Real-time optimization method and system for chemical production device and electronic equipment

Abstract

The invention provides a real-time optimization method and system for a chemical production device and electronic equipment, and relates to the field of data processing. The method comprises the following steps: acquiring mechanism data and real-time operation data of a chemical production device in a production process; according to the mechanism data, a mechanism model is constructed and obtained, and the mechanism model is used for reflecting material transfer, energy balance and chemical reaction in the chemical production device; according to the real-time operation data, a prediction model is constructed and obtained, and the prediction model is used for predicting a nonlinear relation between an input variable and an output variable of the chemical production device; fusing the mechanism model and the prediction model to obtain a fusion model; and processing the target optimization parameter through the fusion model to obtain an optimization result. By implementing the technical scheme provided by the invention, the real-time optimization precision of the chemical production device can be conveniently improved.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and particularly relates to a real-time optimization method, system and electronic device for a chemical production device. Background Art

[0002] With the acceleration of the global industrialization process, the chemical industry occupies an important position in the modern economy. The operating efficiency, economic benefits and safety of chemical production devices directly affect the competitiveness of enterprises and the consumption level of energy resources. In this context, the optimization problem of chemical production devices has gradually become a research hotspot in the industry.

[0003] Currently, most of the related optimization methods are limited to the application of a single model, or a linear model is combined with other models through a simple weighting method. However, the application of a single model or a simple combination often has limitations. For example, a linear model often ignores non-linear characteristics and complex operating conditions, so the accuracy of the optimization results provided is relatively low.

[0004] Therefore, there is an urgent need for a real-time optimization method, system and electronic device for a chemical production device. Summary of the Invention

[0005] This application provides a real-time optimization method, system and electronic device for a chemical production device, which is convenient for improving the accuracy of real-time optimization of chemical production devices.

[0006] In the first aspect of this application, a real-time optimization method for a chemical production device is provided. The method includes: obtaining mechanism data and real-time operation data during the production process of the chemical production device; constructing a mechanism model based on the mechanism data, where the mechanism model is used to reflect the mass transfer, energy balance and chemical reactions inside the chemical production device; constructing a prediction model based on the real-time operation data, where the prediction model is used to predict the non-linear relationship between the input variables and output variables of the chemical production device; fusing the mechanism model and the prediction model to obtain a fusion model; and processing target optimization parameters through the fusion model to obtain optimization results.

[0007] By adopting the above technical solutions, the mechanism model is based on the theoretical basis of mass transfer, energy balance, and chemical reactions, and can accurately reflect the basic principles of chemical processes. Through this model, the core physical and chemical laws in the process can be accurately described. The prediction model, driven by data, can effectively capture the complex non-linear relationship between input and output. Especially when facing highly complex and dynamically changing working conditions, it can provide a fast response and relatively accurate prediction. The fusion model combines the physical nature of the mechanism model and the adaptability of the prediction model. During the optimization process, it can take into account both the basic laws of the process and the actual operating conditions, thus achieving higher-precision optimization and avoiding the error situation of a single model under certain working conditions. In the chemical production process, the operating conditions of the device often change dynamically, making it difficult for traditional static optimization methods to cope. By integrating the mechanism model and the prediction model, the system can process real-time operation data in real time and make rapid adjustments according to process changes and dynamic working conditions. The prediction model can learn from real-time data to cope with changes in working conditions and adjust the optimization strategy to ensure the best operating state under different working conditions. Although the mechanism model can provide an accurate theoretical framework, in practical applications, it often faces uncertainties. The prediction model can enhance the adaptability to these uncertain factors by learning a large amount of actual operation data. By integrating these two models, while retaining the accuracy of the mechanism model, the prediction model can be used for dynamic correction and adjustment, thereby improving the robustness of the system. Even when facing data noise or partial data loss, it can still maintain good optimization performance. In addition, the fusion model can effectively improve the response speed of real-time optimization by balancing the complexity of the mechanism model and the calculation speed of the prediction model in real-time calculations. During the production process, when the working conditions change, the system can quickly respond, adjust the control strategy, and optimize the operating parameters to ensure production efficiency and safety. Therefore, it is convenient to improve the accuracy of real-time optimization of chemical production devices.

[0008] Optionally, the obtaining of the mechanism data and the real-time operation data of the chemical production device during the production process specifically includes: determining the physical principles and chemical principles of the chemical production device; based on the physical principles and the chemical principles, determining the process behavior of the chemical production device during the production process to obtain the mechanism data; obtaining the raw data of the chemical production device during the production process in real time; and preprocessing the raw data to obtain the real-time operation data, where the preprocessing includes denoising, filtering, and normalization processing.

[0009] By adopting the above technical solutions, by determining these basic principles, the accuracy of the mechanism model can be ensured, and predictions or optimization results that do not conform to the actual situation can be avoided in complex production environments, thereby improving the reliability and practical application effect of the model. The capture of this process behavior provides a profound theoretical basis for optimizing the model, enabling it to better reflect the dynamic characteristics and variability of the production process and enhancing the adaptability of the optimization model in dynamic changes. By obtaining the original data in the production process in real time, the operating status of the device can be dynamically monitored. These original data provide valuable input information for the prediction model, enabling it to capture subtle changes in actual production and enhancing the real-time nature of optimization decisions. By preprocessing the original data, the accuracy and consistency of the data can be ensured. Denoising and filtering processes help eliminate random noise caused by equipment errors or environmental fluctuations, making the data more stable and reliable; normalization processing can ensure that data with different dimensions and ranges have the same influence in the model, avoiding inaccurate optimization models due to different data scales. These processed real-time data provide feedback on the actual operation for the mechanism model, helping to make optimization adjustments according to the current actual working conditions, enabling the fusion model to more accurately predict and optimize the operating status of the system. Through the collection and preprocessing of real-time operation data, the optimization process can dynamically adjust optimization decisions according to the latest production data, rather than relying solely on the assumptions of the theoretical model. The real-time nature and accuracy of the data enable the optimization model to flexibly respond to changes in the production process, thereby improving the accuracy of the optimization results. Combining mechanism data with real-time data can address the limitations of mechanism models in the face of uncertainty and complex working conditions.

[0010] Optionally, constructing the mechanism model according to the mechanism data specifically includes: determining the basic reaction relationships between substances inside the chemical production device according to the mechanism data; based on the basic reaction relationships, determining the feed parameters, output parameters, intermediate product parameters, and by-product parameters of the chemical production device; determining the complex reaction relationships between substances inside the chemical production device according to the mechanism data; based on the complex reaction relationships, determining the heat transfer parameters, heat loss parameters, and mass transfer parameters of the chemical production device; under natural constraint conditions, generating the mechanism model according to the feed parameters, output parameters, intermediate product parameters, and by-product parameters, and the heat transfer parameters, heat loss parameters, and mass transfer parameters.

[0011] By adopting the above technical solutions, the basic reaction relationships between substances are the core of constructing the mechanism model, which helps to determine how reactants are converted into products. Through these relationships, reaction kinetic equations can be established, and further information such as the reaction rate and conversion rate of each reaction can be derived. Based on these reaction relationships, by determining the feed parameters, output parameters, as well as the parameters of intermediate products and by-products, the material flow and changes in the entire production process can be accurately simulated. This not only helps to understand the conversion efficiency of the reaction, but also provides support for optimizing aspects such as product quality and yield. Heat transfer, heat loss, and mass transfer parameters are the key factors describing the energy and material exchange processes in chemical plants. The heat and mass transfer in chemical production processes are often the result of the combined action of multiple factors. By defining and calculating these parameters, it can be ensured that the mechanism model not only describes the chemical process of the reaction, but also takes into account the thermal effects of the equipment and the mass transfer efficiency, which plays an important role in optimizing energy efficiency, increasing the reaction rate, and controlling the reaction temperature. By modeling complex reaction relationships, a more realistic chemical reaction process can be simulated, which not only improves the accuracy of the model, but also enables the optimization and control strategies to remain effective under changing operating conditions. When generating the mechanism model, compulsorily following these constraints ensures that the model does not produce results that do not conform to reality. For example, mass in a chemical reaction cannot be created or disappeared out of thin air, and the heat transfer process cannot violate the first law of thermodynamics, etc. By following these natural constraints, the generated mechanism model can be closer to actual operations and avoid unreasonable predictions in theoretical models.

[0012] Optionally, constructing a prediction model according to the real-time operation data specifically includes: obtaining the historical operation data of the chemical production device; determining the real-time operation data and the historical operation data as the input variables, and processing them using a support vector machine algorithm to obtain the output variables; obtaining the input variables and output variables corresponding to each of multiple training processes, and evaluating the training processes through an evaluation algorithm to obtain the prediction model.

[0013] By adopting the above technical solution, the real-time operation data can reflect the state in the current production process, while the historical operation data provides a record of past operating conditions. Through the combination of the two, a more comprehensive understanding and prediction of the behavior of chemical production plants can be achieved. Real-time data can provide immediate feedback on the current operation, while historical data provides long-term behavior patterns under different operating conditions. The combination of the two helps to capture more comprehensive system characteristics. The support vector machine can effectively handle non-linear relationships by using kernel functions, and thus can provide more accurate predictions under complex operating conditions. By obtaining the input variables and output variables during multiple training processes and using an evaluation algorithm to evaluate the training process, it can be ensured that the model reaches the optimal state during training. The evaluation algorithm usually helps to screen out the most suitable parameters, avoid overfitting and underfitting problems, and make the final prediction model more accurate and robust. The evaluation process can help detect and optimize the performance of the model, further improving its prediction accuracy and stability. Through evaluation, potential problems in model training, such as overly complex or inappropriate assumptions for the current dataset, can be identified, and the model can be adjusted to better adapt to the production environment. Through this process, it can be ensured that the model not only performs well on the training set, but also has strong prediction ability in actual applications. By obtaining data in real time and applying the prediction model, immediate feedback and prediction results can be provided for the chemical production process. This real-time nature enables the production process to be dynamically adjusted, optimize operating conditions, and prevent failures or inefficiencies from occurring.

[0014] Optionally, the fusion of the mechanism model and the prediction model to obtain a fusion model specifically includes: constructing a population containing the mechanism model and the prediction model, the population including multiple individuals, and each individual representing a combined structure of one model; performing differential mutation on a target individual to obtain a mutant individual, the target individual being any one of the multiple individuals; performing a crossover operation on the target individual and the mutant individual to obtain a candidate individual; comparing the magnitude relationship between the fitness value corresponding to the candidate individual and the fitness value corresponding to the target individual; if it is determined that the fitness value corresponding to the candidate individual is greater than or equal to the fitness value corresponding to the target individual, then replacing the target individual with the candidate individual and entering the next generation population; repeating the differential mutation, the crossover operation, and the comparison until a preset number of iterations is reached or the fitness value converges, so as to obtain the fusion model.

[0015] By adopting the above technical solutions, a population containing a mechanism model and a prediction model is constructed. The method uses various different model combination methods as individuals. Each individual represents a different model structure, so that various possible model combination methods can be explored, and the fusion potential of the mechanism model and the prediction model can be fully exploited. The differential mutation operation generates new individuals by mutating the target individuals. This mutation operation can introduce new structures and parameters, increasing the diversity of the population. The mutation process enables the algorithm to jump out of the local optimum, explore a wider solution space, and enhance the adaptability and robustness of the fusion model. Through the crossover operation, the target individual and the mutated individual exchange part of their structures to generate candidate individuals. This operation helps to retain the characteristics of excellent individuals and combine the advantages of different individuals, which may produce a better model structure. By comparing the fitness values of the candidate individuals and the target individuals, the differential evolution algorithm can select a more suitable model combination by optimizing the objective function. An individual with a higher fitness value means that the model has a better performance in the actual task, so it is more in line with the optimization goal. Such an evaluation mechanism ensures that each iteration develops in the direction of the optimal fusion model. The fitness evaluation can help the algorithm optimize the model targeted and gradually improve the prediction accuracy and reliability of the fusion model.

[0016] Optionally, processing the target optimization parameter through the fusion model to obtain an optimization result specifically includes: determining a target input variable and a target output variable from the target optimization parameter; inputting the target input variable into the fusion model to obtain a feature variable, where the feature variable includes a feature vector; judging the positional relationship between the feature vector and the hyperplane of the support vector machine model corresponding to the fusion model; calculating a first probability value and a second probability value based on the positional relationship, where the first probability value is used to represent the probability value corresponding to the feature vector on the first side of the hyperplane, and the second probability value is used to represent the probability value corresponding to the feature vector on the second side of the hyperplane, and the first side and the second side are two opposite sides of the hyperplane; if it is determined that the feature vector is located on the first side and the first probability value is greater than or equal to a preset threshold, then determining that the feature variable is the target output variable, and determining that the optimization result is normal; if it is determined that the feature vector is located on the second side and the second probability value is greater than or equal to the preset threshold, then determining that the feature variable is not the target output variable, and determining that the optimization result is abnormal.

[0017] By adopting the above technical solution, by determining the target input variables and output variables from the target optimization parameters, the system can focus on the key parameters in the optimization process, avoid interference from irrelevant data, and thus improve the accuracy of the optimization results. By inputting the target input variables into the fusion model, the advantages of the mechanism model and the prediction model can be fully utilized to obtain more accurate and comprehensive feature variables, ensuring that the optimization results better meet the actual production requirements. The construction of the hyperplane enables the support vector machine to effectively distinguish different categories in the high-dimensional space, thus achieving more efficient decision-making. The support vector machine can accurately perform classification and regression. Through the segmentation of the hyperplane, the optimization results are clearly classified in the model, providing a basis for high-precision judgment. By setting a preset threshold, the system can automatically screen out the optimization results that meet the standards and exclude abnormal situations. The setting of the threshold provides a mechanism for controlling risks, ensuring that only the optimization results that meet certain conditions are considered normal, further enhancing the reliability of the decision-making. By judging the positional relationship between the feature vector and the hyperplane, the abnormal situations of the optimization results can be automatically identified. For example, when the optimization result is on the other side of the hyperplane and the probability value does not reach the preset threshold, the system will automatically judge that the optimization result is abnormal. This function can effectively prevent human omissions and reduce the possibility of incorrect judgments.

[0018] Optionally, the method further includes: receiving an optimization request uploaded by a user through a user device, where the optimization request includes the target optimization parameters; and optimizing the target optimization parameters according to the optimization request.

[0019] By adopting the above technical solution, the user can directly participate in the optimization process by uploading an optimization request through the device. This method allows the user to actively request optimization according to their own needs and actual production situations, enhancing the interactivity and flexibility of the system. The user can submit new optimization goals at any time according to real-time data and working conditions changes, increasing the controllability of the production process. By receiving the uploaded optimization request, the cumbersome interaction steps between the user and the system are avoided, and the optimization process can be started more quickly and efficiently. This saves time for the user during operation and improves work efficiency. The user can upload an optimization request according to specific production requirements or problems encountered. Each optimization request can be targeted at different production goals, thus effectively improving production efficiency. The system processes the target optimization parameters according to the request uploaded by the user, enabling an accurate response to the current production conditions.

[0020] In a second aspect of the present application, a real-time optimization system for a chemical production device is provided. The real-time optimization system includes an acquisition module and a processing module. Among them, the acquisition module is used to acquire mechanism data and real-time operation data during the production process of the chemical production device; the processing module is used to construct a mechanism model based on the mechanism data, and the mechanism model is used to reflect the mass transfer, energy balance, and chemical reactions inside the chemical production device; the processing module is further used to construct a prediction model based on the real-time operation data, and the prediction model is used to predict the non-linear relationship between the input variables and output variables of the chemical production device; the processing module is further used to fuse the mechanism model and the prediction model to obtain a fusion model; the processing module is further used to process the target optimization parameters through the fusion model to obtain an optimization result.

[0021] In a third aspect of the present application, an electronic device is provided. The electronic device includes a processor, a memory, a user interface, and a network interface. The memory is used to store instructions. Both the user interface and the network interface are used to communicate with other devices. The processor is used to execute the instructions stored in the memory so that the electronic device executes the method described above.

[0022] In a fourth aspect of the present application, a computer-readable storage medium is provided. The computer-readable storage medium stores instructions, and when the instructions are executed, the method described above is executed.

[0023] In summary, one or more technical solutions provided in the present application have at least the following technical effects or advantages: The mechanism model is based on the theoretical basis of mass transfer, energy balance, and chemical reactions, and can accurately reflect the basic principles of chemical processes. Through this model, the core physical and chemical laws in the process can be precisely described. The prediction model, driven by data, can effectively capture the complex non-linear relationship between input and output. Especially when facing highly complex and dynamically changing operating conditions, it can provide a rapid response and relatively accurate prediction. The fusion model combines the physical nature of the mechanism model and the adaptability of the prediction model. During the optimization process, it can take into account both the basic laws of the process and the actual operating state, thus achieving higher-precision optimization and avoiding errors that may occur in a single model under certain operating conditions. In the chemical production process, the operating conditions of the device often change dynamically, making it difficult for traditional static optimization methods to cope. By integrating the mechanism model and the prediction model, the system can process real-time operating data in real time and make rapid adjustments according to process changes and dynamic operating conditions. The prediction model can learn from real-time data, respond to changes in operating conditions, and adjust the optimization strategy to ensure the best operating state under different operating conditions. Although the mechanism model can provide an accurate theoretical framework, in practical applications, it often faces uncertainties. The prediction model, through learning a large amount of actual operating data, can enhance its adaptability to these uncertain factors. By integrating these two models, while retaining the accuracy of the mechanism model, the prediction model can be used for dynamic correction and adjustment, thereby improving the robustness of the system. Even when facing data noise or partial data loss, it can still maintain good optimization performance. In addition, the fusion model can effectively improve the response speed of real-time optimization by balancing the complexity of the mechanism model and the calculation speed of the prediction model in real-time calculations. During the production process, when the operating conditions change, the system can quickly respond, adjust the control strategy, and optimize the operating parameters to ensure production efficiency and safety. Therefore, it is convenient to improve the accuracy of real-time optimization of chemical production devices. Description of the Drawings

[0024] Figure 1 It is a schematic flowchart of a real-time optimization method for a chemical production device provided by an embodiment of the present application.

[0025] Figure 2 It is another schematic flowchart of a real-time optimization method for a chemical production device provided by an embodiment of the present application.

[0026] Figure 3 It is a schematic block diagram of a real-time optimization system for a chemical production device provided by an embodiment of the present application.

[0027] Figure 4 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

[0028] Explanation of the reference numerals: 31, acquisition module; 32, processing module; 41, processor; 42, communication bus; 43, user interface; 44, network interface; 45, memory. DETAILED DESCRIPTION

[0029] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0030] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0031] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0032] With the rapid advancement of the global industrialization process, the chemical industry has gradually occupied a vital position in the modern economic system. As one of the basic industries supporting all walks of life around the world, the operating efficiency, economic benefits and safety of chemical production equipment directly determine the market competitiveness of enterprises, and are also related to the rational use of energy resources and environmental protection. In this context, the optimization problem of chemical production equipment has gradually become one of the core issues in the research field of the chemical industry. How to improve the comprehensive benefits of production equipment and reduce energy consumption and waste emissions through effective optimization methods has become the focus of major chemical companies.

[0033] However, most of the current research and application of chemical production device optimization methods are limited to the use of a single model, or simply combine different types of models through weighted methods. For example, many optimization methods still rely on linear models. Although linear models can provide quick solutions when dealing with some simple working conditions, their applicability is often limited by the non-linear characteristics and complex working condition changes in actual production. When facing a complex production environment, the accuracy of their optimization results is relatively low. To solve the above technical problems, this application provides a real-time optimization method for chemical production devices, referring to Figure 1 , Figure 1 FIG. is a schematic flow chart of a real-time optimization method for a chemical production device provided by an embodiment of this application. This real-time optimization method is applied to a server and includes steps S110 to S150. The above steps are as follows: S110. Obtain the mechanism data and real-time operation data of the chemical production device during the production process.

[0034] Specifically, the server is a full-process device for monitoring the production operation of the chemical production device, which can be a computer system or a computer cluster. Mechanism data refers to the data reflecting physical and chemical processes such as mass transfer, heat transfer, and chemical reactions in the chemical production device. These data are obtained through experiments, theoretical derivations, or physical model calculations and are used to describe the basic operation mechanism within the system. Mechanism data can reveal the behavior of the production device under different working conditions, including raw material conversion, reaction rate, and the flow of substances and heat energy. For example, reaction temperature, pressure, flow rate, reactant and product concentrations, etc. all belong to mechanism data. They reflect the basic characteristics of chemical reactions, heat transfer, and material conversion processes in the reactor. Real-time operation data refers to the process parameters and equipment status information collected in real-time from the production device. These data are obtained based on monitoring devices such as sensors and instruments and can dynamically reflect the actual operation situation during the production process. Real-time operation data can reflect the real-time operation status of the equipment, the actual execution situation of the process, and the health status of the system. For example, equipment operating temperature, pressure sensor data, flowmeter readings, energy consumption data, etc. are all real-time operation data. They can display the current process conditions of the production device in real-time and help operation and maintenance personnel determine whether the device is operating normally.

[0035] In a possible implementation manner, obtaining the mechanism data and real-time operation data of the chemical production device during the production process specifically includes: determining the physical and chemical principles of the chemical production device; based on the physical and chemical principles, determining the process behavior of the chemical production device during the production process to obtain mechanism data; obtaining the raw data of the chemical production device during the production process in real-time; and preprocessing the raw data to obtain real-time operation data, where the preprocessing includes denoising, filtering, and normalization processing.

[0036] Specifically, physical principles refer to the basic physical laws involved in chemical production plants. For example, fluid dynamics, laws of thermodynamics, heat and mass transfer processes, etc. These principles describe phenomena such as the flow of substances and heat transfer in the plant. Chemical principles refer to the chemical reaction processes in the plant, including reaction kinetics, equilibrium, catalytic reactions, etc. Through these chemical principles, it is possible to understand how raw materials are converted into products and how chemical reactions occur. Based on the above physical and chemical principles, a mechanism model of the chemical production plant can be established, which can describe the reaction process, heat transfer, and mass transfer inside the plant. For example, in a reaction kettle, how reactants are converted into products, how heat is transferred through a heat exchanger, and how substances flow through pipelines, etc. Mechanism data is obtained from these principles through theoretical calculations or experimental data, and can reflect the basic behaviors in the production process, such as reaction rate, temperature change, pressure distribution, etc. Raw data refers to the process data collected in real-time by monitoring devices such as sensors and instruments, usually raw signals without any processing. For example: temperature sensor data inside the reaction kettle, pressure values read by pressure sensors, liquid or gas flow rates collected by flow meters, and reactant concentrations recorded by material concentration sensors. These data are digital signals that reflect the current state of the production process in real-time, but they directly come from the production site and may contain noise or irregular fluctuations.

[0037] Among them, raw data may be disturbed by equipment errors, environmental factors, etc., resulting in noise in the data. Denoising refers to eliminating these interferences through techniques such as filtering and averaging. For example, the rapid fluctuations in temperature sensor data are eliminated by the moving average method. Filtering is to remove the unwanted frequency components in the data and retain the information meaningful for system analysis. For example, a low-pass filter can be applied to remove the high-frequency noise in temperature sensor data. In order to enable raw data from different sources or with different dimensions to be compared or processed under the same standard, normalization processing is required. Normalization is to convert data with different dimensions into a unified magnitude, for example, converting temperature data, flow rate data, and pressure data into standardized values for subsequent processing.

[0038] S120. According to the mechanism data, a mechanism model is constructed, and the mechanism model is used to reflect the mass transfer, energy balance, and chemical reactions inside the chemical production plant.

[0039] Specifically, a mechanism model is a mathematical model constructed based on mechanism data, which can quantitatively describe the processes of mass transfer, energy balance, and chemical reactions inside a device. It does not solely rely on historical data or empirical rules, but combines physical and chemical principles to deduce the behavior of the system. Mass transfer refers to the process in chemical production where substances transfer from one location or phase to another. For example, in the distillation process, the volatile components in the liquid evaporate into the gas phase and then condense back into a liquid. The mechanism model will detail various parameters in this process, such as gas-liquid equilibrium, mass transfer rate, concentration distribution, etc. Energy balance refers to how various forms of energy, such as thermal energy and mechanical energy, flow and transform within the system in a chemical process. In a reactor, heat is input into the system through a heater, and at the same time, a portion of the heat will flow out through a cooling device. Heat is also released or absorbed during the reaction process. In the mechanism model, the energy balance equation can describe the heat flow, heat loss, temperature change, etc. in this process to help optimize the efficiency of the heating and cooling systems. A chemical reaction refers to the process where reactants are converted into products, usually accompanied by the release or absorption of heat. In a reaction kettle or other reaction equipment, the reaction rate, conversion rate, and generation of by-products in a chemical reaction are closely related to factors such as temperature, pressure, and reactant concentration. The mechanism model describes the changes of various variables during the reaction process by considering factors such as reaction kinetics and reaction rate constants.

[0040] In a possible implementation manner, based on the mechanism data, a mechanism model is constructed, specifically including: determining the basic reaction relationships between substances inside the chemical production device according to the mechanism data; determining the feed parameters, output parameters, intermediate product parameters, and by-product parameters of the chemical production device based on the basic reaction relationships; determining the complex reaction relationships between substances inside the chemical production device according to the mechanism data; determining the heat transfer parameters, heat loss parameters, and mass transfer parameters of the chemical production device based on the complex reaction relationships; and generating the mechanism model under natural constraint conditions according to the feed parameters, output parameters, intermediate product parameters, and by-product parameters, as well as the heat transfer parameters, heat loss parameters, and mass transfer parameters.

[0041] Specifically, the basic reaction relationship refers to the simple and fundamental reaction equations or reaction paths between substances in a chemical reaction. For example, in a chemical reaction, two or more substances react with each other to form products, and the mechanism model helps describe how the reaction occurs through these basic reaction relationships. Feed parameters are the characteristics of the substances entering the reaction system, such as concentration, flow rate, temperature, etc. Through the mechanism model, the input conditions of different reactants can be determined. Output parameters are the characteristics of the products generated by the reaction, including concentration, flow rate, etc. Intermediate product parameters are the concentrations of transient substances generated during the reaction process. By-product parameters are the concentrations of additional substances that are not desired to be generated during the reaction process. For example, in the ethylene cracking reaction, the feed parameters may include the flow rate, temperature, and pressure of ethylene, the output parameters include the concentrations and flow rates of acetylene and hydrogen, and the by-products may be some undesired low-molecular hydrocarbons or carbon black.

[0042] In many chemical reactions, there are not only simple basic reaction relationships, but also complex interactions and reaction paths between multiple substances. These complex reaction relationships involve multi-step reactions and changes in reaction rates. These complex reaction relationships are deduced through more in-depth experimental data, reaction kinetics models, and reactor designs in the mechanism data. For example, in the ethylene cracking reaction, in addition to the basic reaction, there may be multiple reaction paths, and the reaction rate may be strongly affected by factors such as temperature and pressure. The mechanism model needs to consider these complex reactions, such as side reactions and pyrolysis reactions in ethylene cracking.

[0043] In chemical reactions, the input and output of heat directly affect the reaction rate and effect. The mechanism model helps predict the temperature distribution in the reactor by describing the heat transfer. The heat loss parameter refers to the heat lost by the reactor and its components during operation. These lost heats affect the energy efficiency of the entire reaction system. In a chemical reaction, substances need to be transferred between different phases, such as gas phase and liquid phase, solid phase and liquid phase, etc. The mass transfer parameter helps us understand how substances migrate between different regions during the reaction process. For example, in the ethylene cracking reaction, the temperature of the reactor is a key parameter. The mechanism model needs to consider how heat is transferred to the reactants through the heater and how to keep the temperature in the reactor constant. At the same time, there may be a mass transfer process in the reactor, such as the diffusion and flow of gases.

[0044] Among them, natural constraint conditions refer to the operating conditions and physical limitations of equipment during the actual production process. For example, the maximum temperature, pressure, and flow rate limitations of a reactor, etc. Based on known feed parameters, output parameters, intermediate product parameters, complex reaction relationships, heat transfer and mass transfer parameters, etc., a complete mechanism model is generated under actual operating conditions and equipment limitations. For example, in the ethylene cracking reaction, the operating temperature of the reactor is between 800 - 1000 °C, and the pressure is maintained within a certain range, such as under high pressure. The mechanism model needs to incorporate these constraint conditions to ensure that the model reflects the reaction behavior and results under these conditions.

[0045] S130. Based on real-time operation data, a prediction model is constructed. The prediction model is used to predict the non-linear relationship between the input variables and output variables of a chemical production plant.

[0046] Specifically, real-time operation data refers to the data collected in real-time during the actual operation of a chemical production plant, including dynamic data measured by various sensors and instruments. This data includes information reflecting the state of the production process, such as feed flow rate, temperature, pressure, concentration, reaction rate, etc. Real-time operation data is crucial for the prediction model because it reflects the actual operation of the plant. For example, assume that in a reactor, the real-time operation data may include: feed flow rate: 50 m 3 / h, the temperature inside the reactor: 400 °C, pressure: 20 bar, gas concentration: 80%, ethylene: 10%, 10% other impurities.

[0047] Among them, the prediction model infers future states or output results based on existing real-time operation data through machine learning algorithms. The prediction model learns a large amount of historical data to identify complex patterns or relationships between inputs and outputs. During this process, the model automatically adjusts its parameters to minimize the prediction error. The main role of this model is to predict how output variables in the production process, such as product concentration, yield, etc., change with the change of input variables, such as raw material flow rate, reaction temperature, etc. The non-linear relationship means that the relationship between input variables and output variables is not a simple linear relationship, but has a certain degree of complexity. For example, the relationship between temperature and the reaction rate of the product may not be linear; after the temperature rises to a certain critical value, the yield may increase rapidly, while at a certain temperature, the yield is stable. The variable relationships in many chemical production processes are often non-linear because they may be affected by the interaction of multiple factors and reaction conditions.

[0048] For example, in a chemical reaction, the relationship between the reaction temperature and the reaction rate is non-linear. As the temperature increases, the reaction rate may start to accelerate, but above a certain temperature threshold, the reaction rate may tend to level off or even exhibit a thermal runaway phenomenon. The prediction model needs to learn from historical data to understand that such non-linear input variables are factors that can be controlled or measured in a chemical production plant, such as raw material flow rate, temperature, pressure, etc. The output variables are the results of the reaction process, such as product concentration, yield, energy efficiency, etc. By constructing a prediction model, the model can predict the corresponding output values based on the inputs in the real-time operation data.

[0049] In one possible implementation, a prediction model is constructed based on the real-time operation data, which specifically includes: obtaining the historical operation data of the chemical production plant; determining the real-time operation data and the historical operation data as input variables, and using the support vector machine algorithm for processing to obtain output variables; obtaining the input variables and output variables corresponding to multiple training processes, and evaluating the training processes through an evaluation algorithm to obtain the prediction model.

[0050] Specifically, the historical operation data refers to the data collected on the chemical production plant over a past period of time. These data include feed flow rate, temperature, pressure, concentration in the reactor, product quality, etc., and can be used to analyze the performance under different conditions during the production process. The input variables include the real-time operation data and the historical operation data, which will serve as the inputs to the support vector machine model to help the model predict future states or outputs. The support vector machine is a machine learning algorithm used for classification and regression analysis. In the embodiments of the present application, the support vector machine is used for regression analysis, that is, to learn the relationship between the input and the output from the input data to predict the output variables of the chemical production plant, such as product concentration, output, etc.

[0051] Among them, the support vector machine finds an optimal hyperplane in a high-dimensional space so that the categories of input data can be accurately distinguished or predicted. In regression problems, the support vector machine learns the relationship between input variables and output variables, generates a prediction model, and can accurately predict future outputs. For example, in the application of chemical reactors, the support vector machine can be used to predict product concentration or reaction rate based on historical and real-time data. For instance, the input variables may include reaction temperature and flow rate, and the output variable is the concentration of the product. Through the regression analysis of the support vector machine, the model can learn how temperature and flow rate affect the product concentration. When constructing a prediction model, the support vector machine needs to be trained multiple times. Each time during training, the model uses different input variables and learns according to the corresponding output variables. The purpose of training is to enable the support vector machine to gradually find the optimal mapping relationship between input and output. For example, in each training, the input variables may vary. Some trainings use temperature and flow rate as inputs, while others may include more factors, such as reactor pressure or substance concentration. After each training, the model calculates the corresponding output, such as product concentration, and compares it with the actual results to optimize the model.

[0052] In addition, evaluation algorithms are used to verify the accuracy and reliability of the support vector machine model, including cross-validation, mean square error, etc., which are used to measure the performance of the prediction model on different training data. The evaluation process helps to determine whether the trained model can accurately predict on new data. If the evaluation result of the model is good, it indicates that it can accurately predict the relationship between input variables and output variables. For example, assume that the cross-validation during the training process shows that the support vector machine model based on historical and real-time data can accurately predict the product concentration of the reactor. When the evaluation result shows that the model can predict new production data, it can be confirmed that the model is effective. Finally, the support vector machine that has undergone multiple trainings and evaluations will generate a prediction model. This model can predict output variables based on new real-time data and historical data. The output of the model can be used to guide the optimization of the actual production process. For example, this prediction model can be used for the real-time monitoring of chemical production plants. When new real-time data is input into the model, the model can predict the upcoming product concentration, helping operators to adjust production parameters in a timely manner, thereby optimizing the production process.

[0053] S140. Integrate the mechanism model and the prediction model to obtain a fusion model.

[0054] Specifically, by combining the mechanism model with the prediction model, the server can integrate the advantages of both, thereby improving the accuracy and stability of prediction. The mechanism model provides a physical understanding of the behavior of the device, while the prediction model can capture features such as non-linearity and time-variation based on data-driven. The integration of the two helps to more comprehensively describe the operation law of the chemical device and make up for the limitations of a single model. For example, through optimization algorithms such as genetic algorithms and differential evolution algorithms, the output of the mechanism model and the prediction model can be automatically adjusted to make the final integration result best conform to the behavior of the actual device.

[0055] In a possible implementation manner, the mechanism model and the prediction model are fused to obtain a fusion model, which specifically includes: constructing a population containing the mechanism model and the prediction model, the population includes multiple individuals, and each individual represents a combined structure of a model; performing differential mutation on the target individual to obtain a mutant individual, where the target individual is any one of the multiple individuals; performing a crossover operation on the target individual and the mutant individual to obtain a candidate individual; comparing the magnitude relationship between the fitness value corresponding to the candidate individual and the fitness value corresponding to the target individual; if it is determined that the fitness value corresponding to the candidate individual is greater than or equal to the fitness value corresponding to the target individual, then the candidate individual replaces the target individual and enters the next generation population; repeating differential mutation, crossover operation and comparison until the preset number of iterations is reached or the fitness value converges to end, so as to obtain the fusion model.

[0056] Specifically, the population refers to a set of possible model combined structures. Each individual represents a combination of a mechanism model and a prediction model. For example, you can combine different weights of the mechanism model and the prediction model into different individuals. Each individual represents a potential solution in the population. For example, assume that the mechanism model is used to describe the temperature change of the reactor, and the prediction model predicts the concentration of the product in the reactor through historical data. An individual in the population may represent the combination of the mechanism model and the prediction model, where the weight of the mechanism model accounts for 70% and the weight of the prediction model accounts for 30%. The server performs differential mutation on each target individual to generate a mutant individual. Differential mutation means creating a new candidate solution by adding the differences between other individuals in the population to the target individual. The mutant individual may make certain adjustments to the structure or parameters of the model. Assume that the weight of the mechanism model of the target individual is 70% and the weight of the prediction model is 30%. Differential mutation can add the combined weight difference of other individuals, such as the weight of the mechanism model of an individual is 60% and the weight of the prediction model is 40% to the target individual to generate a new mutant individual, for example, the weight of the mechanism model is 75% and the weight of the prediction model is 25%.

[0057] Secondly, the server performs a crossover operation on the target individual and the mutated individual to generate a candidate individual. The crossover operation means combining some features of the target individual and the mutated individual to form a new individual. For example, some parameters of the mechanism model can be alternately copied from the target individual and the mutated individual. Suppose the target individual and the mutated individual have different values in certain parameters, such as reaction rate constant, temperature change, etc. Through the crossover operation, the temperature parameter of the target individual can be combined with the reaction rate constant of the mutated individual to produce a new candidate individual. Among them, each individual has a fitness value, which is used to measure how good the individual is in solving the problem. The fitness value is calculated through an objective function, such as minimizing error, maximizing prediction accuracy, etc. By comparing the fitness value of the candidate individual with that of the target individual, it is judged whether to replace the target individual with the candidate individual. Suppose the fitness value reflects the error of the model prediction, the fitness value of the candidate individual is 0.01, and the fitness value of the target individual is 0.03. Since the error of the candidate individual is smaller, the candidate individual will replace the target individual.

[0058] Next, if the fitness value of the candidate individual is greater than or equal to that of the target individual, then the candidate individual replaces the target individual and enters the next generation population. This indicates that this combined structure is better. If it is found after evaluation that the prediction result of the candidate individual is more accurate and has a higher fitness, then the candidate individual is introduced into the population to replace the original target individual. Repeat the above differential mutation, crossover operation and fitness comparison until the preset number of iterations is reached or the fitness value converges, that is, the prediction accuracy of the model no longer improves significantly. For example, through multiple rounds of iteration, each time the individuals in the population are updated, making the model combination more and more accurate until sufficient accuracy is achieved or the optimization stop condition is reached, such as the number of iterations reaches the upper limit or the error reaches an acceptable level. Finally, after multiple iterations of optimization, the best individual, that is, the combination with the optimal fitness value, will constitute the final fusion model. Suppose after multiple cycles, the best model combination is the combination of the mechanism model and the prediction model, where the weight of the mechanism model is 80% and the weight of the prediction model is 20%. After optimization, this combined model can predict the product concentration of the chemical reactor more accurately.

[0059] S150. Process the target optimization parameter through the fusion model to obtain an optimization result.

[0060] Specifically, the target optimization parameters refer to the specific parameters in a chemical production plant that need to be optimized. They may be the input parameters of the plant, such as feed flow rate, temperature, etc., or the output parameters, such as the concentration of the product, reaction efficiency, etc., or the overall performance indicators of the plant, such as energy efficiency, safety, economy, etc. For example, in a certain chemical production process, the target optimization parameters may be reaction temperature, reaction time, feed flow rate, product concentration, etc. The purpose of optimizing these parameters is to increase product output, optimize resource utilization, or reduce energy consumption. Processing the target optimization parameters means using the fusion model to simulate and calculate different operating conditions to find the optimal operating point. The calculation results of the fusion model will provide predicted values for each possible operating condition, and then the best parameter configuration will be selected according to the optimization objectives, such as minimizing energy consumption, maximizing output, improving product quality, etc. For example, in a chemical production process, the server can use the fusion model to simulate the impact of different feed flow rates, temperatures, etc. on product quality and calculate the optimization results under different parameter combinations. Through the processing of the target optimization parameters, the server will finally obtain a set of optimal solutions, that is, the optimization results. The optimization results can be the optimal values of certain operating variables or the best operating scheme under certain process conditions. For example, the optimization results may show that under specific reaction conditions, such as a temperature of 300 °C and a feed flow rate of 100 L / min, the concentration of the chemical reaction product is the highest, and the energy efficiency and reaction rate are the best.

[0061] In a possible implementation, the target optimization parameters are processed through the fusion model to obtain the optimization results, specifically including: determining the target input variables and target output variables from the target optimization parameters; inputting the target input variables into the fusion model to obtain feature variables, where the feature variables include feature vectors; judging the positional relationship between the feature vectors and the hyperplane of the support vector machine model corresponding to the fusion model; calculating the first probability value and the second probability value based on the positional relationship, where the first probability value is used to represent the probability value corresponding to the feature vectors on the first side of the hyperplane, and the second probability value is used to represent the probability value corresponding to the feature vectors on the second side of the hyperplane, and the first side and the second side are two opposite sides of the hyperplane; if it is determined that the feature vectors are located on the first side and the first probability value is greater than or equal to the preset threshold, then determine that the feature variables are the target output variables, and then determine that the optimization results are normal; if it is determined that the feature vectors are located on the second side and the second probability value is greater than or equal to the preset threshold, then determine that the feature variables are not the target output variables, and then determine that the optimization results are abnormal.

[0062] Specifically, the server inputs the target input variables into the fusion model. The fusion model combines a mechanism model and a prediction model, can process these variables, and outputs some characteristic variables. The characteristic variables are processed variables that contain the potential relationship between the input variables and the output results. For example, if the target input variables are reaction temperature and flow rate, the fusion model may output characteristic variables that represent the predicted results of the reaction under specific conditions. A feature vector refers to the multi-dimensional representation of the characteristic variables, which contains each feature calculated by the input variables through the fusion model. The support vector machine model constructs a hyperplane to distinguish different classes of data. The hyperplane divides the data space into two regions, each corresponding to a class. In this process, the output of the fusion model will be compared with the hyperplane of the support vector machine model. This hyperplane is used to determine whether the target input variables lead to the optimization objective reaching the expected result.

[0063] Among them, the server determines the optimization result by judging the position of the feature vector on both sides of the hyperplane of the support vector machine model. The first probability value represents the probability that the feature vector is on the first side of the support vector machine hyperplane, and the second probability value represents the probability that the feature vector is on the second side. These two probability values respectively represent whether the feature vector conforms to the expected optimization result. If the feature vector is on the first side of the hyperplane and the first probability value is greater than or equal to the preset threshold, the system considers the optimization result to be normal. At this time, the characteristic variables are considered to meet the expected target, and the system confirms that the optimization result is successful. If the feature vector is on the second side of the hyperplane and the second probability value is greater than or equal to the preset threshold, the optimization result is considered abnormal. This means that the input parameters fail to achieve the expected optimization effect, which may lead to poor output or an inefficient production process.

[0064] In a possible implementation manner, referring to Figure 2 , Figure 2 is another flowchart of a real-time optimization method for a chemical production device provided by an embodiment of the present application, including steps S210 to S220. The above steps are as follows: S210. Receive an optimization request uploaded by the user through the user device. The optimization request includes target optimization parameters; S220. Optimize the target optimization parameters according to the optimization request.

[0065] Specifically, the user device refers to computers, consoles, or other terminal devices in a chemical plant or laboratory through which users can send optimization requests. An optimization request is a requirement put forward by users to the server, indicating that they hope to optimize certain process parameters to improve production efficiency, save costs, improve product quality, or achieve other goals. In the optimization request, the target optimization parameter is the variable or metric that the user hopes to optimize. Depending on the different goals of chemical production, the target optimization parameters may include: reaction temperature, pressure, flow rate, raw material concentration, product output, energy efficiency, cost, etc. For example, users may hope to maximize the product output by adjusting the reaction temperature and flow rate, or hope to reduce energy consumption without affecting quality. After receiving the request, the server will perform optimization calculations based on the target optimization parameters provided by the user. This process may involve adjusting the chemical production device or updating the control strategy. The optimization methods may include based on mechanism models, prediction models, machine learning algorithms, or by means of simulation, experimental design, etc. to find the optimal solution. For example, if the user requests to increase the output while ensuring product quality, the server may analyze the mutual relationship between parameters such as reaction temperature, pressure, and flow rate, obtain the most suitable operating conditions, and then generate a control strategy.

[0066] This application also provides a real-time optimization system for a chemical production device. Referring to Figure 3 , Figure 3 is a schematic diagram of the modules of a real-time optimization system for a chemical production device provided by an embodiment of this application. The real-time optimization system is a server, and the server includes an acquisition module 31 and a processing module 32. Among them, the acquisition module 31 acquires mechanism data and real-time operation data during the production process of the chemical production device; the processing module 32 constructs a mechanism model based on the mechanism data, and the mechanism model is used to reflect the mass transfer, energy balance, and chemical reactions inside the chemical production device; the processing module 32 constructs a prediction model based on the real-time operation data, and the prediction model is used to predict the non-linear relationship between the input variables and output variables of the chemical production device; the processing module 32 fuses the mechanism model and the prediction model to obtain a fusion model; the processing module 32 processes the target optimization parameters through the fusion model to obtain an optimization result.

[0067] In a possible implementation manner, the acquisition module 31 acquires mechanism data and real-time operation data during the production process of the chemical production device, specifically including: the processing module 32 determines the physical principle and chemical principle of the chemical production device; the processing module 32 determines the process behavior of the chemical production device during the production process based on the physical principle and chemical principle to obtain mechanism data; the acquisition module 31 acquires the original data of the chemical production device during the production process in real time; the processing module 32 preprocesses the original data to obtain real-time operation data, and the preprocessing includes denoising, filtering, and normalization processing.

[0068] In a possible implementation manner, the processing module 32 constructs a mechanism model according to mechanism data, specifically including: the processing module 32 determines the basic reaction relationships among substances inside the chemical production device according to the mechanism data; the processing module 32 determines the feed parameters, output parameters, intermediate product parameters, and by-product parameters of the chemical production device based on the basic reaction relationships; the processing module 32 determines the complex reaction relationships among substances inside the chemical production device according to the mechanism data; the processing module 32 determines the heat transfer parameters, heat loss parameters, and mass transfer parameters of the chemical production device based on the complex reaction relationships; the processing module 32 generates a mechanism model under natural constraint conditions according to the feed parameters, output parameters, intermediate product parameters, and by-product parameters, as well as the heat transfer parameters, heat loss parameters, and mass transfer parameters.

[0069] In a possible implementation manner, the processing module 32 constructs a prediction model according to real-time operation data, specifically including: the acquisition module 31 acquires the historical operation data of the chemical production device; the processing module 32 determines the real-time operation data and the historical operation data as input variables, and processes them using a support vector machine algorithm to obtain output variables; the acquisition module 31 acquires the input variables and output variables corresponding to each of multiple training processes, and evaluates the training processes through an evaluation algorithm to obtain a prediction model.

[0070] In a possible implementation manner, the processing module 32 fuses the mechanism model and the prediction model to obtain a fusion model, specifically including: the processing module 32 constructs a population containing the mechanism model and the prediction model, the population includes multiple individuals, and each individual represents a combined structure of a model; the processing module 32 performs differential mutation on a target individual to obtain a mutant individual, and the target individual is any one of the multiple individuals; the processing module 32 performs a crossover operation on the target individual and the mutant individual to obtain a candidate individual; the processing module 32 compares the magnitude relationship between the fitness value corresponding to the candidate individual and the fitness value corresponding to the target individual; if the processing module 32 determines that the fitness value corresponding to the candidate individual is greater than or equal to the fitness value corresponding to the target individual, it replaces the target individual with the candidate individual and enters the next generation population; the processing module 32 repeats differential mutation, crossover operation, and comparison until a preset number of iterations is reached or the fitness value converges, so as to obtain a fusion model.

[0071] In a possible implementation, the processing module 32 processes the target optimization parameters through a fusion model to obtain an optimization result, specifically including: the processing module 32 determines a target input variable and a target output variable from the target optimization parameters; the processing module 32 inputs the target input variable into the fusion model to obtain a feature variable, and the feature variable includes a feature vector; the processing module 32 determines the positional relationship between the feature vector and the hyperplane of the support vector machine model corresponding to the fusion model; the processing module 32 calculates a first probability value and a second probability value based on the positional relationship, where the first probability value is used to represent the probability value corresponding to the feature vector on the first side of the hyperplane, and the second probability value is used to represent the probability value corresponding to the feature vector on the second side of the hyperplane, and the first side and the second side are two opposite sides of the hyperplane; if the processing module 32 determines that the feature vector is located on the first side and the first probability value is greater than or equal to a preset threshold, it determines that the feature variable is the target output variable, and then determines that the optimization result is normal; if the processing module 32 determines that the feature vector is located on the second side and the second probability value is greater than or equal to a preset threshold, it determines that the feature variable is not the target output variable, and then determines that the optimization result is abnormal.

[0072] In a possible implementation, the acquisition module 31 receives an optimization request uploaded by the user through the user device, and the optimization request includes target optimization parameters; the processing module 32 optimizes the target optimization parameters according to the optimization request.

[0073] It should be noted that when the system provided in the above embodiments implements its functions, only the division of the above functional modules is used for illustration. In practical applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be seen in the method embodiments, which will not be elaborated here.

[0074] This application also provides an electronic device, referring to Figure 4 , Figure 4 is a schematic structural diagram of an electronic device provided by an embodiment of this application. The electronic device may include: at least one processor 41, at least one network interface 44, a user interface 43, a memory 45, and at least one communication bus 42.

[0075] Among them, the communication bus 42 is used to realize the connection and communication between these components.

[0076] Among them, the user interface 43 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 43 may further include a standard wired interface and a wireless interface.

[0077] Among them, the network interface 44 may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface).

[0078] Among them, the processor 41 may include one or more processing cores. The processor 41 connects various parts within the entire server through various interfaces and lines. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 45, and by calling data stored in the memory 45, it performs various functions of the server and processes data. Optionally, the processor 41 may be implemented in at least one of the following hardware forms: digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). The processor 41 may integrate a combination of one or several of a central processing unit (CPU), a graphics processing unit (GPU), and a modem, etc. Among them, the CPU mainly processes the operating system, user interface, application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 41 and may be implemented separately by a single chip.

[0079] Among them, the memory 45 may include random access memory (RAM) and may also include read-only memory. Optionally, the memory 45 includes a non-transitory computer-readable storage medium. The memory 45 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 45 may include a program storage area and a data storage area. Among them, the program storage area may store instructions for implementing the operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store data involved in the above-mentioned method embodiments. Optionally, the memory 45 may also be at least one storage system located far from the aforementioned processor 41. As Figure 4 shown, the memory 45, as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for a real-time optimization method of a chemical production device.

[0080] In Figure 4In the electronic device shown, the user interface 43 is mainly used to provide an interface for the user to input and obtain the data input by the user; while the processor 41 can be used to call the application program stored in the memory 45 for a real-time optimization method of a chemical production device. When executed by one or more processors, the electronic device is caused to execute the method as described in one or more of the above embodiments.

[0081] It should be noted that, for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present application is not limited by the described action sequence, because according to the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the present application.

[0082] The present application also provides a computer-readable storage medium storing instructions. When executed by one or more processors, the electronic device is caused to execute the method as described in one or more of the above embodiments.

[0083] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0084] In several embodiments provided by the present application, it should be understood that the disclosed system can be implemented in other ways. For example, the system embodiments described above are only illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some service interfaces. The indirect couplings or communication connections of the systems or units can be in electrical or other forms.

[0085] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place, or can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0086] In addition, in each embodiment of the present application, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0087] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present application. The aforementioned memory includes: various media such as USB flash drives, mobile hard disks, magnetic disks, or optical discs that can store program codes.

[0088] The above are only exemplary embodiments of the present disclosure, and the scope of the present disclosure cannot be limited thereby. That is, any equivalent changes and modifications made in accordance with the teachings of the present disclosure still fall within the scope covered by the present disclosure. Those skilled in the art will readily think of other implementation manners of the present disclosure after considering the specification and the practice of the present disclosure. The present application aims to cover any variations, uses, or adaptive changes of the present disclosure, and these variations, uses, or adaptive changes follow the general principles of the present disclosure and include the common general knowledge or conventional technical means in the technical field not recorded in the present disclosure. The specification and the embodiments are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A real-time optimization method for a chemical production device, characterized in that: The method comprises: Obtain mechanism data and real-time operation data of chemical production equipment during the production process; A mechanism model is constructed based on the mechanism data, and the mechanism model is used to reflect the material transfer, energy balance and chemical reaction inside the chemical production device; A prediction model is constructed based on the real-time operation data, wherein the prediction model is used to predict the nonlinear relationship between the input variables and the output variables of the chemical production device; Fusing the mechanism model and the prediction model to obtain a fusion model; The target optimization parameters are processed through the fusion model to obtain the optimization results.

2. The real-time optimization method for a chemical production device according to claim 1, characterized in that: The acquisition of mechanism data and real-time operation data of the chemical production device during the production process specifically includes: Determine the physical and chemical principles of the chemical production plant; Based on the physical principle and the chemical principle, determining the process behavior of the chemical production device during the production process to obtain the mechanism data; Real-time acquisition of raw data of the chemical production device during the production process; The raw data is preprocessed to obtain the real-time operation data, wherein the preprocessing includes denoising, filtering and normalization.

3. The real-time optimization method for a chemical production device according to claim 1, characterized in that: The mechanism model is constructed based on the mechanism data, and specifically includes: Determining the basic reaction relationship between substances inside the chemical production device based on the mechanism data; Based on the basic reaction relationship, determining the feed parameters, output parameters, intermediate product parameters and by-product parameters of the chemical production device; Determining the complex reaction relationships between substances within the chemical production device based on the mechanism data; Based on the complex reaction relationship, determining heat transfer parameters, heat loss parameters and mass transfer parameters of the chemical production device; Under natural constraints, the mechanism model is generated according to the feed parameters, output parameters, intermediate product parameters and by-product parameters, as well as the heat transfer parameters, heat loss parameters and mass transfer parameters.

4. The real-time optimization method for a chemical production device according to claim 1, characterized in that: The constructing of a prediction model according to the real-time operation data specifically includes: Obtaining historical operation data of the chemical production device; Determine the real-time operation data and the historical operation data as the input variables, and process them using a support vector machine algorithm to obtain the output variable; The input variables and output variables corresponding to the multiple training processes are obtained, and the training processes are evaluated by an evaluation algorithm to obtain the prediction model.

5. The real-time optimization method for a chemical production device according to claim 1, characterized in that: The step of fusing the mechanism model and the prediction model to obtain a fusion model specifically includes: Constructing a population including the mechanism model and the prediction model, wherein the population includes a plurality of individuals, each of which represents a combination structure of a model; Performing differential mutation on a target individual to obtain a mutant individual, wherein the target individual is any one of the multiple individuals; Performing a crossover operation on the target individual and the variant individual to obtain a candidate individual; Comparing the fitness value corresponding to the candidate individual and the fitness value corresponding to the target individual; If it is determined that the fitness value corresponding to the candidate individual is greater than or equal to the fitness value corresponding to the target individual, the candidate individual replaces the target individual and enters the next generation population; The differential mutation, the crossover operation and the comparison are repeated until a preset number of iterations is reached or the fitness value converges to obtain the fusion model.

6. The real-time optimization method for a chemical production device according to claim 1, characterized in that: The target optimization parameters are processed by the fusion model to obtain the optimization results, which specifically include: Determining a target input variable and a target output variable from the target optimization parameters; Inputting the target input variable into the fusion model to obtain a feature variable, wherein the feature variable includes a feature vector; Determining the positional relationship between the feature vector and the hyperplane of the support vector machine model corresponding to the fusion model; Based on the positional relationship, a first probability value and a second probability value are calculated, wherein the first probability value is used to represent the probability value of the feature vector corresponding to the first side of the hyperplane, and the second probability value is used to represent the probability value of the feature vector corresponding to the second side of the hyperplane, wherein the first side and the second side are two opposite sides of the hyperplane; If it is determined that the feature vector is located on the first side and the first probability value is greater than or equal to a preset threshold, then it is determined that the feature variable is the target output variable, and then it is determined that the optimization result is normal; If it is determined that the feature vector is located on the second side and the second probability value is greater than or equal to the preset threshold, it is determined that the feature variable is not the target output variable, and the optimization result is determined to be abnormal.

7. The real-time optimization method for a chemical production device according to claim 1, characterized in that: The method further comprises: Receiving an optimization request uploaded by a user through a user device, wherein the optimization request includes the target optimization parameter; According to the optimization request, the target optimization parameters are optimized.

8. A real-time optimization system for a chemical production device, characterized in that: The real-time optimization system comprises an acquisition module (31) and a processing module (32), wherein: The acquisition module (31) is used to acquire the mechanism data and real-time operation data of the chemical production device during the production process; The processing module (32) is used to construct a mechanism model based on the mechanism data, and the mechanism model is used to reflect the material transfer, energy balance and chemical reaction inside the chemical production device; The processing module (32) is further used to construct a prediction model based on the real-time operation data, wherein the prediction model is used to predict the nonlinear relationship between the input variables and the output variables of the chemical production device; The processing module (32) is also used to fuse the mechanism model and the prediction model to obtain a fusion model; The processing module (32) is also used to process the target optimization parameters through the fusion model to obtain the optimization results.

9. An electronic device, characterized in that: The electronic device comprises a processor (41), a memory (45), a user interface (43) and a network interface (44), wherein the memory (45) is used to store instructions, the user interface (43) and the network interface (44) are both used to communicate with other devices, and the processor (41) is used to execute the instructions stored in the memory (45) so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is performed.

Images