Xenon oscillation automatic suppression auxiliary early warning prompting method based on 3AO method

Abstract

The invention discloses a xenon oscillation automatic suppression auxiliary early warning prompting method based on a 3AO method, and belongs to the technical field of nuclear power reactor core monitoring. Firstly, various high-precision sensors are arranged at key parts of a nuclear reactor, occurrence signs of xenon oscillation are captured in time, and the system can accurately identify early signals of xenon oscillation by analyzing data of the sensors, so that a basis is provided for formulation of a subsequent control strategy. The core of the control strategy is fusion of a 3AO method and a deep Q network algorithm, the algorithm comprehensively considers the real-time state of the reactor core, and a control decision is optimized by learning historical data, so that the optimal control rod position and boron concentration adjustment scheme is calculated. And then, the system automatically prompts an operator to adjust the control rod and the boron concentration according to a control instruction generated by an intelligent algorithm so as to respond to the change requirement of the nuclear reactor in real time, and an efficient closed-loop control system is formed by feeding back an adjustment effect in real time, so that the accuracy and the reaction speed of a control strategy can be continuously improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent monitoring of nuclear power cores, and specifically discloses a xenon oscillation automatic suppression auxiliary early warning prompt method based on a 3AO method. Background Art

[0002] At present, domestic nuclear power units are increasingly frequently adjusting their loads in response to grid requirements. Frequent load change requirements will bring certain challenges to reactor control and core safety management. At the same time, load changes at the end of the service life may cause xenon oscillations, which pose certain challenges to the operator's control capabilities. There is a risk of local power distortion in the core due to improper control. For the second-generation and second-generation plus units that have been in operation for many years, there is no possibility of in-core self-powered detectors and similar technical transformation. In order to have the core online monitoring function of the third-generation pressurized water reactor unit and enhance the unit's control ability to cope with load tracking changes, the existing xenon oscillation suppression methods mostly rely on the operator to manually adjust the control rods and boron concentration, which have a slow reaction speed and are easily affected by human factors, and there are certain safety hazards. Summary of the invention

[0003] The present invention aims to solve the problem of monitoring and early warning of xenon oscillations in nuclear reactor operations. In the operation of traditional nuclear reactors, the occurrence of xenon oscillations is often due to the untimely response of manual or semi-automatic control systems, resulting in unstable nuclear reactor power, which may cause safety accidents in severe cases. In addition, in traditional methods, operators need to continuously monitor the status of the reactor, which not only increases the complexity of operation, but also increases the risk of human error. Therefore, the present invention needs to solve the following technical problems:

[0004] 1. How to accurately monitor and predict the occurrence of xenon oscillations in real time so as to respond in time.

[0005] 2. How to adjust the control rod position and boron concentration with auxiliary early warning prompts, reduce manual operations, and improve control accuracy and efficiency.

[0006] 3. How to ensure the safety and stability of nuclear reactors under different operating conditions and avoid reactor accidents caused by improper control.

[0007] In order to overcome the deficiencies in the prior art, the present invention provides a xenon oscillation automatic suppression auxiliary early warning prompt method based on the 3AO method, which can respond quickly when xenon oscillation occurs, and automatically prompt the operator to adjust the control rod and boron concentration to suppress the occurrence of xenon oscillation and ensure the safe and stable operation of the core. The technical solution of the present invention is:

[0008] A xenon oscillation automatic suppression auxiliary early warning prompt method based on the 3AO method comprises the following steps:

[0009] Step 1: monitor the core operating parameters in real time and use a convolutional neural network to identify the occurrence of xenon oscillations;

[0010] Step 2: Based on the 3AO algorithm, calculate the optimal control rod and boron concentration adjustment plan to adapt to the dynamic changes under different operating conditions.

[0011] Step 3: According to the instructions of the control module, the operator is automatically prompted to adjust the control rod position and boron concentration.

[0012] Step 4: Provide real-time feedback on the adjustment effects, analyze the differences between the adjustment results and the expected goals, and dynamically optimize subsequent adjustment strategies.

[0013] Further, the specific process of step 1 is:

[0014] First, high-precision sensors are installed at key locations of the core to comprehensively monitor the operating status of the core.

[0015] Temperature sensors: Temperature sensors are evenly arranged around the central axis to monitor the temperature changes of the core and identify hot spots and temperature anomalies.

[0016] Pressure sensors: Pressure sensors are installed at the top and bottom of the core, distributed in four quadrants, to monitor coolant pressure and identify pressure fluctuations and anomalies.

[0017] Neutron flux (power): RPN power range detector of the extra-core nuclear measurement system.

[0018] Coolant flow sensor: Install flow sensors at the coolant inlet and outlet to monitor the coolant flow and ensure stable operation of the cooling system.

[0019] The collected raw sensor data is denoised, normalized and feature extracted; a moving average filter is used to remove high-frequency noise, and feature extraction is performed after normalization to eliminate the influence of different dimensions. The time domain features of each sensor are extracted: mean, variance, kurtosis, skewness, maximum value, minimum value; frequency domain features: spectrum energy, main frequency and time-frequency domain features: wavelet transform coefficients, to form a comprehensive feature vector. Then the sliding window technology is used to divide the continuous data into multiple overlapping windows, each window As a sample input convolutional neural network model.

[0020]

[0021] in, is the time step of the window length, For the samples, is the time step data.

[0022] The convolutional neural network architecture is as follows:

[0023] Input layer: shape , that is, each sample contains time steps of data, each time step has Sensor data.

[0024] Convolutional layer 1: 64 (3,3) convolution kernels, activation function ReLU;

[0025] Pooling layer 1: max pooling, window size (2,2);

[0026] Convolutional layer 2: 128 (3,3) convolution kernels, activation function ReLU;

[0027] Pooling layer 2: max pooling, window size (2,2);

[0028] Fully connected layer 1: 256 nodes, activation function ReLU;

[0029] Output layer: 1 node, activation function Sigmoid, output probability of xenon oscillation ;

[0030] Loss Function and Optimization Algorithm

[0031] Binary cross entropy is used as the loss function to measure the difference between the model's predicted probability and the actual label:

[0032]

[0033] in, As the binary cross entropy loss function to guide model optimization, is the sample size, is the true label of the sample, 0 means no xenon oscillation, 1 means xenon oscillation occurs, Represents the model for The predicted probability of a sample.

[0034] Adopt Adam optimizer and combine gradient descent method to adaptively adjust learning rate:

[0035]

[0036]

[0037]

[0038]

[0039]

[0040] in, and are the estimates of the first and second order moments, respectively. and are the deviation-corrected forms of the first-order and second-order moments, respectively, represents the gradient, and are the decay rates of the first and second order moments, is the learning rate, ϵ is a small constant to prevent division by zero, is the value of the parameter before updating, The updated value of the parameter.

[0041] After updating the model parameters, the model performance is evaluated on the validation set. The model learning rate is adjusted according to the validation results to improve the model convergence speed and stability. A threshold is set for the real-time monitoring data. The probability of output xenon oscillation is predicted through a convolutional neural network to predict whether xenon oscillation will occur.

[0042] Further, the specific process of step 2 is:

[0043] The deep Q network algorithm is selected to be integrated with the 3AO method. The initial control scheme provided by the 3AO method is used, and the DQN algorithm performs autonomous learning and strategy optimization on this basis to calculate the optimal control rod and boron concentration adjustment scheme to adapt to the dynamic changes under different operating conditions, effectively improve the suppression effect of xenon oscillations, and ensure the safety and stability of core operation.

[0044] The fusion process is as follows:

[0045] Action space initialization

[0046] The basic control actions defined by the 3AO method are used as the initial action set of the reinforcement learning agent. On the basis of the 3AO method, the action space of the agent is expanded, more sophisticated control actions are introduced, and the control rod position and boron concentration are slightly adjusted to improve the control accuracy.

[0047]

[0048]

[0049]

[0050] in, Indicates the basic control actions defined by the 3AO method. represents the action extended by reinforcement learning, is the comprehensive action space, They are control rod raising, holding and lowering. The new actions are: control rod rising 0.5%, control rod falling 0.5%, boron concentration increasing 0.1ppm, boron concentration remaining unchanged, and boron concentration decreasing 0.1ppm.

[0051] Strategy initialization

[0052] The initial strategy of the agent prioritizes the actions provided by the 3AO method. , by introducing the ε-greedy strategy, gradually exploring the expansion action ; The hybrid control strategy is:

[0053]

[0054] in, Indicates the specific control action selected for execution. represents a decision rule, is an action selected from the action space A so that in the current state Under this condition, the Q function value is maximized. That is, under the state Next, select the one based on the current value function parameters Evaluate the action that will yield the greatest expected reward;

[0055] Strategy Optimization Process

[0056] Action selection, use - Greedy strategy, chooses random actions part of the time , to facilitate exploration.

[0057]

[0058] The Q network is updated, and the loss function is used to measure the difference between the target network and the main Q network:

[0059]

[0060] in, is the loss function of the DQN algorithm, Used to express the expected value of a variable. For the experience replay buffer, Represents the experience replay buffer A set of randomly sampled transfer samples , is the expectation on the experience replay buffer, It’s an instant reward. is the discount factor, is the current network parameter, are the parameters of the target Q network, Indicates the next state The maximum Q value under Indicates the current status Take action Q value;

[0061] The target network is synchronized, and the main Q network parameters are adjusted every certain number of steps. Copy to the target network: ,

[0062] Experience replay, transfer state Save experience replay buffer , randomly select small batches of samples for training. Through multiple rounds of training, the agent gradually optimizes the control strategy.

[0063] The system receives sensor data in real time, including the core temperature, pressure, neutron power, and coolant flow. The agent learns to choose the best action in different states to maximize the cumulative reward through interaction with the environment. The system adjusts the control rod position and boron concentration according to the best action. The action includes raising, lowering, or maintaining the current position of the control rod and adjusting the specific value of the boron concentration. The feedback module monitors the adjustment effect and dynamically optimizes the control strategy.

[0064] Parameters, the algorithm can ensure the smoothness and continuity of the entire adjustment process.

[0065] Further, the specific process of step 3 is:

[0066] The control rod position adjustment and boron concentration change calculated by the control module are converted into specific execution instructions through the intelligent control algorithm. After parsing the instructions, the specific control rod position adjustment and boron concentration change requirements are determined. The intelligent algorithm is responsible for strategy optimization and decision-making, while the PID controller is responsible for specific execution and fine-tuning, ensuring the efficient combination of the intelligent algorithm and the execution device, working together to achieve efficient xenon oscillation suppression.

[0067] Execution instruction generation: After receiving the optimal execution plan, the algorithm calculates the specific execution steps and necessary control parameters, and continuously adjusts Parameters, algorithms can ensure the stability and continuity of the entire adjustment process; calculate real-time control instructions :

[0068]

[0069] in, For real-time control instructions, is the systematic error, i.e., the difference between the target position or boron concentration and the current reading; They are proportional, integral and differential coefficients respectively; the PID controller judges and adjusts the position of the control rod and the boron concentration by inputting the current system error e(t), the integral of the error and the differential of the error, that is, the PID controller judges and adjusts the position of the control rod and the boron concentration by inputting the current position of the control rod and the current value of the boron concentration.

[0070] Control rod drive: A high-precision servo motor is used to drive the control rod to ensure high precision and fast response of the control rod position adjustment; the servo motor can achieve fine-tuning according to the instructions of the control module to reduce position error; in addition, the servo motor is also equipped with a position encoder to provide real-time feedback of the control rod position data for monitoring and optimization by the feedback module;

[0071] Boron concentration adjustment: Use intelligent flow control valves to control the mixing and distribution of boron solution to ensure accurate adjustment of boron concentration. Intelligent flow control valves can achieve fast and accurate flow control according to algorithm instructions; the actuator valve is equipped with a flow sensor to monitor the flow and concentration of the boron solution in real time, and feed back to the control module for dynamic adjustment;

[0072] In this method, the PID controller plays a key role in converting the adjustment output by the intelligent algorithm, i.e., the control rod position and the change in boron concentration, into specific execution instructions. Specifically, the PID controller converts the input data into the adjustment amount through the following workflow:

[0073] Data acquisition: The PID controller receives data from sensors, including the current position of the control rod and the current value of the boron concentration. These data are acquired in real time through the high-precision position encoder and boron concentration sensor installed on the control rod to ensure the accuracy and timeliness of the data.

[0074] Error calculation: The PID controller compares the current system state with the preset target state and calculates the system error. The system error includes the deviation of the control rod position and the boron concentration.

[0075] Error processing: The PID controller processes the current value, cumulative value and rate of change of the system error according to the three control links of proportion, integration and differentiation: the proportion P is to make corresponding adjustments immediately according to the size of the current error to reduce the error; the integral I is to accumulate past errors and eliminate long-standing steady-state errors; the differential D is to predict the changing trend of future errors and increase the stability and response speed of the system.

[0076] Adjustment amount calculation: Through the comprehensive effect of data acquisition, error calculation and error processing, the PID controller calculates accurate control instructions , that is, the position adjustment of the control rod and the change in boron concentration. These adjustments ensure that the adjustment of the control rod and boron concentration can accurately respond to changes in the system state and maintain the core operating parameters within the preset target range.

[0077] Execution and feedback: The PID controller sends the calculated control instructions to the servo motor and intelligent flow control valve to perform specific control rod position adjustment and boron concentration adjustment. The real-time feedback data of the actuator is input into the PID controller again to form a closed-loop control to ensure the continuous optimization and stability of the adjustment process.

[0078] Further, the specific process of step 4 is:

[0079] After the control adjustment is completed, the system will continue to collect the operating parameters of the reactor core, temperature, pressure, neutron flow and coolant flow, as well as the current position of the control rods and the current value of the boron concentration; these data provide the system with a basis for judging the effect of the adjustment.

[0080] The xenon oscillation monitoring interface provides clear xenon oscillation information feedback to operators by displaying the core's operating parameters and control status in real time. Operators can use this interface to monitor the core's real-time status, identify the occurrence of xenon oscillation in a timely manner, and manually intervene or adjust the control strategy based on the feedback information to ensure stable and safe operation of the reactor.

[0081] Dynamic optimization adjustment

[0082] Data acquisition: After the control rod position and boron concentration are adjusted, the system continuously monitors and records the core operating parameters, namely temperature, pressure, neutron flux, coolant flow, and the current position of the control rods and the current value of the boron concentration.

[0083] Data comparison: Compare the feedback data collected in real time with the preset target state and calculate the deviation of various parameters.

[0084] Error identification: Through comparative analysis, the source of deviation is identified to determine whether the xenon oscillation suppression is sufficient.

[0085] Adjustment recommendations: If the xenon oscillations are found to be insufficiently suppressed, the system will recommend further lowering of the control rods or increasing the injection of boron to achieve more stable nuclear chain reaction control.

[0086] System performance evaluation and optimization

[0087] The system calculates the overall performance index of the system by measuring errors multiple times within the evaluation cycle to measure the effectiveness of the adjustment.

[0088] Optimization process: Based on the latest operation data and feedback results, the system regularly performs the following steps:

[0089] Data update: Input the latest collected operating parameters and control status data into the system.

[0090] Process adjustment: According to the performance evaluation results, adjust the control strategy and operation steps, including the convolutional neural network model and the DQN model, to improve the adaptability and stability of the system.

[0091] Continuous improvement: Through multiple cycles, the control process is continuously optimized to ensure the continuous improvement of the xenon oscillation suppression effect.

[0092] Compared with the prior art, the present invention has the following beneficial effects:

[0093] 1. The present invention installs high-precision sensors at key locations of a nuclear reactor, and the system is able to monitor and accurately capture early signals of xenon oscillations in real time. This real-time monitoring function enables the system to quickly identify and predict the occurrence of xenon oscillations before they develop to the point where they may affect the stability of the nuclear reactor.

[0094] 2. The present invention analyzes the collected sensor data in real time. The system can not only accurately predict xenon oscillations, but also prompt the operator to adjust the control strategy in a timely manner, which reduces the monitoring and adjustment burden of the operator and reduces human operation errors caused by fatigue or monitoring errors. The operator can focus more on system monitoring and decision support rather than daily parameter adjustment, thereby improving overall operational efficiency and safety.

[0095] 3. The present invention realizes automatic adjustment prompts for control rod position and boron concentration by integrating the 3AO method with the deep Q network algorithm. Based on real-time data and historical operating experience, the intelligent algorithm dynamically calculates the optimal control strategy and automatically adjusts the operating parameters of the nuclear reactor to effectively suppress xenon oscillations. The closed-loop feedback system in this process further enhances the accuracy and response speed of the control strategy, ensuring the stability of the nuclear reactor under different operating conditions. This stable operating state helps to protect the core components of the nuclear reactor, reduce wear and aging, thereby extending the life of the unit and improving economic benefits. In addition, this stability allows the system to automatically adjust parameters to maintain optimal operating efficiency and safety when responding to changing load conditions, thereby enhancing the reliability and flexibility of the unit under unstable power grid conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0096] Figure 1 is a schematic diagram of the xenon oscillation monitoring interface;

[0097] Figure 2 is a schematic diagram of an early warning prompt;

[0098] FIG3 is a flowchart of the execution of xenon oscillation suppression;

[0099] FIG4 is a flow chart of a method for automatically suppressing xenon oscillations. DETAILED DESCRIPTION

[0100] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part 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 creative work are within the scope of protection of the present invention.

[0101] As shown in FIG1-4, a xenon oscillation automatic suppression auxiliary warning prompt method based on the 3AO method includes the following steps:

[0102] Step 1: monitor the core operating parameters in real time to identify the occurrence of xenon oscillations;

[0103] Step 2: Calculate the optimal control rod and boron concentration adjustment scheme based on the 3AO method algorithm to adapt to the dynamic changes under different operating conditions;

[0104] Step 3, according to the instruction of the control module, the operator is automatically prompted to adjust the control rod position and boron concentration;

[0105] Step 4: Provide real-time feedback on the adjustment effects, analyze the differences between the adjustment results and the expected goals, and dynamically optimize subsequent adjustment strategies.

[0106] The specific process of step 1 above is:

[0107] First, high-precision sensors are installed at key locations of the core to comprehensively monitor the operating status of the core.

[0108] Temperature sensors: Temperature sensors are evenly arranged around the central axis to monitor the temperature changes of the core and identify hot spots and temperature anomalies.

[0109] Pressure sensors: Pressure sensors are installed at the top and bottom of the core, distributed in four quadrants, to monitor coolant pressure and identify pressure fluctuations and anomalies.

[0110] Neutron flux: The RPN power range detector of the extra-core nuclear measurement system measures the distribution and power output of neutrons in the core and identifies power fluctuations.

[0111] Coolant flow sensor: Install flow sensors at the coolant inlet and outlet to monitor the coolant flow and ensure stable operation of the cooling system.

[0112] The collected raw sensor data is denoised, normalized and feature extracted; a moving average filter is used to remove high-frequency noise, and feature extraction is performed after normalization to eliminate the influence of different dimensions. The time domain features of each sensor are extracted: mean, variance, kurtosis, skewness, maximum value, minimum value; frequency domain features: spectrum energy, main frequency and time-frequency domain features: wavelet transform coefficients, to form a comprehensive feature vector. Then the sliding window technology is used to divide the continuous data into multiple overlapping windows, each window As a sample input convolutional neural network model.

[0113]

[0114] in, is the time step of the window length, For the samples, is the time step data.

[0115] The convolutional neural network architecture is as follows:

[0116] Input layer: shape , that is, each sample contains time steps of data, each time step has Sensor data.

[0117] Convolutional layer 1: 64 (3,3) convolution kernels, activation function ReLU;

[0118] Pooling layer 1: max pooling, window size (2,2);

[0119] Convolutional layer 2: 128 (3,3) convolution kernels, activation function ReLU;

[0120] Pooling layer 2: max pooling, window size (2,2);

[0121] Fully connected layer 1: 256 nodes, activation function ReLU;

[0122] Output layer: 1 node, activation function Sigmoid, output probability of xenon oscillation ;

[0123] Loss Function and Optimization Algorithm

[0124] Binary cross entropy is used as the loss function to measure the difference between the model's predicted probability and the actual label:

[0125]

[0126] in, As the binary cross entropy loss function to guide model optimization, is the sample size, is the true label of the sample, 0 means no xenon oscillation, 1 means xenon oscillation occurs, Represents the model for The predicted probability of a sample.

[0127] Adopt Adam optimizer and combine gradient descent method to adaptively adjust learning rate:

[0128]

[0129]

[0130]

[0131]

[0132]

[0133] in, and are the estimates of the first and second order moments, respectively. and are the deviation-corrected forms of the first-order and second-order moments, respectively, represents the gradient, and are the decay rates of the first and second order moments, is the learning rate, ϵ is a small constant to prevent division by zero, is the value of the parameter before updating, The updated value of the parameter.

[0134] After updating the model parameters, the model performance is evaluated on the validation set. The model learning rate is adjusted according to the validation results to improve the model convergence speed and stability. The probability of xenon oscillation is output to predict whether xenon oscillation occurs.

[0135] The specific process of step 2 above is:

[0136] The deep Q network algorithm is selected to be integrated with the 3AO method. The initial control scheme provided by the 3AO method is used, and the DQN algorithm performs autonomous learning and strategy optimization on this basis to calculate the optimal control rod and boron concentration adjustment scheme to adapt to the dynamic changes under different operating conditions, effectively improve the suppression effect of xenon oscillations, and ensure the safety and stability of core operation.

[0137] The fusion process is as follows:

[0138] Action space initialization

[0139] The basic control actions defined by the 3AO method are used as the initial action set of the reinforcement learning agent. On the basis of the 3AO method, the action space of the agent is expanded, more sophisticated control actions are introduced, and the control rod position and boron concentration are slightly adjusted to improve the control accuracy.

[0140]

[0141]

[0142]

[0143] in, Indicates the basic control actions defined by the 3AO method. represents the action extended by reinforcement learning, is the comprehensive action space, They are control rod raising, holding and lowering. The new actions are: control rod rising 0.5%, control rod falling 0.5%, boron concentration increasing 0.1ppm, boron concentration remaining unchanged, and boron concentration decreasing 0.1ppm.

[0144] Strategy initialization

[0145] The initial strategy of the agent prioritizes the actions provided by the 3AO method. , by introducing the ε-greedy strategy, gradually exploring the expansion action ; The hybrid control strategy is:

[0146]

[0147] in, Indicates the specific control action selected for execution. represents a decision rule, is an action selected from the action space A so that in the current state Under this condition, the Q function value is maximized. That is, under the state Next, select the one based on the current value function parameters Evaluate the action that will yield the greatest expected reward;

[0148] Strategy Optimization Process

[0149] Action selection, use - Greedy strategy, chooses random actions part of the time , to facilitate exploration.

[0150]

[0151] The Q network is updated, and the loss function is used to measure the difference between the target network and the main Q network:

[0152]

[0153] in, is the loss function of the DQN algorithm, Used to express the expected value of a variable. For the experience replay buffer, Represents the experience replay buffer A set of randomly sampled transfer samples , is the expectation on the experience replay buffer, It’s an instant reward. is the discount factor, is the current network parameter, are the parameters of the target Q network, Indicates the next state The maximum Q value under Indicates the current status Take action Q value;

[0154] The target network is synchronized, and the main Q network parameters are adjusted every certain number of steps. Copy to the target network: ,

[0155] Experience replay, transfer state Save experience replay buffer , randomly select small batches of samples for training. Through multiple rounds of training, the agent gradually optimizes the control strategy.

[0156] The system receives sensor data in real time, including the core temperature, pressure, neutron power, and coolant flow. The agent learns to choose the best action in different states to maximize the cumulative reward through interaction with the environment. The system adjusts the control rod position and boron concentration according to the best action. The action includes raising, lowering, or maintaining the current position of the control rod and adjusting the specific value of the boron concentration. The feedback module monitors the adjustment effect and dynamically optimizes the control strategy.

[0157] Parameters, the algorithm can ensure the smoothness and continuity of the entire adjustment process.

[0158] The specific process of step 3 above is:

[0159] The control rod position adjustment and boron concentration change calculated by the control module are converted into specific execution instructions through the intelligent control algorithm. After parsing the instructions, the specific control rod position adjustment and boron concentration change requirements are determined. The intelligent algorithm is responsible for strategy optimization and decision-making, while the PID controller is responsible for specific execution and fine-tuning, ensuring the efficient combination of the intelligent algorithm and the execution device, working together to achieve efficient xenon oscillation suppression.

[0160] Execution instruction generation: After receiving the optimal execution plan, the algorithm calculates the specific execution steps and necessary control parameters, and continuously adjusts Parameters, algorithms can ensure the stability and continuity of the entire adjustment process; calculate real-time control instructions :

[0161]

[0162] in, For real-time control instructions, is the systematic error, i.e., the difference between the target position or boron concentration and the current reading; They are proportional, integral and differential coefficients respectively; the PID controller judges and adjusts the position of the control rod and the boron concentration by inputting the current system error e(t), the integral of the error and the differential of the error, that is, the PID controller judges and adjusts the position of the control rod and the boron concentration by inputting the current position of the control rod and the current value of the boron concentration.

[0163] Control rod drive: A high-precision servo motor is used to drive the control rod to ensure high precision and fast response of the control rod position adjustment; the servo motor can achieve fine-tuning according to the instructions of the control module to reduce position error; in addition, the servo motor is also equipped with a position encoder to provide real-time feedback of the control rod position data for monitoring and optimization by the feedback module;

[0164] Boron concentration adjustment: Use intelligent flow control valves to control the mixing and distribution of boron solution to ensure accurate adjustment of boron concentration. Intelligent flow control valves can achieve fast and accurate flow control according to algorithm instructions; the actuator valve is equipped with a flow sensor to monitor the flow and concentration of the boron solution in real time, and feed back to the control module for dynamic adjustment;

[0165] In this method, the PID controller plays a key role in converting the adjustment output by the intelligent algorithm, i.e., the control rod position and the change in boron concentration, into specific execution instructions. Specifically, the PID controller converts the input data into the adjustment amount through the following workflow:

[0166] Data acquisition: The PID controller receives data from sensors, including the current position of the control rod and the current value of the boron concentration. These data are acquired in real time through the high-precision position encoder and boron concentration sensor installed on the control rod to ensure the accuracy and timeliness of the data.

[0167] Error calculation: The PID controller compares the current system state with the preset target state and calculates the system error. The system error includes the deviation of the control rod position and the boron concentration.

[0168] Error processing: The PID controller processes the current value, cumulative value and rate of change of the system error according to the three control links of proportion, integration and differentiation: the proportion P is to make corresponding adjustments immediately according to the size of the current error to reduce the error; the integral I is to accumulate past errors and eliminate long-standing steady-state errors; the differential D is to predict the changing trend of future errors and increase the stability and response speed of the system.

[0169] Adjustment amount calculation: Through the comprehensive effect of data acquisition, error calculation and error processing, the PID controller calculates accurate control instructions , that is, the position adjustment of the control rod and the change in boron concentration. These adjustments ensure that the adjustment of the control rod and boron concentration can accurately respond to changes in the system state and maintain the core operating parameters within the preset target range.

[0170] Execution and feedback: The PID controller sends the calculated control instructions to the servo motor and intelligent flow control valve to perform specific control rod position adjustment and boron concentration adjustment. The real-time feedback data of the actuator is input into the PID controller again to form a closed-loop control to ensure the continuous optimization and stability of the adjustment process.

[0171] The specific process of step 4 above is:

[0172] The specific process of step 4 is as follows:

[0173] After the control adjustment is completed, the system will continue to collect the operating parameters of the reactor core, temperature, pressure, neutron flow and coolant flow, as well as the current position of the control rods and the current value of the boron concentration; these data provide the system with a basis for judging the effect of the adjustment.

[0174] The xenon oscillation monitoring interface provides clear xenon oscillation information feedback to operators by displaying the core's operating parameters and control status in real time. Operators can use this interface to monitor the core's real-time status, identify the occurrence of xenon oscillation in a timely manner, and manually intervene or adjust the control strategy based on the feedback information to ensure stable and safe operation of the reactor.

[0175] Dynamic optimization adjustment

[0176] Data acquisition: After the control rod position and boron concentration are adjusted, the system continuously monitors and records the core operating parameters, namely temperature, pressure, neutron flux, coolant flow, and the current position of the control rods and the current value of the boron concentration.

[0177] Data comparison: Compare the feedback data collected in real time with the preset target state and calculate the deviation of various parameters.

[0178] Error identification: Through comparative analysis, the source of deviation is identified to determine whether the xenon oscillation suppression is sufficient.

[0179] Adjustment recommendations: If the xenon oscillations are found to be insufficiently suppressed, the system will recommend further lowering of the control rods or increasing the injection of boron to achieve more stable nuclear chain reaction control.

[0180] System performance evaluation and optimization

[0181] The system calculates the overall performance index of the system by measuring errors multiple times within the evaluation cycle to measure the effectiveness of the adjustment.

[0182] Optimization process: Based on the latest operation data and feedback results, the system regularly performs the following steps:

[0183] Data update: Input the latest collected operating parameters and control status data into the system.

[0184] Process adjustment: According to the performance evaluation results, adjust the control strategy and operation steps, including the convolutional neural network model and the DQN model, to improve the adaptability and stability of the system.

[0185] Continuous improvement: Through multiple cycles, the control process is continuously optimized to ensure the continuous improvement of the xenon oscillation suppression effect.

[0186] Numerical example:

[0187] Sensor Installation

[0188] Temperature sensor: Platinum resistance temperature detectors are installed at 6 locations around the central axis of the core, one every 60 degrees, to monitor the temperature distribution inside the core.

[0189] Pressure sensor: Four piezoelectric pressure sensors are installed at the top and bottom of the core, distributed in four quadrants, to monitor coolant pressure changes in real time.

[0190] Neutron flux density measurement: An RPN power range detector of the extra-core nuclear measurement system is installed outside the core to monitor the distribution of neutrons and power output in the core.

[0191] Coolant flow sensor: Two vortex flow meters are installed at the coolant inlet and outlet to monitor the changes in coolant flow.

[0192] Data collection and processing

[0193] Sampling frequency: set to 10 times per second (10 Hz) to ensure high-frequency monitoring.

[0194] Denoising: Apply a moving average filter with a window size of N = 5 seconds to smooth the sensor data.

[0195]

[0196] Normalization: Standardize each sensor data and calculate the mean and standard deviation .

[0197] The mean, variance, kurtosis, skewness, maximum value, minimum value, spectrum energy, and wavelet transform coefficient of each sensor are extracted to form a comprehensive feature vector.

[0198] Data segmentation: Using sliding window technology, the window length Time step, step size , generating 10,000 samples for subsequent model training.

[0199]

[0200] CNN model training

[0201] The convolutional neural network architecture is as follows:

[0202] Input layer: Each sample contains time steps of data, each time step has Sensor data.

[0203] Convolutional layer 1: 64 (3,3) convolution kernels, activation function ReLU;

[0204] Pooling layer 1: max pooling, window size (2,2);

[0205] Convolutional layer 2: 128 (3,3) convolution kernels, activation function ReLU;

[0206] Pooling layer 2: max pooling, window size (2,2);

[0207] Fully connected layer 1: 256 nodes, activation function ReLU;

[0208] Output layer: 1 node, activation function Sigmoid, output probability of xenon oscillation ;

[0209] Training parameters

[0210] Loss Function: Binary Cross Entropy

[0211] Optimization algorithm: Adam, learning rate

[0212] Batch size: 64

[0213] Training rounds: 50

[0214] Target network update frequency: update once every 1000 steps

[0215] Model optimization: Regularly update model parameters and retrain with new operating data to predict the probability of xenon oscillations , improve the generalization ability and prediction accuracy of the model.

[0216] CNN output: The trained CNN model outputs the probability of a xenon oscillation for each input sample. , whose value ranges from 0 to 1. This probability is used to determine whether xenon oscillation is likely to occur and serves as the basis for subsequent control strategies.

[0217] Due to the xenon oscillation performance: the axial power deviation AOref exceeds the normal operating range of ±3%FP. Under normal circumstances, the average core temperature is 309.5 degrees Celsius, the pressure is 15.48 MP, and AOref±3 is the safety control range (the coolant flow rate is not an important reference in actual applications). The local temperature rise and pressure increase of the core, and the deviation of AOref beyond ±3 (this indicator is the most important) determine the probability of xenon oscillation: through the local temperature rise and pressure increase of the core, AOref deviates from the normal value by 2%, 2.5%, 3%, 3.5%, and 4%, corresponding to CNN outputs of 0.2, 0.4, 0.6, 0.8, and 1 respectively); Sample A: After the input data is processed by the CNN model, the output , indicating that there is an 85% probability of xenon oscillation. Sample B: After the input data is processed by the CNN model, the output , indicating that there is a 30% probability of xenon oscillation. Sample C: After the input data is processed by the CNN model, the output , indicating that there is a 60% probability of xenon oscillation. Based on these probability values, the system can set a threshold of 50%. When this threshold is exceeded, the control rods and boron concentration are automatically adjusted to suppress the occurrence of xenon oscillations.

[0218] DQN model training

[0219] Environment setup:

[0220] State space (S): Contains 20 sensor data (temperature, pressure, neutron flux, coolant flow) and 5 control variables (4 control rod positions and boron concentration).

[0221] Initial action set: based on 3AO method , each control rod has 3 possible actions (up 1%, hold, down 1%)

[0222] Expanding the action set: Reinforcement learning algorithms add more sophisticated control actions , there are 2 possible actions for the control rod (increase 0.5%, decrease 0.5%), and 3 actions for the boron concentration (increase 0.1 ppm, maintain, decrease 0.1 ppm).

[0223] Comprehensive action space (A): , boron concentration has 3 actions (increase 0.1 ppm, maintain, decrease 0.1 ppm). The total number of actions is .

[0224] Reward function (R): Comprehensively consider power fluctuations, safety parameters, and control energy consumption, and set different weight coefficients.

[0225] Reward mechanism: The reward function takes into account power fluctuations and safety parameters (axial power deviation AOref exceeds the normal operating range of ±3% FP) and sets different weight coefficients to guide the agent to optimize the control strategy.

[0226] Network structure:

[0227] Input layer: shape (20), i.e. 20 sensor data.

[0228] Hidden layer: two fully connected layers, with 128 and 64 nodes, and ReLU activation function.

[0229] Output layer: 6561 nodes, corresponding to the Q value of each action in the action space.

[0230] Training parameters:

[0231] Learning rate (η): 0.001

[0232] Discount factor (γ): 0.99

[0233] Experience replay buffer size: 100,000

[0234] Batch size: 64

[0235] Target network update frequency: update once every 1000 steps

[0236] Exploration strategy: ε-greedy strategy, initial =1.0, and gradually decrease to 0.1.

[0237] Algorithm fusion and deployment: The 3AO method integrates the DQN algorithm and deploys the trained model to the real-time monitoring system, which receives sensor data in real time, outputs the optimal control action, i.e., the adjustment of the control rod and the boron concentration, and monitors the control effect and system stability. The output control action is as follows:

[0238] (1) Under normal operation: When the system runs stably and the xenon oscillation probability output by the CNN model is lower than the set threshold, the DQN model will choose to maintain the current control rod position and boron concentration without making any adjustments.

[0239] (2) Responding to xenon oscillation: Based on the prediction results output by CNN, the predicted xenon oscillation threshold is set to 50%. If the predicted xenon oscillation probability exceeds 50%, the adjustment is automatically performed.

[0240] Control rod adjustment strategy: The control rod is lowered by 0.5% to suppress the reaction rate and mitigate the impact of xenon oscillations.

[0241] Boron concentration adjustment strategy: The boron concentration is increased by 0.1 ppm to absorb excess neutrons and further stabilize the nuclear reaction.

[0242] (3) Routine maintenance adjustment: During routine maintenance checks, the system detected that the position of the control rods deviated slightly from the preset optimal position.

[0243] Control rod adjustment strategy: Automatically fine-tune to return to the optimal position, with an adjustment range of .

[0244] Boron concentration adjustment strategy: Boron concentration adjustment .

[0245] (4) Rapid response to abnormal situations: The sensor suddenly detects an abnormally high neutron flux, indicating a potential safety risk.

[0246] Control rod adjustment strategy: Immediately reduce the reactor power, and the system quickly and continuously adjusts the control rods down by 0.5%.

[0247] Boron concentration adjustment strategy: To quickly stabilize the reactor, the boron concentration is increased systematically .

[0248] Adjustment strategy

[0249] a. Data acquisition: The PID controller receives data from sensors, including the current position of the control rod and the current value of the boron concentration. These data are acquired in real time through the high-precision position encoder and boron concentration sensor installed on the control rod to ensure the accuracy and timeliness of the data.

[0250] b Error calculation: The PID controller compares the current system state with the preset target state and calculates the system error. The system error includes the deviation of the control rod position and the boron concentration.

[0251] c. Error handling:

[0252] Proportional (P): According to the current error size, make corresponding adjustments immediately to reduce the error.

[0253] Integration (I): Accumulates past errors and eliminates long-standing steady-state errors.

[0254] Differentiation (D): Predict future error trends and increase system stability and response speed.

[0255] d. Calculation of adjustment amount: Through the combined effect of the above three links, the PID controller calculates precise control instructions, namely the position adjustment amount of the control rod and the change in boron concentration. These adjustment amounts ensure that the adjustment of the control rod and boron concentration can accurately respond to changes in the system state and maintain the core operating parameters within the preset target range.

[0256] e. Execution and feedback: The PID controller sends the calculated control instructions to the servo motor and intelligent flow control valve to perform specific control rod position adjustment and boron concentration adjustment. The real-time feedback data of the actuator is input into the PID controller again to form a closed-loop control to ensure the continuous optimization and stability of the adjustment process.

[0257] Collaboration between intelligent algorithms and PID controllers: The 3AO method integrates the DQN algorithm to optimize strategies and make decisions, providing preliminary control instructions (control rod position adjustment and boron concentration change). The PID controller fine-tunes these instructions based on real-time feedback data to ensure the accuracy and stability of the adjustment process. The two work together to achieve efficient suppression of xenon oscillations and ensure the safe and stable operation of the nuclear reactor.

[0258] Dynamic optimization adjustment

[0259] Error calculation: For the measured value of the temperature sensor in the input parameters, the error between it and the target is calculated. Since the xenon oscillation cannot be completely suppressed, the system will analyze this error and recommend further adjustment of the control rods.

[0260] Performance evaluation: , Close to 1, this means that the system control accuracy is very high, the error is very small, and it almost reaches the ideal state. The evaluation value is between 0.8 and 0.9, which shows that the control effect of the system is good, the error is small, and most safety and efficiency requirements are met. The evaluation value is between 0.7 and 0.8. Although the control effect is not ideal, it is still within an acceptable range. The system needs further optimization to improve performance. The evaluation value is lower than 0.7, which usually means that the control strategy of the system has a large room for improvement, the error is large, and the control logic or parameter adjustment may need to be reconsidered.

[0261] The control algorithm is updated regularly based on the latest collected operating data and feedback results. The parameters of the convolutional neural network and DQN model are adjusted to improve the prediction accuracy and control effect.

[0262] Finally, it should be noted that in the present invention, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0263] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention is described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A xenon oscillation automatic suppression auxiliary warning prompt method based on 3AO method, characterized in that: The following steps are involved: Step 1: monitor the core operating parameters in real time and use a convolutional neural network to identify the occurrence of xenon oscillations; Step 2: Based on the 3AO algorithm, calculate the optimal control rod and boron concentration adjustment scheme to adapt to the dynamic changes under different operating conditions; Step 3: According to the instruction of the control module, an automatic warning is issued to prompt the operator to adjust the control rod position and boron concentration; Step 4: Provide real-time feedback on the adjustment effects, analyze the differences between the adjustment results and the expected goals, and dynamically optimize subsequent adjustment strategies.

2. According to claim 1, a xenon oscillation automatic suppression auxiliary warning prompt method based on the 3AO method is characterized in that: The specific process of step 1 is: When xenon oscillation occurs, it will cause the local temperature and power of the core to rise, which will then cause the fuel elements to melt. First, high-precision sensors are installed at key locations in the core to fully monitor the operating status of the core. Temperature sensors: Temperature sensors are evenly arranged around the central axis to monitor the temperature changes of the core and identify hot spots and temperature anomalies; Pressure sensors: Pressure sensors are installed at the top and bottom of the core, distributed in four quadrants, to monitor coolant pressure and identify pressure fluctuations and anomalies; Neutron flux: RPN power range detector of the out-of-core nuclear measurement system; Coolant flow sensor: Install flow sensors at the coolant inlet and outlet to monitor the coolant flow and ensure the stable operation of the cooling system; The collected raw sensor data is denoised, normalized and feature extracted; a moving average filter is used to remove high-frequency noise, and feature extraction is performed after normalization to eliminate the influence of different dimensions, and the time domain characteristics of each sensor are extracted: mean, variance, kurtosis, skewness, maximum value, and minimum value; Frequency domain features: spectrum energy, main frequency and time-frequency domain features: wavelet transform coefficients, forming a comprehensive feature vector; using sliding window technology to divide continuous data into multiple overlapping windows, each window As a sample input convolutional neural network model; in, is the time step of the window length, For the samples, is the time step data; The convolutional neural network architecture is as follows: Input layer: shape , that is, each sample contains time steps of data, each time step has Sensor data; Convolutional layer 1: 64 (3,3) convolution kernels, activation function ReLU; Pooling layer 1: max pooling, window size (2,2); Convolutional layer 2: 128 (3,3) convolution kernels, activation function ReLU; Pooling layer 2: max pooling, window size (2,2); Fully connected layer 1: 256 nodes, activation function ReLU; Output layer: 1 node, activation function Sigmoid, output probability of xenon oscillation ; Loss Function and Optimization Algorithm Binary cross entropy is used as the loss function to measure the difference between the model's predicted probability and the actual label: in, As the binary cross entropy loss function to guide model optimization, is the sample size, is the true label of the sample, 0 means no xenon oscillation, 1 means xenon oscillation occurs, Represents the model for The predicted probability of samples; Adopt Adam optimizer and combine gradient descent method to adaptively adjust learning rate: in, and are the estimates of the first and second order moments, respectively. and are the deviation-corrected forms of the first-order and second-order moments, respectively, represents the gradient, and are the decay rates of the first and second order moments, is the learning rate, ϵ is a small constant to prevent division by zero, is the value of the parameter before updating, is the updated value of the parameter; After updating the model parameters, the model performance is evaluated on the validation set. The model learning rate is adjusted according to the validation results to improve the model convergence speed and stability. A threshold is set for the real-time monitoring data. The probability of output xenon oscillation is predicted through a convolutional neural network to predict whether xenon oscillation will occur.

3. The xenon oscillation automatic suppression auxiliary warning prompt method based on the 3AO method according to claim 1 is characterized in that: The specific process of step 2 is: The deep Q network algorithm is selected to be integrated with the 3AO method. The DQN algorithm uses the initial control scheme provided by the 3AO method to perform autonomous learning and strategy optimization on this basis, and calculates the optimal control rod and boron concentration adjustment scheme to adapt to the dynamic changes under different operating conditions, effectively improve the suppression effect of xenon oscillation, and ensure the safety and stability of core operation. The fusion process is as follows: Action space initialization The basic control actions defined by the 3AO method are used as the initial action set of the reinforcement learning agent. On the basis of the 3AO method, the action space of the agent is expanded, more sophisticated control actions are introduced, and the control rod position and boron concentration are finely adjusted to improve the control accuracy. in, Indicates the basic control actions defined by the 3AO method. represents the action extended by reinforcement learning, is the comprehensive action space, They are control rod raising, holding and lowering. They are newly added actions; Strategy initialization The initial strategy of the agent prioritizes the actions provided by the 3AO method. , by introducing the ε-greedy strategy, gradually exploring the expansion action ; The hybrid control strategy is: in, Indicates the specific control action selected for execution. represents a decision rule, is an action selected from the action space A so that in the current state Under this condition, the Q function value is maximized, that is, in the state Next, select the one based on the current value function parameters Evaluate the action that will yield the greatest expected reward; Strategy Optimization Process Action selection, use - Greedy strategy, chooses random actions part of the time , to promote exploration; The Q network is updated, and the loss function is used to measure the difference between the target network and the main Q network: in, is the loss function of the DQN algorithm, Used to express the expected value of a variable. For the experience replay buffer, Represents the experience replay buffer A set of randomly sampled transfer samples , is the expectation on the experience replay buffer, It’s an instant reward. is the discount factor, is the current network parameter, are the parameters of the target Q network, Indicates the next state The maximum Q value under Indicates the current status Take action Q value; The target network is synchronized, and the main Q network parameters are adjusted every certain number of steps. Copy to the target network: , Experience replay, transfer state Save experience replay buffer , randomly select small batches of samples for training; through multiple rounds of training, the agent gradually optimizes the control strategy; The system receives sensor data in real time, including the core's temperature, pressure, neutron power and coolant flow rate; the intelligent agent learns to choose the optimal action in different states through interaction with the environment to maximize the cumulative reward, and the system adjusts the control rod position and boron concentration according to the optimal action; the action includes raising, lowering or maintaining the current position of the control rod and adjusting the specific value of the boron concentration; the adjustment effect is monitored through the feedback module, and the control strategy is dynamically optimized.

4. The xenon oscillation automatic suppression auxiliary warning prompt method based on the 3AO method according to claim 3 is characterized in that: They are: control rod rises 0.5%, control rod drops 0.5%, boron concentration increases 0.1ppm, boron concentration remains constant, and boron concentration decreases 0.1ppm.

5. The xenon oscillation automatic suppression auxiliary warning prompt method based on the 3AO method according to claim 3 is characterized in that: The control actions output by the agent's step-by-step optimization control strategy are as follows: 1) Under normal operation: When the system runs stably and the xenon oscillation probability output by the convolutional neural network CNN model is lower than the set threshold, the DQN model will choose to maintain the current control rod position and boron concentration without adjustment; 2) Responding to xenon oscillation: Based on the prediction results output by CNN, the predicted xenon oscillation threshold is set to 50%. If the predicted xenon oscillation probability exceeds 50%, the adjustment is automatically performed; Control rod adjustment strategy: The control rod is lowered by 0.5% to suppress the reaction rate and slow down the impact of xenon oscillation. Boron concentration adjustment strategy: The boron concentration is increased by 0.1ppm to absorb excess neutrons and further stabilize the nuclear reaction.

6. The xenon oscillation automatic suppression auxiliary warning prompt method based on the 3AO method according to claim 1 is characterized in that: The specific process of step 3 is as follows: The control rod position adjustment and boron concentration change calculated by the control module are converted into specific execution instructions through the intelligent control algorithm. After parsing the instructions, the specific control rod position adjustment and boron concentration change requirements are determined. The intelligent algorithm is responsible for strategy optimization and decision-making, while the PID controller is responsible for specific execution and fine adjustment, ensuring the efficient combination of the intelligent algorithm and the execution device, working together to achieve efficient xenon oscillation suppression. Execution instruction generation: After receiving the optimal execution plan, the algorithm calculates the specific execution steps and necessary control parameters, and continuously adjusts Parameters,the algorithm can ensure the stability and continuity of the entire adjustment process; Calculate real-time control instructions : in, For real-time control instructions, is the systematic error, i.e., the difference between the target position or boron concentration and the current reading; are proportional, integral and differential coefficients respectively; the PID controller judges and adjusts the position of the control rod and the boron concentration by inputting the current system error e(t), the integral of the error and the differential of the error, that is, the PID controller judges and adjusts the position of the control rod and the boron concentration by inputting the current position of the control rod and the current value of the boron concentration; Control rod drive: A high-precision servo motor is used to drive the control rod to ensure high precision and fast response of the control rod position adjustment; the servo motor can achieve fine-tuning according to the instructions of the control module to reduce position error; in addition, the servo motor is also equipped with a position encoder to provide real-time feedback of the control rod position data for monitoring and optimization by the feedback module; Boron concentration adjustment: Use intelligent flow control valve to control the mixing and distribution of boron solution to ensure accurate adjustment of boron concentration. Intelligent flow control valve can achieve fast and accurate flow control according to algorithm instructions; the actuator valve is equipped with a flow sensor to monitor the flow and concentration of the boron solution in real time, and feed back to the control module for dynamic adjustment.

7. The xenon oscillation automatic suppression auxiliary warning prompt method based on the 3AO method according to claim 6 is characterized in that: The PID controller converts input data into regulation quantity through the following workflow: Data acquisition: The PID controller receives data from sensors, including the current position of the stack control rod and the current value of the boron concentration. These data are acquired in real time through the high-precision position encoder and boron concentration sensor installed on the control rod to ensure the accuracy and timeliness of the data. Error calculation: The PID controller compares the current system state with the preset target state and calculates the system error; the system error includes the deviation of the control rod position and the boron concentration; Error processing: PID controller processes the current value, cumulative value and change rate of system error according to the three control links of proportion, integration and differentiation: proportion P is to make corresponding adjustments immediately according to the size of the current error to reduce the error; integration I is to accumulate past errors and eliminate long-standing steady-state errors; differentiation D is to predict the changing trend of future errors and increase the stability and response speed of the system; Adjustment amount calculation: Through the comprehensive effect of data acquisition, error calculation and error processing, the PID controller calculates accurate control instructions , i.e., the control rod position adjustment and the change in boron concentration; These adjustments ensure that the control rod and boron concentration adjustments can accurately respond to changes in system status and maintain core operating parameters within the preset target range; Execution and feedback: The PID controller sends the calculated control instructions to the servo motor and intelligent flow control valve to execute specific control rod position adjustment and boron concentration adjustment; the real-time feedback data of the actuator is input into the PID controller again to form a closed-loop control to ensure the continuous optimization and stability of the adjustment process.

8. The xenon oscillation automatic suppression auxiliary warning prompt method based on the 3AO method according to claim 1 is characterized in that: The specific process of step 4 is as follows: After the control adjustment is completed, the system will continue to collect the operating parameters of the reactor core, temperature, pressure, neutron flow and coolant flow, as well as the current position of the control rods and the current value of the boron concentration; these data provide the system with a basis for judging the effect of the adjustment; The xenon oscillation monitoring interface provides clear xenon oscillation information feedback to operators by displaying the core's operating parameters and control status in real time. Operators can use this interface to monitor the core's real-time status, promptly identify the occurrence of xenon oscillation, and manually intervene or adjust the control strategy based on the feedback information to ensure stable and safe operation of the reactor. Dynamic optimization adjustment Data acquisition: After the control rod position and boron concentration are adjusted, the system continuously monitors and records the core operating parameters, namely, temperature, pressure, neutron flux, coolant flux, and the current position of the control rods and the current value of the boron concentration; Data comparison: Compare the feedback data collected in real time with the preset target state and calculate the deviation of various parameters; Error identification: Through comparative analysis, the source of deviation is identified to determine whether the xenon oscillation suppression is sufficient; Adjustment suggestions: If it is found that the xenon oscillation suppression is insufficient, the system will recommend further lowering of the control rods or increasing the injection of boron to achieve more stable nuclear chain reaction control; System performance evaluation and optimization The system calculates the overall performance index of the system by measuring errors multiple times within the evaluation cycle to measure the effectiveness of the adjustment; Optimization process: Based on the latest operation data and feedback results, the system regularly performs the following steps: Data update: input the latest collected operating parameters and control status data into the system; Process adjustment: According to the performance evaluation results, adjust the control strategy and operation steps, including the convolutional neural network model and the DQN model, to improve the adaptability and stability of the system; Continuous improvement: Through multiple cycles, the control process is continuously optimized to ensure the continuous improvement of the xenon oscillation suppression effect.

Images