OPTIMIPRO | Copyright 2025.
The adaptive real-time optimization with Dynamic Correlation Integration (DCI) technology is developed based on the principles of random process mathematics. It uses real-time dynamic data collected from the production process to achieve closed-loop, online, and real-time optimization without modeling. It has strong adaptive capabilities and does not hinder its optimization function regardless of changes in raw materials or production changes caused by technological transformation. Moreover, the one-time investment is not high, and the enterprise can achieve long-term returns.
The novelty of this technology has been internationally recognized, and currently it has obtained invention patents in China, the United States, Canada, South Korea, and Singapore.
? Application conditions：
The Dynamic Correlation Integral (DCI) optimization method is a purely mathematical approach that is independent of the specific process mechanism and has universality. In principle, as long as the following conditions are met, the DCI optimzation method can be applied:
1、It must be a continuous production process. This method cannot be applied to intermittent production processes.
2、The optimized objective function is measurable or computable online. Due to the need for a continuous curve of the objective function over time for DCI, laboratory test data that occurs every few hours cannot be used.
3、The process is controlled by DCS. Due to the 24-hour continuous closed-loop operation of optimization control, high reliability requirements are placed on computer control systems. Distributed control systems can meet these requirements.
? Technical Features
The DCI optimization methods have the following characteristics that differ from previous real time optimization techniques:
1. It is not necessary to establish static and dynamic models of the process in advance.
2. Strong automatic adaptability, able to automatically track the optimal point under changes in process conditions, raw material changes, equipment modifications, and other situations.
3. Utilize the natural fluctuations of the normal operation of the process for work, without the need to add test signals to the process, thus causing minimal interference to the process operation.
4. This method has strong anti-interference characteristics, and can even work normally under harsh conditions such as dynamic strong interference, where other factors such as raw material properties cause changes in the objective function that are greater than the useful signal (changes in the objective function caused by tuning variables).
5. This method has strong robustness and can still work normally when the dynamic and static characteristics of the process drift and change.
6. The maintenance workload is very small.
7. The main operating conditions can be smoothly adjusted to the optimal level while ensuring the safety of the production process.
8. The computational workload is large and requires the cooperation of DCS.
9. It has been running for a long time in many large petrochemical plants, and has undergone changes in raw materials, equipment modifications, etc., but still can adapt automatically in practical applications.
? The system aims to achieve the overall economic benefits, single product yield, comprehensive product yield, energy consumption, etc. of the Fluid Catalytic Cracking (FCC) unit. It automatically adjusts the key process conditions, such as reaction temperature, feed temperature, catalyst addition amount, etc., adjusts and optimizes these parameters in real time, and finds the best match online under the specified optimization objectives.
? This system does not require pre-established reaction models and has strong adaptive capabilities. It can automatically adapt to equipment modifications, changes in raw materials and catalyst, etc., and automatically searches for new process optimal conditions based on these alternations.
? This system can maximal the yield of dewaxing oil or dewaxing wax, and automatically adjust key process conditions such as primary solvent ratio, secondary solvent ratio, tertiary solvent ratio, and cold wash ratio of dewaxing filter based on total solvent ratio as constraints. During the optimization process, control the pressure drop of the crystallization system within an appropriate range.
? This system does not require the establishment of a model. Based on the measured dynamic data of various solvent ratios and crystallizer pressure drops, the optimization target is optimized in real time using the Dynamic Correlation Integration optimization technology. It can self adapt to changes in the properties of various raw materials and automatically track the optimal operating conditions according to the changes in raw materials.
? The system aims to optimize the aromatic hydrocarbon yield, octane number yield, or overall economic benefits of the catalytic reforming product. By optimizing the four reaction temperatures of the reforming reactor in real-time, the reaction proceeds in the desired direction, thereby improving the aromatic hydrocarbon yield, octane number yield, or economic benefits of the device.
? The system is equipped with an online Raman spectrometer to analyze the composition of the materials after the reforming reaction, assisting in the calculation of optimization objectives and inference of reaction direction online.
? This system does not require the establishment of a system reaction kinetics model and can automatically adapt to changes in raw material properties, automatically find the best match for four reaction temperatures, and if the device is modified or the catalyst is changed, it will not affect its optimization function.
? The Dynamic Correlation Integration self adaptive real-time optimization method is a universal real-time optimization technique for continuous production processes, and therefore can be applied to many practical production devices, such as most petroleum processing processes.
? This system can optimize key operating conditions or process operating parameters in real-time based on specific production equipment and processing flow, after selecting optimization goals and processing plans.
? Due to the fact that the DCI optimization system does not require modeling, it is particularly suitable for systems that are difficult to model, such as the optimization of complex chemical reaction processes, or processes with particularly complex structures and unclear mechanisms.
? One characteristic of this system is its strong adaptive ability, which can automatically adapt to situations where the actual model drifts or changes in equipment modification, raw material changes, etc., without the need for human intervention, thus ensuring the long-term stable operation of the online optimization system in engineering practice.