Green ammonia online optimization method based on improved case-based reasoning algorithm

By improving the case-based reasoning algorithm, a steady-state case library was established based on green ammonia production data. Compactly correlated variables were identified and posterior estimation was performed, which solved the problems of high model complexity, slow computation, and poor adaptability in the existing technology. This enabled rapid and accurate online optimization of green ammonia production and improved production efficiency.

CN116755412BActive Publication Date: 2026-06-12SICHUAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SICHUAN UNIV
Filing Date
2023-07-24
Publication Date
2026-06-12

AI Technical Summary

⚠Technical Problem

In existing technologies, the selection of model variables for large-scale renewable energy electrolysis of water to produce hydrogen and synthesize ammonia relies on human expert experience, which may ignore influencing variables, increase model complexity, slow calculation speed, low convergence, and make it difficult to obtain accurate optimization results under various variable load conditions. Furthermore, the robustness and interpretability are not strong.

⚗Method used

An improved case-based reasoning algorithm is adopted. A steady-state case library is established based on DCS and LIMS data of the green ammonia production process. The CMIM-GIEF method of information entropy is used to identify compactly correlated variables. Posterior estimation is performed by combining MCMC sampling to optimize the product yield target, reduce model complexity, and improve computational efficiency and adaptability.

🎯Benefits of technology

It enables rapid and accurate online optimization in the green ammonia production process, reduces model complexity, improves calculation speed and convergence, has strong adaptability, and can provide accurate optimization suggestions under variable load conditions, improving production efficiency by more than 10%.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116755412B_ABST
    Figure CN116755412B_ABST
Patent Text Reader

Abstract

The application provides a kind of green ammonia online optimization method based on improved case reasoning algorithm, comprising: based on green ammonia production process DCS and LIMS data, input, production process and output variable data are collected and processed, and steady state working condition case library is established;Based on the test input and similar input near neighbor case, the posterior distribution of the optimal yield of the process variable corresponding to the input is calculated, and then the maximum posterior estimation of each compact process variable is obtained;Compact posterior estimation based on MCMC sampling is carried out.The application not only effectively identifies and analyzes the key process variables, greatly reduces the complexity of the process model, but also improves the accuracy and calculation efficiency of process optimization, can intelligently select the variables strongly related to the target variables from a large number of variables, reduce the model complexity to the minimum, fast operation speed, high convergence, strong adaptability.
Need to check novelty before this filing date? Find Prior Art