pattern-based differentiation.
Solution Approach 2:
The patent applies partial action by selecting only a subset of events that are sufficient for meaningful analysis. Rather than using all available events or a fixed threshold, the system dynamically determines the optimal number and type of events needed to capture batch progression, avoiding both over-inclusion of noise and under-inclusion of relevant information.
3Ease of operation
If batch executions are aligned using arbitrary events, then alignment can be performed without selective event identification, but the comparability of operational phases deteriorates leading to less meaningful piece-wise comparison
Solution Approach 1:
The patent performs preliminary action by pre-identifying and selecting meaningful batch events before the alignment process. This preliminary selection creates a curated set of reference points that guide the subsequent alignment operation, ensuring that the ease of alignment does not compromise the quality of phase comparison. The meaningful events serve as anchors that structure the alignment process.
Data Source
AI summary
A method for automatic identification of important batch events for a batch execution alignment algorithm, including receiving historical batch data of a batch process, wherein the historical batch data comprises a plurality of batch executions, and wherein each of the plurality of batch executions comprises a plurality of batch events, indicating a specific event of the batch process, and at least one time series of a process variable, indicating a development of the process variable during the batch process; determining a distance between the at least one time series of the plurality of the batch executions for each of the plurality of batch events; and identifying at least one important batch event for a batch execution alignment algorithm with a smallest distance using the determined distances.