Closed-loop controller parameter optimization method and system applied to uranium concentration auxiliary control process

Abstract

The invention provides a closed-loop controller parameter optimization method and system applied to a uranium concentration auxiliary control process. The method comprises the steps that according to different scenes, historical system control parameters corresponding to the different scenes are captured from a uranium concentration auxiliary control process control system running in real time at regular intervals; according to the latest historical system control parameters corresponding to different scenes, different system large models are constructed regularly; respectively simulating the current system large models of different scenes, and correspondingly obtaining simplified system models respectively corresponding to the different scenes after simulation; and comparing the precision of the simplified system model and the precision of the original system simplified model of the closed-loop controller in the same scene, and determining whether to update the current parameters of the closed-loop controller according to a comparison result. According to the method, the defect of manually updating the model parameters is avoided, and the model parameters are automatically updated.

Description

Technical Field

[0001] The present invention relates to the technical field of uranium enrichment, and in particular to a closed-loop controller parameter optimization method and system applied to a uranium enrichment auxiliary control process. Background Art

[0002] The uranium enrichment auxiliary control system is a critical auxiliary system in the uranium enrichment production process. It consists of hardware equipment such as cooling towers, air conditioning units, fans, pumps, sensors, actuators, and related control logic. Its core function is to provide process chilled water at a constant temperature and flow rate for uranium enrichment production and maintain a constant temperature and humidity environment in the equipment production workshop. This directly ensures stable heat exchange within the main process production equipment, thereby ensuring the smooth operation of the uranium enrichment process. In actual production, the uranium enrichment auxiliary control system uses closed-loop controllers to achieve closed-loop control of certain workshops. However, due to external factors such as seasonal changes and weather changes, the operating conditions of the uranium enrichment auxiliary control system are complex and variable. Therefore, appropriate adjustment of the closed-loop controller parameters is required to adapt to disturbances caused by external factors and more accurately control the constant temperature and humidity environment in the equipment production workshop.

[0003] However, the parameter tuning of existing closed-loop controllers relies on manual work and cannot be automatically updated according to real-time operating conditions. This manual adjustment method has the problem of response lag, which affects the stability and energy-saving efficiency of the uranium enrichment auxiliary control system.

[0004] Therefore, developing a method that can automatically adjust the closed-loop controller parameters of the uranium enrichment auxiliary control system has become the key to improving system stability, control accuracy and energy saving. Summary of the Invention

[0005] The purpose of the present invention is to solve the defects of the above-mentioned prior art and provide a closed-loop controller parameter optimization method and system for uranium enrichment auxiliary control process.

[0006] The present invention provides a closed-loop controller parameter optimization method applied to a uranium enrichment auxiliary control process, comprising: Step 1: Based on different scenarios, periodically capture the historical system control parameters corresponding to different scenarios from the real-time uranium enrichment auxiliary control system; Step 2: Regularly build different system models based on the latest historical system control parameters corresponding to different scenarios; Step 3: Simulate the current system large model for different scenarios respectively, and obtain the corresponding simplified system models for different scenarios after simulation; Step 4: Compare the accuracy of the simplified system model and the original simplified system model of the closed-loop controller under the same scenario, and determine whether to update the current parameters of the closed-loop controller based on the comparison result.

[0007] According to the closed-loop controller parameter optimization method for a uranium enrichment auxiliary control process provided by the present invention, step three includes: Step 31: For different scenarios, first system input data within a preset time period is customized respectively, and the first system input data corresponding to each scenario is used as input to the current system macro model, thereby obtaining first system output data of the current system macro model corresponding to each scenario within a future preset time period; Step 32: Using the first system input data and the first system output data of the same scenario, a simplified system model of the corresponding scenario is constructed, thereby obtaining simplified system models corresponding to different scenarios after simulation.

[0008] According to the closed-loop controller parameter optimization method for a uranium enrichment auxiliary control process provided by the present invention, step four includes: Step 41: Using the first system input data under the same scenario as the input of the original system simplified model of the closed-loop controller, and correspondingly obtaining the second system output data of the original system simplified model under the corresponding scenario; using the first system input data corresponding to the same scenario as the input of the current system simplified model, and correspondingly obtaining the third system output data of the current system simplified model under the same scenario; Step 42: Determine the difference between the second system output data and the first system output data to obtain a first difference value; determine the difference between the third system output data and the first system output data to obtain a second difference value; Step 43: comparing the first difference value and the second difference value to see whether the difference is within a preset range; if it is outside the preset range, updating the original simplified system model of the controller to the current simplified system model; Step 44: Update the current parameters of the closed-loop controller according to the current system simplified model.

[0009] According to the closed-loop controller parameter optimization method for uranium enrichment auxiliary control process provided by the present invention, the historical system control parameters include: historical water system control parameters, historical air conditioning system control parameters; The historical water system control parameters include: cooling water pump frequency, cooling tower return water temperature, cooling tower inlet water temperature set according to time series; The historical air conditioning system control parameters include: valve opening, supply fan frequency, return fan frequency, return air dew point temperature, hall air dry bulb temperature and relative humidity set in time series.

[0010] According to the closed-loop controller parameter optimization method for a uranium enrichment auxiliary control process provided by the present invention, step 2 includes: Using the historical water system control parameters corresponding to different scenarios, large water system models corresponding to different scenarios are regularly constructed; based on the historical air conditioning system control parameters corresponding to different scenarios, large air conditioning system models corresponding to different scenarios are regularly constructed; The input data of the water system large model include: cooling water pump frequency, cooling tower return water temperature, cooling tower inlet water temperature set in time series; The input data of the large air conditioning system model include: valve opening, supply fan frequency, return fan frequency, return air dew point temperature, hall air dry bulb temperature and relative humidity set in time series.

[0011] According to the closed-loop controller parameter optimization method for uranium enrichment auxiliary control process provided by the present invention, the large system model is constructed using the BP-P model; and the simplified system model is constructed using the ARMAX system identification training method.

[0012] The present invention also provides a closed-loop controller parameter optimization system applied to a uranium enrichment auxiliary control system, comprising: The capture unit is used to periodically capture the latest historical system control parameters corresponding to different scenarios from the real-time uranium enrichment auxiliary control system according to different scenarios; The construction unit is used to regularly build different system models based on the latest historical system control parameters corresponding to different scenarios; The simulation unit is used to simulate the current system large model in different scenarios respectively, and obtain the simplified system models corresponding to the different scenarios after simulation; A comparison unit, configured to compare the accuracy of the simplified system model and the original simplified system model of the closed-loop controller under the same scenario; The updating unit is used to determine whether to update the current parameters of the closed-loop controller according to the comparison result.

[0013] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the closed-loop controller parameter optimization method for a uranium enrichment auxiliary control process as described above is implemented.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the closed-loop controller parameter optimization method applied to a uranium enrichment auxiliary control process as described above is implemented.

[0015] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the closed-loop controller parameter optimization method applied to a uranium enrichment auxiliary control process as described above is implemented.

[0016] The closed-loop controller parameter optimization method and system for uranium enrichment auxiliary control processes, provided by this invention, achieves an automated closed-loop process of "data-driven → model update → parameter optimization" by capturing historical data, building a large system model, and then streamlining model simulation, accuracy comparison, and parameter updates without requiring human intervention. Furthermore, through a regular update mechanism (e.g., periodic data capture and modeling), it can promptly capture changes such as outdoor temperature and humidity disturbances and equipment performance degradation, enabling parameter adjustments to be completed without human intervention, thus avoiding system fluctuations caused by delayed adjustments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flow chart of the closed-loop controller parameter optimization method for a uranium enrichment auxiliary control process provided by the present invention; Figure 2 This is a schematic diagram of the structure of the closed-loop controller parameter optimization system applied to the uranium enrichment auxiliary control system provided by the present invention; Figure 3 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0018] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0019] The uranium enrichment auxiliary control system is an auxiliary process system that supports uranium enrichment production. It primarily includes the water system and the air conditioning system, responsible for providing a stable cooling load and heat exchange environment. Closed-loop controllers directly operate on auxiliary control system equipment (such as cooling towers, fans, and pumps), adjusting their operating parameters (frequency, number of units, etc.) to control cooling output, ultimately maintaining stable temperature and humidity within the uranium enrichment plant. The core role of the closed-loop controller in the uranium enrichment auxiliary control system is to precisely control key parameters such as temperature and humidity (ensuring minimal fluctuations) through dynamic parameter updates and performance tuning. For example, the closed-loop controller for the water system receives inputs of the target temperature and the current actual temperature. The closed-loop controller, through internal calculations, outputs valve openings and fan frequency, thereby ensuring a stable, constant temperature and humidity environment for key uranium enrichment equipment.

[0020] In the uranium enrichment auxiliary control system, the working principle of the closed-loop controller is as follows: The closed-loop controller dynamically adjusts auxiliary control equipment (such as cooling towers, fans, and pumps) by integrating historical production data with an embedded streamlined system model. Input data includes the current actual temperature, target temperature, current valve opening, current fan frequency, prediction time, and execution interval. Based on this input, the controller leverages the embedded system model to simulate system responses under different control strategies within a preset prediction timeframe. For example, when input parameters such as outdoor temperature and humidity changes and the current fan frequency are used, the system model predicts temperature fluctuation trends over a period of time. Using an optimization algorithm (such as model predictive control) and combining its own parameters (such as weights and constraints), the controller calculates the cumulative deviation and ultimately selects the control solution that minimizes the deviation. The controller then outputs commands for valve opening, fan frequency, and other parameters at preset execution intervals to drive the auxiliary control equipment to adjust their operating states.

[0021] It is worth noting that there is an inverse relationship between the input and output data of the closed-loop controller and the input and output data of the embedded system model: the input of the embedded system model is control instructions (such as changes in valve opening) and disturbance factors (such as outdoor temperature and humidity), and the output is the system state (such as temperature changes), which is used to simulate the process of "control action → system response"; while the closed-loop controller uses the current state of the system (such as the deviation between the actual temperature and the target temperature) as input and outputs the corresponding control instructions, realizing the reverse deduction of "system state → control action".

[0022] Figure 1 The flow chart of the closed-loop controller parameter optimization method for uranium enrichment auxiliary control process provided by the present invention is as follows: Figure 1 As shown, the method includes the following steps: Step 1: Based on different scenarios, periodically capture historical system control parameters corresponding to different scenarios from the real-time uranium enrichment auxiliary control system. Automatically separate the data by scenario and then construct a data structure.

[0023] Specifically, the uranium enrichment auxiliary control system is primarily responsible for maintaining a constant temperature and humidity environment for key uranium enrichment equipment, as well as heat exchange during the production process. The uranium enrichment auxiliary control system is divided into a water system and an air conditioning system. Within the uranium enrichment auxiliary control system, both the water system and the air conditioning system require effective cooling capacity to maintain constant temperature and humidity conditions. Therefore, the historical system control parameters provided by embodiments of the present invention include: historical water system control parameters and historical air conditioning system control parameters. The historical water system control parameters include: cooling water pump frequency, cooling tower return water temperature, and cooling tower inlet water temperature, set in a time series. The historical air conditioning system control parameters include: valve opening, supply fan frequency, return fan frequency, return air dew point temperature, hall air dry-bulb temperature, and relative humidity, set in a time series.

[0024] As a core component of the refrigeration system, the compressor's performance directly impacts the cooling effect, which in turn affects the stability of the uranium enrichment process and product quality. In this embodiment of the present invention, different scenarios correspond to different numbers of compressors. One compressor corresponds to one uranium enrichment auxiliary control scenario; two compressors correspond to another uranium enrichment auxiliary control scenario. For example, if there are 10 compressors, there are a total of 10 uranium enrichment auxiliary control scenarios.

[0025] Specifically, the primary task of the uranium enrichment auxiliary control system is to ensure that key uranium enrichment equipment maintains a constant temperature and humidity environment, which is crucial for ensuring production process stability and product quality. The compressor, the "heart" of the refrigeration system, compresses the refrigerant through circulation, transferring heat and maintaining the required low temperature. In the uranium enrichment auxiliary control system, compressor performance directly affects the cooling effect, and thus the accuracy of temperature and humidity control.

[0026] In the embodiments of this invention, different numbers of compressors correspond to different auxiliary control scenarios for uranium enrichment. For example, one compressor may be used for a specific production scenario or operating condition, while two or more compressors may be used to meet the auxiliary control needs of higher production capacity or more complex operating conditions. This correspondence demonstrates the flexibility and scalability of the system design, enabling the auxiliary control system to adjust cooling capacity based on actual production needs, ensuring an optimal constant temperature and humidity environment in different scenarios.

[0027] When capturing historical system control parameters from the uranium enrichment auxiliary control process control system, first capture the historical system control parameters of all scenarios from the relevant storage media, and then classify all historical system control parameters according to different scenarios to obtain the latest historical system control parameters corresponding to different scenarios.

[0028] Historical system control parameters include historical water system control parameters and historical air conditioning system control parameters. Historical water system control parameters include: cooling water pump frequency, cooling tower return water temperature, and cooling tower inlet water temperature, all set in a time series. Historical air conditioning system control parameters include: valve opening, supply fan frequency, return fan frequency, return air dew point temperature, hall air dry-bulb temperature, and relative humidity, all set in a time series.

[0029] The historical water system control parameters include time-series data on the cooling water pump frequency, cooling tower return water temperature, and cooling tower inlet water temperature. These parameters directly reflect the operating status of the water system (e.g., closed tower and open tower). The cooling water pump frequency is the actuator's operating variable, determining the water circulation rate; the return water temperature and inlet water temperature are the outputs of the controlled object, reflecting the cooling effect and system heat exchange efficiency.

[0030] Historical air conditioning system control parameters include time-series data on valve opening, supply fan frequency, return fan frequency, return air dew point temperature, hall air dry-bulb temperature, and relative humidity. Valve opening and supply / return fan frequency are the system's operating variables, used to adjust air volume and refrigerant flow. Return air dew point temperature and hall air temperature and humidity are the outputs of the controlled objects, directly related to the constant temperature and humidity required for key uranium enrichment equipment.

[0031] Step 2: Regularly build different system models based on the latest historical system control parameters corresponding to different scenarios.

[0032] Specifically, for each scenario, define corresponding historical system control parameters. Under each scenario, regularly collect data on these historical system control parameters. The frequency and duration of data collection should be determined based on actual needs and system characteristics. Preprocess the collected raw data, including discrete and continuous signal identification, continuous signal correlation analysis, noise removal, and steady-state data extraction, to improve data quality and validity, providing a reliable foundation for subsequent modeling.

[0033] Select an appropriate modeling method based on the preprocessed data for each scenario. This paper recommends using a neural network-based modeling approach, particularly the PSO-BP method, which combines the particle swarm optimization (PSO) algorithm with a back-propagation neural network (BP). The selected model is trained using the preprocessed data. During training, the PSO algorithm is used to find the optimal initial parameters and number of hidden layer nodes in the neural network, thereby reducing the number of random training runs and improving training efficiency.

[0034] In this embodiment of the present invention, a large system model is constructed using the BP-SO training method. Due to the strong nonlinearity, time-varying disturbances (such as changes in outdoor temperature and humidity), and discrete events (such as compressor additions and subtractions) present in the uranium enrichment auxiliary control system, traditional BP neural networks are prone to falling into local optimal solutions, resulting in the model's inability to accurately characterize the dynamic relationship between control commands (such as valve opening and pump frequency) and system responses (such as temperature and humidity, and return water temperature). The BP-PSO training method uses the PSO algorithm to optimize the initial weights and number of hidden layer nodes of the BP neural network, reducing errors caused by random initialization. This allows the model to more efficiently exploit nonlinear relationships in historical control parameters (such as time series data on valve opening and temperature and humidity), thereby improving the accuracy of the fitting of the system's dynamic characteristics and providing a reliable foundation for the subsequent construction of a streamlined model.

[0035] The following describes how to regularly build different system models based on the latest historical system control parameters corresponding to different scenarios. Specific solutions include the following: Using the historical water system control parameters corresponding to different scenarios, large water system models corresponding to different scenarios are regularly constructed; based on the historical air-conditioning system control parameters corresponding to different scenarios, large air-conditioning system models corresponding to different scenarios are regularly constructed; the input data of the water system large model include: cooling water pump frequency and cooling tower inlet water temperature set according to time series; the output data include: cooling tower return water temperature; the input data of the air-conditioning system large model include: valve opening, air supply probability, return fan frequency, return air dew point temperature set according to time series; the output data include: hall air dry-bulb temperature and relative humidity.

[0036] The construction principle of the large model of the air-conditioning system is similar to that of the large model of the water system, so it will not be repeated here.

[0037] The simplified system model provided by the present invention is constructed by adopting the system identification training method of ARMAX.

[0038] Production data from uranium enrichment auxiliary control systems primarily consists of closed-loop operational data (e.g., temperature and humidity during equipment dynamic adjustment, actuator frequency, etc.). Traditional model identification relies on open-loop experimental data (e.g., step response), while the ARMAX model can directly utilize closed-loop data for identification. The primary disturbances in the auxiliary control system (e.g., outdoor temperature and humidity) are random and uncontrollable. The ARMAX model, by introducing "autoregressive terms" and "sliding average terms," ​​can quantify the impact of disturbances on system output (e.g., the lagged effect of outdoor temperature fluctuations on cooling tower return water temperature). The ARMAX model is a parametric identification method that balances model complexity and accuracy by adjusting the order (e.g., the order of the autoregressive term and the order of the sliding average term), ultimately outputting a low-order model with a clear mathematical expression.

[0039] Step 3: Simulate the current system large model for different scenarios respectively, and obtain the simplified system models corresponding to the different scenarios after simulation.

[0040] Specifically, in the uranium enrichment auxiliary control system, directly using the current system large model as the embedded simplified model of the closed-loop controller will cause the following problems: First, taking the large air conditioning system model as an example, since temperature data is strictly controlled within a preset range, when the temperature deviates from the preset value (for example, by 0.1 degrees), the system immediately adjusts the valve opening and fan frequency to restore the temperature to normal. This makes it impossible to directly observe the complete impact of changes in valve opening or fan frequency on temperature. As a result, the input and output data of the large air conditioning system model suffer from unclear mapping relationships. Secondly, if the large system model is directly embedded in a closed-loop controller, the large data volume and complex calculations will result in slow calculations when the closed-loop controller runs the model, failing to meet the requirements of real-time control. To meet real-time control requirements, the closed-loop controller must complete calculations and output control commands in an extremely short time (e.g., seconds). However, due to the large data volume and complex calculations of the large system model, the real-time calculation requirements of the closed-loop controller cannot be met.

[0041] Therefore, the present invention uses a simulation model to construct a simplified system model. Since the simplified system model ensures the control accuracy of the model through simulation optimization and feature extraction while retaining the main dynamic characteristics of the large system model, the simplified system model significantly improves the computing efficiency by reducing the amount of data and simplifying the calculation process, so that the closed-loop controller can complete the calculation and output control instructions in a very short time, so it can better adapt to the changes in the dynamic characteristics of the system and improve the stability of the system.

[0042] Step 4: Compare the accuracy of the simplified system model and the original simplified system model of the closed-loop controller under the same scenario, and determine whether to update the current parameters of the closed-loop controller based on the comparison results.

[0043] Specifically, the closed-loop controller's parameters (such as weights, constraints, and adjustment coefficients) must match the dynamic characteristics of the controlled object. The embedded streamlined model is a mathematical representation of the dynamic characteristics of the controlled object (the uranium enrichment auxiliary control system equipment). When the closed-loop controller's embedded system model is updated, the controller parameters must be adjusted to accommodate the changes in equipment performance reflected by the new model, ultimately optimizing control accuracy and stability.

[0044] The present invention verifies whether the newly constructed streamlined system model is better than the original streamlined model through accuracy comparison, and then determines whether the controller parameters need to be updated. Specifically, since the parameters of the closed-loop controller need to match the dynamic characteristics of the embedded streamlined model, and the streamlined model is obtained based on the latest production data and simulation optimization, it is more in line with the current working conditions. If its accuracy is significantly better than the original system streamlined model, it means that the parameters of the original closed-loop controller are no longer suitable for the new working conditions and need to be optimized by updating the parameters of the model, ultimately improving the control accuracy and stability.

[0045] The closed-loop controller parameter optimization method for uranium enrichment auxiliary control processes, provided by this invention, achieves an automated closed-loop process of "data-driven → model update → parameter optimization" without human intervention, from capturing historical data, building a large system model, to streamlined model simulation, accuracy comparison, and parameter updates. This allows controller parameters to dynamically adapt over the system lifecycle (e.g., equipment aging, process upgrades). Furthermore, through a regular update mechanism (e.g., periodic data capture and modeling), it can promptly capture changes such as outdoor temperature and humidity disturbances and equipment performance degradation, allowing parameter adjustments to be completed without human intervention, thus avoiding system fluctuations caused by delayed regulation.

[0046] Furthermore, the following introduces how to simulate the current system big model, which specifically includes the following solutions: First, for different scenarios, the first system input data within the preset time period is customized respectively, and the first system input data corresponding to different scenarios are used as the input of the current system big model to obtain the first system output data of the current system big model corresponding to different scenarios in the future preset time period; secondly, the first system input data and the first system output data of the same scenario are used to construct a simplified system model of the corresponding scenario, thereby obtaining the simplified system models corresponding to different scenarios after simulation.

[0047] Specifically, for different scenarios of the uranium enrichment auxiliary control system (i.e., operating conditions corresponding to different numbers of compressors), customized "first system input data" are defined. These data are sequences of control instructions within a preset time period (e.g., a time series of valve opening, fan frequency, and water pump frequency within the next 30 minutes). In this embodiment of the present invention, for the large water system model, the input is a sequence of valve opening and fan frequency, and the output is a sequence of changes in hall air temperature and humidity. For the large water system model, the input is a sequence of pump frequency, and the output is a sequence of changes in cooling tower return water temperature. The essence of this process is to simulate the dynamic process of "control instruction → system response," compensating for the limitation of strict temperature control within ±0.1°C in actual production (which makes it impossible to observe the complete regulation effect) and obtaining a complete input-output mapping relationship.

[0048] This invention utilizes the "first system input data" and "first system output data" obtained in the same scenario to construct a "streamlined system model" through simulation optimization methods such as feature extraction and dimensionality compression. Compared with the large system model, the streamlined model retains core dynamic characteristics (such as the trend of the impact of valve opening changes on temperature), but significantly improves computational efficiency by reducing the amount of data and simplifying the calculation logic (such as converting complex neural network models into low-order transfer function models). In addition, due to the differences in system characteristics between different scenarios (such as different cooling capacities due to different numbers of compressors), a separate streamlined model must be constructed for each scenario to ensure that the model and scenario are compatible.

[0049] This invention solves the problem of "unclear input-output mapping relationship" caused by too small a temperature fluctuation range (for example, a temperature fluctuation of ±0.1°C) in actual production through customized input and simulation, providing a complete data foundation for model construction. Moreover, the large system model cannot meet the second-level response requirements of the closed-loop controller due to the large amount of data and complex calculations. The streamlined model, by simplifying the design, can quickly complete the calculation and output control instructions, adapting to real-time control scenarios. In addition, the streamlined model is constructed based on the simulation data of the large system model, inheriting the large model's accurate description of the system's dynamic characteristics, ensuring that the core accuracy is not lost while simplifying, and providing a reliable model foundation for subsequent controller parameter optimization.

[0050] Furthermore, after constructing a large system model based on historical data, simulations were performed using custom inputs (such as preset valve opening sequences) to generate output data (first system output data) covering various operating conditions, thus compensating for the incompleteness of actual data. This streamlined model, built from simulation data, can cover rare operating conditions such as extreme weather and equipment additions and subtractions, ensuring stable controller operation in unexpected scenarios and enhancing system robustness.

[0051] Furthermore, the following describes how to determine whether to update the current parameters of the closed-loop controller, specifically including the following solutions: First, the first system input data under the same scenario is used as the input of the original system simplified model of the closed-loop controller and the input of the current system simplified model, respectively, to obtain the second system output data of the original system simplified model under the corresponding scenario and the third system output data of the current system simplified model; then, the difference between the second system output data and the first system output data is determined to obtain a first difference; the difference between the third system output data and the first system output data is determined to obtain a second difference; then, the difference between the first difference and the second difference is compared to see whether it is outside the preset range. If it is outside the preset range, the original system simplified model of the controller is updated to the current system simplified model; finally, the current parameters of the closed-loop controller are updated according to the current system simplified model.

[0052] Specifically, this invention aims to address the issue of "degradation of the original model's accuracy" in uranium enrichment auxiliary control systems due to operating condition changes (such as outdoor temperature and humidity disturbances). By scientifically comparing the performance of a newly constructed streamlined model with the original model, it determines whether closed-loop controller parameters need to be updated, ultimately achieving "control accuracy that matches the system's dynamic characteristics." This implementation assumes that the comparison is based on the same scenario (e.g., operating conditions corresponding to a specific number of compressors), ensuring fairness and pertinence.

[0053] Since the first system output data is the output generated by the current system macromodel after inputting the "first system input data," the macromodel is constructed using methods such as the PSO-BP algorithm based on the latest historical system control parameters (such as time series data on valve opening, fan frequency, temperature and humidity) under the same scenario. This model accurately depicts the dynamic characteristics of the uranium enrichment auxiliary control system under current operating conditions. Its construction relies on regular updates of historical production data, reflecting changes in system characteristics caused by operating conditions such as seasonal variations and equipment performance degradation. Therefore, the first system output data is essentially a "simulated reproduction" of the system's actual response under current operating conditions, providing a reliable benchmark. To this end, this application simultaneously inputs customized "first system input data" (such as valve opening and fan frequency series within a preset time period) into two models: the "original system simplified model" currently in use in the closed-loop controller, and the newly constructed "current system simplified model." Based on the original system simplified model, the "second system output data" (such as temperature and humidity changes predicted by the old model) are output; based on the current system simplified model, the "third system output data" (such as temperature and humidity changes predicted by the new model) are output. Then, two sets of key differences are calculated using mathematical methods (such as mean square error and cumulative absolute error). The first difference reflects the deviation between the original model and the actual system characteristics—that is, the difference between the "second system output data" (predicted by the old model) and the "first system output data" (the baseline output of the large system model simulation, representing the actual response under the current operating conditions). The second difference reflects the deviation between the new streamlined model and the actual system characteristics—that is, the difference between the "third system output data" (predicted by the new model) and the "first system output data." For example, if the first difference is 0.5°C (the old model error) and the second difference is 0.1°C (the new model error), the streamlined model more accurately describes the current operating conditions.

[0054] The so-called preset range is a threshold set based on the system stability and control accuracy requirements, which is used to determine whether the streamlined model is significantly better than the original model. For example, if the difference between the first difference and the second difference is greater than or equal to 10%, if this condition is met, it means that the original model can no longer adapt to the current operating conditions, and the embedded streamlined model of the closed-loop controller needs to be updated to the "current system streamlined model". In addition, due to the large differences in system characteristics in different scenarios (different numbers of compressors), the threshold judgment needs to be limited to the "same scenario" to avoid misjudgment caused by cross-scenario comparison (such as the model accuracy standard for low-load scenarios is not applicable to high-load scenarios). After the model is updated, the parameters of the closed-loop controller (such as adjustment weights, prediction time domain, execution interval, etc.) need to be recalculated and updated according to the dynamic characteristics of the new streamlined model.

[0055] The overall solution of the present invention is introduced below: The following is an overall introduction to the solution of the present invention, taking a single scenario as an example, which specifically includes the following steps: S1: Regularly capture the latest historical system control parameters from the real-time uranium enrichment auxiliary control process control system; the historical system control parameters include: historical water system control parameters, historical air conditioning system control parameters.

[0056] S2: Regularly construct different water system large models based on the historical water system control parameters obtained regularly; regularly construct the air conditioning system large model based on the historical air conditioning system control parameters; S3: Customize the first water system input data within a preset time period and use it as input to the current water system macromodel, and correspondingly obtain the first water system output data of the current water system macromodel in the future preset time period; customize the first air conditioning input data within the preset time period and use it as input to the current air conditioning system macromodel, and correspondingly obtain the first air conditioning output data of the current air conditioning system macromodel in the future preset time period; S4: constructing a simplified model of the current water system based on the first water system input data and the first water system output data; constructing a simplified model of the current air conditioner based on the first air conditioner input data and the first air conditioner output data; S5: Using the first water system input data as input to the original water system simplified model of the closed-loop controller, and correspondingly obtaining second water system output data of the original water system simplified model; using the first water system input data as input to the current water system simplified model, and correspondingly obtaining third water system output data of the current water system simplified model; Using the first air-conditioning input data as the input of the original air-conditioning simplified model of the closed-loop controller, and correspondingly obtaining the second air-conditioning output data of the original air-conditioning simplified model; using the first air-conditioning input data as the input of the current air-conditioning simplified model, and correspondingly obtaining the third air-conditioning output data of the current air-conditioning simplified model; S6: performing a mean square error operation on the second water system output data and the first water system output data to obtain a first difference; performing a mean square error operation on the third water system output data and the first water system output data to obtain a second difference; performing a mean square error operation on the second air conditioner output data and the first air conditioner output data to obtain a third difference; performing a mean square error operation on the third air conditioner output data and the first air conditioner output data to obtain a fourth difference; S7: comparing whether the difference between the first difference and the second difference is within a preset range; if it is outside the preset range, updating the original water system simplified model of the closed-loop controller to the current water system simplified model; Compare the third difference value and the fourth difference value to determine whether the difference is outside a preset range. If so, update the original air-conditioning simplified model of the closed-loop controller to the current air-conditioning simplified model.

[0057] S8: Update the current parameters of the closed-loop controller according to the current water system simplified model or the current air conditioning simplified model.

[0058] The following describes a closed-loop controller parameter optimization system for a uranium enrichment auxiliary control system provided by the present invention. The closed-loop controller parameter optimization system for a uranium enrichment auxiliary control system described below and the closed-loop controller parameter optimization method for a uranium enrichment auxiliary control process described above can be referenced to each other.

[0059] Figure 2 This is a schematic diagram of the structure of the closed-loop controller parameter optimization system for the uranium enrichment auxiliary control system provided by the present invention, as shown in FIG. Figure 2 As shown, the system includes: The capture unit 201 is used to periodically capture the latest historical system control parameters corresponding to different scenarios from the real-time uranium enrichment auxiliary control system according to different scenarios; A construction unit 202 is used to periodically construct different system macro models based on the latest historical system control parameters corresponding to different scenarios; The simulation unit 203 is used to simulate the current system large model in different scenarios respectively, and obtain the simplified system models corresponding to the different scenarios after simulation; A comparison unit 204 is configured to compare the accuracy of the simplified system model and the original simplified system model of the closed-loop controller under the same scenario; The updating unit 205 is configured to determine whether to update the current parameters of the closed-loop controller according to the comparison result.

[0060] Figure 3 An example of a physical structure diagram of an electronic device is shown below. Figure 3As shown, the electronic device may include: a processor 310, a communications interface 320, a memory 330, and a communications bus 340, wherein the processor 310, the communications interface 320, and the memory 330 communicate with each other via the communications bus 340. The processor 310 may call logic instructions in the memory 330 to execute a closed-loop controller parameter optimization method applied to a uranium enrichment auxiliary control process, the method comprising: Step 1: Based on different scenarios, regularly capture the latest historical system control parameters corresponding to different scenarios from the real-time uranium enrichment auxiliary control system; Step 2: Regularly build different system models based on the latest historical system control parameters corresponding to different scenarios; Step 3: Simulate the current system large model for different scenarios respectively, and obtain the corresponding simplified system models for different scenarios after simulation; Step 4: Compare the accuracy of the simplified system model and the original simplified system model of the closed-loop controller under the same scenario, and determine whether to update the current parameters of the closed-loop controller based on the comparison result.

[0061] In addition, the logic instructions in the aforementioned memory 330 can be implemented in the form of a software functional unit and, when sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or the portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0062] In another aspect, the present invention further provides a computer program product, comprising a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the closed-loop controller parameter optimization method for a uranium enrichment auxiliary control process provided by the above methods, the method comprising: Step 1: Based on different scenarios, regularly capture the latest historical system control parameters corresponding to different scenarios from the real-time uranium enrichment auxiliary control system; Step 2: Regularly build different system models based on the latest historical system control parameters corresponding to different scenarios; Step 3: Simulate the current system large model for different scenarios respectively, and obtain the corresponding simplified system models for different scenarios after simulation; Step 4: Compare the accuracy of the simplified system model and the original simplified system model of the closed-loop controller under the same scenario, and determine whether to update the current parameters of the closed-loop controller based on the comparison result.

[0063] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the closed-loop controller parameter optimization method for a uranium enrichment auxiliary control process provided by the above methods is implemented. The method comprises: Step 1: Based on different scenarios, regularly capture the latest historical system control parameters corresponding to different scenarios from the real-time uranium enrichment auxiliary control system; Step 2: Regularly build different system models based on the latest historical system control parameters corresponding to different scenarios; Step 3: Simulate the current system large model for different scenarios respectively, and obtain the corresponding simplified system models for different scenarios after simulation; Step 4: Compare the accuracy of the simplified system model and the original simplified system model of the closed-loop controller under the same scenario, and determine whether to update the current parameters of the closed-loop controller based on the comparison result.

[0064] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units. That is, they may be located in one place or distributed across multiple network units. Some or all of these modules may be selected based on actual needs to achieve the objectives of this embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0065] Through the description of the above embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A closed-loop controller parameter optimization method for a uranium enrichment auxiliary control process, characterized in that: include: Step 1: Based on different scenarios, regularly capture the latest historical system control parameters corresponding to different scenarios from the real-time uranium enrichment auxiliary control system; Step 2: Regularly build different system models based on the latest historical system control parameters corresponding to different scenarios; Step 3: Simulate the current system large model for different scenarios respectively, and obtain the corresponding simplified system models for different scenarios after simulation; Step 4: Compare the accuracy of the simplified system model and the original simplified system model of the closed-loop controller under the same scenario, and determine whether to update the current parameters of the closed-loop controller based on the comparison result.

2. The closed-loop controller parameter optimization method for a uranium enrichment auxiliary control process according to claim 1, characterized in that: The step three includes: Step 31: For different scenarios, first system input data within a preset time period is customized respectively, and the first system input data corresponding to each scenario is used as input to the current system macro model, thereby obtaining first system output data of the current system macro model corresponding to each scenario within a future preset time period; Step 32: Using the first system input data and the first system output data of the same scenario, a simplified system model of the corresponding scenario is constructed, thereby obtaining simplified system models corresponding to different scenarios after simulation.

3. The closed-loop controller parameter optimization method for a uranium enrichment auxiliary control process according to claim 2, characterized in that: The fourth step includes: Step 41: Using the first system input data under the same scenario as the input of the original system simplified model of the closed-loop controller, and correspondingly obtaining the second system output data of the original system simplified model under the corresponding scenario; using the first system input data corresponding to the same scenario as the input of the current system simplified model, and correspondingly obtaining the third system output data of the current system simplified model under the same scenario; Step 42: Determine the difference between the second system output data and the first system output data to obtain a first difference value; determine the difference between the third system output data and the first system output data to obtain a second difference value; Step 43: comparing the first difference value and the second difference value to see whether the difference is within a preset range; if the difference is outside the preset range, updating the original simplified system model of the controller to the current simplified system model; Step 44: Update the current parameters of the closed-loop controller according to the current system simplified model.

4. The closed-loop controller parameter optimization method for a uranium enrichment auxiliary control process according to any one of claims 1 to 3, characterized in that: The historical system control parameters include: historical water system control parameters, historical air conditioning system control parameters; The real-time air conditioning system control parameters include: valve opening, supply fan frequency, return fan frequency, return air dew point temperature, hall air dry bulb temperature and relative humidity set in time sequence; The historical water system control parameters include: cooling water pump frequency, cooling tower return water temperature, and cooling tower inlet water temperature set according to time series.

5. The closed-loop controller parameter optimization method for a uranium enrichment auxiliary control process according to claim 4, characterized in that: The second step includes: Using the historical water system control parameters corresponding to different scenarios, large water system models corresponding to different scenarios are regularly constructed; based on the historical air conditioning system control parameters corresponding to different scenarios, large air conditioning system models corresponding to different scenarios are regularly constructed; The input data of the water system large model include: cooling water pump frequency and cooling tower inlet water temperature set according to time series; output data include: cooling tower return water temperature; The input data of the large air conditioning system model include: valve opening, air supply probability, return fan frequency, and return air dew point temperature set in time series; the output data include: hall air dry bulb temperature and relative humidity.

6. The closed-loop controller parameter optimization method for a uranium enrichment auxiliary control process according to claim 4, wherein the large system model is constructed using a BP-PSO training method; and the simplified system model is constructed using an ARMAX system identification training method.

7. A closed-loop controller parameter optimization system applied to a uranium enrichment auxiliary control system, characterized in that: include: The capture unit is used to periodically capture the latest historical system control parameters corresponding to different scenarios from the real-time uranium enrichment auxiliary control system according to different scenarios; The construction unit is used to regularly build different system models based on the latest historical system control parameters corresponding to different scenarios; The simulation unit is used to simulate the current system large model in different scenarios respectively, and obtain the simplified system models corresponding to the different scenarios after simulation; A comparison unit, configured to compare the accuracy of the simplified system model and the original simplified system model of the closed-loop controller under the same scenario; The updating unit is used to determine whether to update the current parameters of the closed-loop controller according to the comparison result.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the closed-loop controller parameter optimization applied to the uranium enrichment auxiliary control system as claimed in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the closed-loop controller parameter optimization applied to the uranium enrichment auxiliary control system as claimed in any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the closed-loop controller parameter optimization method applied to a uranium enrichment auxiliary control process as claimed in any one of claims 1 to 6 is implemented.

Images