Artificial intelligence-based steam-water quality optimization control method and system
By constructing an AI-based soft drink quality optimization and control system, training a dynamic prediction model using multi-source datasets, and combining chemical reagent mapping rules, the system solved the control problem of CCGT soft drink quality in complex marine environments, achieving accurate prediction and dynamic optimization, and improving system operating efficiency and safety.
Patent Information
- Application Number
- CN202511736386.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-24
- Publication Date
- 2026-01-27
AI Technical Summary
In the existing technology, CCGT steam and water quality control is difficult to cope with the multi-factor interference under complex marine conditions, resulting in equipment damage, reduced heat exchange efficiency, increased energy consumption and carbon dioxide emissions, and failing to achieve accurate prediction and dynamic optimization control of steam and water quality.
By constructing an AI-based soft drink quality optimization and control system, a dynamic prediction model is trained using multi-source datasets. Combined with chemical reagent mapping rules, the system predicts the soft drink status in real time and generates reagent plans. The reagent dosage is then iteratively optimized to regulate the soft drink status.
It enables accurate prediction and dynamic optimization of steam and water quality in complex marine environments, effectively preventing corrosion and scaling, improving system operating efficiency and safety, and reducing energy consumption and carbon emissions.
Smart Images
Figure CN121414224A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of artificial intelligence technology, and in particular to a method, system, computer device, computer-readable storage medium, and computer program product for optimizing and controlling the quality of soft drinks based on artificial intelligence. Background Technology
[0002] Driven by global energy demand growth and the goals of "carbon peaking and carbon neutrality," offshore floating platform combined cycle power systems (CCGTs) are gaining increasing popularity due to their advantages in combined operation of gas turbines and steam turbines and waste heat recovery. In the complex marine environment, the quality of steam and water directly impacts system safety, operational efficiency, and equipment lifespan.
[0003] However, the CCGT steam and water quality control technology relies on traditional methods, which are difficult to cope with the multi-factor interference under complex marine conditions. It cannot achieve accurate prediction and dynamic optimization of steam and water quality, and is prone to problems such as corrosion, scaling and deposition, and chemical conflicts, which in turn lead to equipment damage, reduced heat exchange efficiency, increased energy consumption and carbon dioxide emissions, thus restricting the efficient, low-carbon and safe operation of the system. Summary of the Invention
[0004] Therefore, it is necessary to provide an artificial intelligence-based method, system, computer equipment, computer-readable storage medium, and computer program product that can accurately predict and dynamically optimize the quality of soft drinks, thereby addressing the aforementioned technical problems.
[0005] In a first aspect, this application provides a method for optimizing and controlling the quality of soft drinks based on artificial intelligence, including:
[0006] Obtain multi-source datasets for the combined cycle power system; the multi-source datasets include the operating parameters of the combined cycle power system, steam and water state indices, and state parameters; the state parameters include: steam and water temperature, pressure, flow rate, liquid level, and marine environmental parameters;
[0007] The steps involve using a pre-trained dynamic prediction model with multi-source datasets as input to predict the state indicators of soft drinks and obtain the predicted state results of the soft drinks.
[0008] The mapping rules between chemical reagents and the state of carbonated drinks are input into a dynamic prediction model to generate corresponding reagent schemes.
[0009] When the predicted result exceeds the set threshold, adjust the boiler blowdown volume or implement the chemical dosing according to the plan to improve the steam-water state index.
[0010] If the deviation of the regulated soda state index exceeds the preset deviation range, the reagent scheme will be iteratively optimized based on the current prediction results, soda state index, and mapping rules.
[0011] In one embodiment, the dynamic prediction model is trained using a historical multi-source dataset of a combined cycle power system, including:
[0012] Based on the mechanism modeling and simulation engine, a thermodynamic-hydrochemical-heat and mass transfer model is constructed;
[0013] Historical multi-source datasets are input into the thermodynamic-water chemistry-heat and mass transfer model to calculate operating parameters, steam and water state indices, and the correlation between state parameters.
[0014] Based on the correlation and historical multi-source datasets, an initial dynamic prediction model is established.
[0015] In one embodiment, the steam-water mixture includes: condensate, feedwater, boiler drum, high-pressure steam, and low-pressure steam;
[0016] The properties of carbonated beverages include pH value, conductivity, silica content, dissolved oxygen concentration, phosphate concentration, and sodium ion concentration.
[0017] Mapping rules, at least including:
[0018] The relationship between ammonia dosage and condensate pH value;
[0019] The matching relationship between phosphate dosage and phosphate concentration in boiler water.
[0020] In one embodiment, if the deviation of the regulated soda state index exceeds a preset deviation range, the reagent scheme is iteratively optimized based on the current prediction results and the soda state index, including:
[0021] Obtain the actual pesticide application parameters; these parameters include: pesticide type, dosage, application rate, and application duration.
[0022] The drug delivery parameters are updated based on the current multi-source dataset, prediction results, and mapping rules.
[0023] Secondly, this application also provides an artificial intelligence-based soft drink quality optimization control system, comprising a data acquisition layer, an AI analysis and optimization platform, a control layer, and an execution layer; wherein,
[0024] The data acquisition layer is used to acquire multi-source datasets of the combined cycle power system. The multi-source datasets include the operating parameters of the combined cycle power system, steam and water state indicators, and state parameters. The state parameters include: steam and water temperature, pressure, flow rate, liquid level, and marine environmental parameters.
[0025] The AI analysis and optimization platform is used to predict the changing trends of soft drink status based on multi-source datasets and generate corresponding adjustment instructions.
[0026] The control layer is used to parse and convert adjustment instructions to generate control signals that can drive the actions of the execution layer;
[0027] The execution layer is used to receive control signals transmitted from the control layer and perform operations such as adding chemicals or adjusting boiler blowdown to regulate the quality of steam and water.
[0028] In one embodiment, the AI analysis and optimization platform further includes:
[0029] The parameter acquisition module is used to receive multi-source datasets;
[0030] The data prediction module is used to predict the state indicators of soft drinks using a pre-trained dynamic prediction model and multiple source datasets as input, and obtain the prediction results of the soft drink state.
[0031] The reagent optimization module is used to input the mapping rules between chemical reagents and the state of the soda into the dynamic prediction model to generate the corresponding reagent scheme;
[0032] The dynamic optimization module is used to generate adjustment instructions for adjusting boiler blowdown or for administering chemical dosing based on the scheme when the prediction result exceeds the set threshold.
[0033] The feedback learning module is used to iteratively optimize the reagent scheme based on the current prediction results, the state index of the carbonated beverage, and the mapping rules if the deviation of the regulated carbonated beverage state index exceeds the preset deviation range.
[0034] In one embodiment, the control layer includes:
[0035] The early warning module is used to automatically trigger an alarm and switch to backup control logic when an abnormality is detected in a sensor, analyzer, or actuator.
[0036] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the soft drink quality optimization control method described in the above embodiments;
[0037] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the soft drink quality optimization control method described in the above embodiments.
[0038] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the soft drink quality optimization control method described in the above embodiments.
[0039] The aforementioned AI-based methods, systems, computer equipment, computer-readable storage media, and computer program products address the issue of manual intervention in the regulation of soft drink quality in related technologies. By introducing an AI analysis and optimization platform and building a data prediction module within it, and using a pre-trained dynamic prediction model, the platform takes the operating parameters of the combined cycle power system, soft drink status indicators, and marine environmental parameters as inputs to achieve intelligent prediction and adaptive regulation of soft drink quality (i.e., status indicators such as pH, dissolved oxygen, conductivity, and silicon content).
[0040] Based on this, a dynamic optimization module is introduced, which has built-in mapping rules between soda and added chemicals, generates prediction results and adjustment schemes for soda quality fluctuation trends, and iteratively optimizes the scheme when the quality deviation exceeds the limit, thereby effectively solving the technical problem of soda quality control in complex marine environments. Attached Figure Description
[0041] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0042] Figure 1 Here is a structural block diagram of an AI-based soft drink quality optimization control system in one embodiment. Figure 1 ;
[0043] Figure 2 Here is a structural block diagram of an AI-based soft drink quality optimization control system in one embodiment. Figure 2 ;
[0044] Figure 3 This is a flowchart illustrating an artificial intelligence-based soft drink quality optimization and control method in one embodiment.
[0045] Figure 4 for Figure 3 A flowchart illustrating step S2;
[0046] Figure 5 for Figure 3 A flowchart illustrating step S5 in the middle section;
[0047] Figure 6 This is an internal structural diagram of a computer device in one embodiment.
[0048] Explanation of reference numerals in the attached figures:
[0049] 1. Data Acquisition Layer; 11. Various Sensors; 12. Soft Drink Analysis System; 13. Integrated Control and Safety System; 2. AI Analysis and Optimization Platform; 21. Parameter Acquisition Module; 22. Data Prediction Module; 23. Reagent Optimization Module; 24. Dynamic Optimization Module; 25. Feedback Learning Module; 3. Control Layer; 4. Execution Layer. Detailed Implementation
[0050] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0051] like Figure 1 As shown, an artificial intelligence-based soft drink quality optimization control system is provided, comprising: a data acquisition layer 1, an AI analysis and optimization platform 2, a control layer 3, and an execution layer 4; wherein,
[0052] Data acquisition layer 1 is used to acquire multi-source datasets of the combined cycle power system. The multi-source datasets include the operating parameters of the combined cycle power system, steam and water state indicators, and state parameters. The state parameters include the temperature, pressure, flow rate, and liquid level of the steam and water, as well as marine environmental parameters.
[0053] AI analysis and optimization platform 2 is used to predict the changing trend of the state of soft drinks based on multi-source datasets and generate corresponding adjustment instructions;
[0054] Control layer 3 is used to parse and convert adjustment instructions to generate control signals that can drive the actions of the execution layer;
[0055] Execution layer 4 is used to receive control signals transmitted from the control layer and perform operations such as adding chemicals and adjusting boiler blowdown to regulate the quality of steam and water.
[0056] Thus, in the aforementioned steam and water quality optimization control system, the AI analysis and optimization platform can predict the steam and water quality in real time under complex operating conditions, and combine the interaction between agents, the correlation between agent concentration and steam and water state indicators and thresholds to perform operations such as agent addition and boiler blowdown adjustment to regulate the steam and water quality.
[0057] It should be understood that, as those skilled in the art can understand, Figure 1The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the soda quality optimization control system to which the present application is applied. The specific computer equipment may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.
[0058] The following is combined with Figure 2 The hierarchical architecture of the aforementioned soft drink quality optimization control system will be further discussed.
[0059] like Figure 2 As shown, in one embodiment, the data acquisition layer 1 includes various sensors 11, a steam and water analysis system 12 (SWAS), and an integrated control and safety system 13 (ICSS).
[0060] Specifically, in accordance with the data sampling requirements of SWAS and ICSS, monitoring devices such as pH analyzers, dissolved oxygen analyzers, conductivity probes, and silicon content probes are deployed at key locations including condensate, feedwater, boiler drum, high-pressure steam (HPSteam), and low-pressure steam (LPSteam) to acquire real-time steam and water condition indicators such as pH value, conductivity, silica content, dissolved oxygen concentration, phosphate concentration, and sodium ion concentration. Simultaneously, various sensors are installed on offshore floating platforms to collect steam and water temperature, pressure, flow rate, liquid level, as well as marine-specific environmental parameters such as salt spray concentration, ambient temperature, and humidity, which are used as condition parameters. In addition, ICSS also interacts with the combined cycle gas turbine (CCGT) system via an interface to collect operating parameters such as unit load, boiler operating parameters, and chemical dosing data.
[0061] Each monitoring point (such as PH_Cond and Cond_CationCond) is equipped with high / low (Hi / Lo) and high-high / low-low (HiHi / LoLo) alarm limits, which are the thresholds mentioned later in the text.
[0062] In the above implementation scheme, SWAS extracts samples from the key points using various analyzers and performs testing and processing, continuously monitoring key chemical indicators in the water-steam cycle to ensure compliance with technical specifications. This effectively prevents corrosion and scaling risks inside the waste heat boiler (HRSG), condenser, and steam turbine, providing continuous, accurate, and reliable online measurement of water quality and steam chemical composition. ICSS is responsible for real-time storage of status parameters from various sensors, steam and water status indicators from analyzers, and operating parameters obtained through interfaces, forming a multi-source dataset to provide data support for the continuous iteration and optimization of the artificial intelligence module.
[0063] Please continue reading. Figure 2In one embodiment, the AI analysis and optimization platform 2 further includes: a parameter acquisition module 21, a data prediction module 22, a drug optimization module 23, a dynamic optimization module 24, and a feedback learning module 25.
[0064] Specifically, the parameter acquisition module 21 receives the multi-source dataset transmitted by the data acquisition layer 1. In offline mode, it uses the historical operating data of the CCGT power system of the offshore floating platform as the training dataset for the dynamic prediction model. In real-time mode, it performs dynamic control based on the real-time acquisition of multi-source data.
[0065] In real-time mode, the data prediction module 22 uses a pre-trained dynamic prediction model, taking multi-source datasets as input, to predict the steam-water state indicators for future periods. Upon receiving the prediction structure, the reagent optimization module 23 invokes the built-in mapping rules between chemical reagents and steam-water state, combining the prediction results to generate an initial reagent plan. When the steam-water state indicators reach a threshold, the dynamic optimization module 24 generates an adjustment instruction for the corresponding reagent plan or boiler blowdown.
[0066] After the control is implemented, the data acquisition layer 1 provides real-time feedback on the actual soda status indicators. If the indicator deviation exceeds the preset range, the feedback learning module 25 is triggered. Combining the dosage of the agent fed back by the execution layer 4 with the current soda status indicators, the initial agent plan is iteratively optimized.
[0067] Please continue reading. Figure 2 In one embodiment, control layer 3 includes a distributed control system (DCS) or a supervisory control and data acquisition and monitoring system (SCADA), as well as a programmable logic controller (PLC) and control logic. The DCS / SCADA receives adjustment instructions from the AI analysis and optimization platform 2 and directly drives the built-in devices of execution layer 4 through the PLC logic. Simultaneously, control layer 3 integrates an early warning module (not shown), which can automatically trigger an alarm and switch to backup control logic upon detecting abnormalities in sensors, analyzers, or actuators, such as sensor signal interruption or dosing pump failure, providing crucial safety redundancy for the system.
[0068] The execution layer 4 includes, but is not limited to, the dosing pumps of the Chemical Dosing System (CDS), the electric or pneumatic regulating valves of the boiler blowdown system, and the coolers and flow control valves of the steam and water sampling system. The control signals of the control layer 3 directly act on these devices. Through the control layer 3 and the execution layer 4, the adjustment schemes issued by AI are reliably and safely translated into specific equipment actions, realizing real-time prediction and adaptive adjustment of steam and water quality based on AI, which can reduce the fluctuation range of key indicators to within ±1%.
[0069] Please see Figure 3 In some embodiments, this application also provides an artificial intelligence-based method for optimizing and controlling the quality of soft drinks. This embodiment illustrates the application of this method to a terminal. It is understood that this method can also be applied to a server, and further to a system including both a terminal and a server, and is implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:
[0070] Step S1: Obtain the multi-source dataset of the combined cycle power system; the multi-source dataset includes the operating parameters, steam and water state indices, and state parameters of the combined cycle power system.
[0071] Step S2: Using a pre-trained dynamic prediction model and a multi-source dataset as input, predict the state indicators of the soda and obtain the prediction results of the soda state.
[0072] Step S3: Input the mapping rules between chemical reagents and the state of the soda into the dynamic prediction model to generate the corresponding reagent scheme.
[0073] Step S4: When the prediction result exceeds the set threshold, adjust the boiler blowdown volume or perform chemical dosing according to the plan to improve the steam-water state indicators.
[0074] Step S5: If the deviation of the adjusted soda state index exceeds the preset deviation range, the reagent scheme is iteratively optimized based on the current prediction results, soda state index, and mapping rules.
[0075] Please see Figure 4 Step S2 in this embodiment includes:
[0076] Step S202: Based on the mechanism modeling and simulation engine, construct a thermodynamic-hydrochemical-heat and mass transfer model.
[0077] Using a mechanistic modeling simulation engine (e.g., Aspen Plus, Unisim, Petrosim, VxOTS, MATLAB / Simulink, or a self-developed simulation platform), a thermodynamic-hydraulic-heat-mass transfer model is constructed that couples thermodynamics, water chemistry, and heat and mass transfer principles. This model can accurately simulate the transfer and transformation processes of mass, energy, and chemical components in a steam-water circulation system, such as the quantitative correlation between pH value, phosphate concentration, metal corrosion rate, and scale formation, or calculate the effects of steam temperature and pressure on dissolved oxygen solubility.
[0078] Step S204: Input the historical multi-source dataset into the thermodynamic-water chemistry-heat and mass transfer model to calculate the correlation between operating parameters, steam and water state indicators, and marine environmental parameters.
[0079] The historical multi-source dataset of the combined cycle power system collected in step S1 (i.e., operating parameters, steam-water state indices, and state parameters covering various typical operating conditions and extreme conditions) is input into the above mechanism model, and a large number of simulation calculations are performed to obtain the correlation between the three.
[0080] Step S206: Based on the correlation and historical multi-source datasets, establish the initialized dynamic prediction model.
[0081] Using a large historical dataset as training samples, temporal deep learning algorithms (such as LSTM long short-term memory networks and Transformers) or hybrid mechanism-data-driven modeling methods are employed to train and calibrate the model parameters. The trained model becomes the initialized dynamic prediction model, enabling the artificial intelligence module to output predictions of the floating trends in soda quality based on real-time input parameters. The prediction results must include the future variation range of key indicators such as pH value, conductivity, and dissolved oxygen concentration.
[0082] In this embodiment, the extended step of step S3 includes inputting mapping rules, verified through experiments and operational experience, that clearly define the quantitative or qualitative relationship between chemical agents and the state parameters of the carbonated beverage, into the dynamic prediction model. After obtaining the prediction result of step S2, the system compares this predicted value with the target setpoint and calls these mapping rules for reverse calculation, thereby generating a preliminary, quantitative reagent plan. This plan specifically specifies the type of reagent (such as ammonia or phosphate), the recommended dosage, and the dosing rate.
[0083] For example, the mapping rules include, but are not limited to, the correspondence between ammonia dosage and condensate pH value, and the matching relationship between phosphate dosage and phosphate concentration in boiler water.
[0084] The extended steps of step S4 include: comparing the prediction results with the preset threshold, and when the threshold is reached, starting the control program to increase the boiler blowdown to quickly reduce the impurity concentration, or according to the reagent scheme generated in step S3, controlling the flow rate and dosing rate of the corresponding dosing pump in the chemical dosing system (CDS) through the control layer, so as to intervene before the problem actually occurs and actively improve the steam and water status indicators.
[0085] Please see Figure 5 Step S5 in this embodiment includes:
[0086] Step S502: If the deviation of the adjusted soda state index exceeds the preset deviation range, obtain the actual reagent dosing parameters; the reagent dosing parameters include: reagent type, dosage, dosing rate and dosing time.
[0087] After a control action is executed, the system continuously monitors the actual values of the regulated carbonated water state indicators and compares them with the expected target values to calculate the deviation. If the deviation persists and exceeds the preset deviation range (for example, the actual pH value does not reach the target range after the addition of chemicals), step S504 is executed.
[0088] Step S504: Update the drug delivery plan based on the drug delivery parameters, combined with the current multi-source dataset, prediction results, and mapping rules.
[0089] By combining current forecast results, real-time monitoring data, actual reagent dosing parameters, and built-in mapping rules, the system dynamically compensates for and updates existing reagent plans in real time. This generates more adaptive new reagent plans, enabling the system to self-evolve and continuously improve, forming a newer and more accurate plan to ensure the timely adoption of updated reagent plans or adjustment strategies. For example, when the system identifies the impact of high salt spray concentrations on water quality, it automatically introduces a dynamic compensation coefficient into the ammonia dosing plan to increase the ammonia dosage, thereby ensuring that the pH value can be accurately controlled within the target range.
[0090] In practical applications, taking the monitoring of boiler water pH value through a steam-water analysis system (SWAS) as an example, when the pH value shows a slow downward trend and the cationic conductivity begins to increase, the dynamic prediction model, based on real-time data and historical operating conditions, will issue a warning that the boiler water quality will exceed the first-level control limit (pH < 9.0, cationic conductivity > 3) within one hour, and then implement corrective measures to increase the boiler blowdown volume.
[0091] When boiler water quality is expected to exceed the secondary control limits (pH < 8.5, cationic conductivity > 4.5), there is a risk of scaling and corrosion. The AI analysis and optimization platform generates and sends an adjustment command to start the phosphate dosing pump based on the chemically generated reagent scheme in offline mode. Simultaneously, upon receiving the command, the control layer controls the execution layer to drive the dosing pump, injecting a specific concentration of phosphate solution into the boiler drum. After the adjustment is executed, the system continuously tracks changes in actual pH and conductivity. If the water quality recovery rate is slower than expected due to the high salt spray environment at sea, the feedback learning module will be triggered. It will automatically fine-tune the phosphate dosing acceleration rate and intelligently connect to the boiler blowdown system, appropriately increasing the blowdown volume to accelerate the water quality recovery to normal. The entire process achieves closed-loop autonomous control from intelligent early warning and precise dosing to dynamic optimization, effectively preventing corrosion and scaling inside the boiler and ensuring the safe and efficient operation of the unit under harsh sea conditions.
[0092] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.
[0093] In some embodiments, in an exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 6As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements an artificial intelligence-based method for optimizing and controlling the quality of soft drinks. The display unit is used to create a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.
[0094] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.
[0095] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.
[0096] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.
Claims
1. A method for optimizing and controlling the quality of soft drinks based on artificial intelligence, characterized in that, include: Obtain a multi-source dataset of the combined cycle power system; the multi-source dataset includes the operating parameters, steam and water state indices, and state parameters of the combined cycle power system. The state parameters include the temperature, pressure, flow rate, and liquid level of the steamed water, as well as marine environmental parameters; Using a pre-trained dynamic prediction model and the multi-source dataset as input, the soda state index is predicted to obtain the predicted soda state result. The mapping rules between chemical agents and the state of the soda are input into the dynamic prediction model to generate the corresponding agent scheme. When the predicted result exceeds the set threshold, adjust the boiler blowdown volume or implement the chemical dosing according to the plan to improve the steam-water state index. If the deviation of the adjusted soda state index exceeds the preset deviation range, the reagent scheme is iteratively optimized based on the current prediction results, the soda state index, and the mapping rule.
2. The method for optimizing and controlling the quality of soft drinks according to claim 1, characterized in that, The dynamic prediction model is trained using historical multi-source datasets of combined cycle power systems, including: Based on the mechanism modeling and simulation engine, a thermodynamic-hydrochemical-heat and mass transfer model is constructed; The historical multi-source dataset is input into the thermodynamic-water chemistry-heat and mass transfer model to calculate the operating parameters, the steam-water state index, and the correlation between the state parameters. Based on the aforementioned correlations and the historical multi-source dataset, an initialized dynamic prediction model is established.
3. The method for optimizing and controlling the quality of soft drinks according to claim 1, characterized in that, The steam and water system includes: condensate, feedwater, boiler drum, high-pressure steam, and low-pressure steam; The soda's state indicators include pH value, conductivity, silica content, dissolved oxygen concentration, phosphate concentration, and sodium ion concentration. The mapping rules include at least the following: The relationship between ammonia dosage and condensate pH value; The matching relationship between phosphate dosage and phosphate concentration in boiler water.
4. The method for optimizing and controlling the quality of soft drinks according to claim 1, characterized in that, If the deviation of the adjusted soda state index exceeds a preset deviation range, the reagent scheme is iteratively optimized based on the current prediction results and the soda state index, including: Obtain the actual drug delivery parameters; the drug delivery parameters include: drug type, dosage, delivery rate and delivery time; The drug delivery parameters are used to update the drug delivery plan, taking into account the current multi-source dataset, prediction results, and the mapping rules.
5. A soft drink quality optimization control system based on artificial intelligence, characterized in that, include: The system comprises a data acquisition layer, an AI analysis and optimization platform, a control layer, and an execution layer; among which, The data acquisition layer is used to acquire multi-source datasets of the combined cycle power system; the multi-source datasets include operating parameters, steam and water state indices, and state parameters of the combined cycle power system. The AI analysis and optimization platform is used to predict the changing trend of the soda state based on the multi-source dataset and generate corresponding adjustment instructions. The control layer is used to parse and convert adjustment instructions to generate control signals that can drive the action of the execution layer. The execution layer is used to receive control signals transmitted from the control layer and perform operations such as adding chemicals or adjusting boiler blowdown to regulate the quality of steam and water.
6. The soft drink quality optimization control system according to claim 5, characterized in that, The AI analysis and optimization platform also includes: The parameter acquisition module is used to receive the multi-source dataset; The data prediction module is used to predict the soda state index using the multi-source dataset as input through a pre-trained dynamic prediction model, and obtain the prediction result of the soda state. The reagent optimization module is used to input the mapping rules between chemical reagents and the state of the soda into the dynamic prediction model to generate the corresponding reagent scheme. The dynamic optimization module is used to generate adjustment instructions for adjusting boiler blowdown or for administering chemical dosing based on the scheme when the prediction result exceeds the set threshold. The feedback learning module is used to iteratively optimize the reagent scheme based on the current prediction results, the water state index, and the mapping rules if the deviation of the regulated soda state index exceeds a preset deviation range.
7. The soft drink quality optimization control system according to claim 5, characterized in that, The control layer includes: The early warning module is used to automatically trigger an alarm and switch to backup control logic when an abnormality is detected in a sensor, analyzer, or actuator.
8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-4.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-4.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-4.