Intelligent electron accelerator for industrial irradiation

Abstract

The invention discloses an intelligent electron accelerator for industrial irradiation, which comprises a conveying unit for conveying a material to be irradiated into an irradiation chamber, a pre-irradiation detection unit for performing pre-irradiation detection on the material and recording initial parameters of the material, and a calibration unit for adjusting calibration parameters of the accelerator based on a pre-irradiation detection result. The accelerator unit starts an accelerator after the calibration parameters are adjusted to accurately irradiate the material, and the maintenance calibration unit performs parameter calibration again after maintenance to ensure that the precision deviation of the accelerator is within an allowable range; the intelligent electron accelerator for industrial irradiation solves the problems that errors easily exist in accurate operation in the prior art, especially in complex parameter setting, due to improper manual operation or insufficient system detection precision, energy and distribution of electron beams deviate, the precision deviation not only affects the effect of the irradiation process, but also improves the accuracy of the irradiation process. And the quality of a final product is unfavorably influenced, and even potential production safety hazards are caused.

Description

Technical Field

[0001] The invention relates to the field of nuclear technology application, and in particular to an intelligent electron accelerator for industrial irradiation. Background Art

[0002] In the field of industrial irradiation, electron accelerators are commonly used equipment, widely used in many fields such as material modification, food disinfection, and sterilization of medical supplies. This type of equipment accelerates electrons to form a high-energy electron beam, which processes the target substance to achieve the desired modification, sterilization or disinfection effect. With the development of industrial automation and intelligent technology, intelligent electron accelerators are gradually introduced to achieve precise control of the stability and accuracy of the electron beam, improve production efficiency and ensure product quality.

[0003] The intelligent electron accelerator systems currently on the market usually use automated control technology, which can automatically adjust and optimize the distribution of electron beam energy according to the set parameters. Through the intelligent system, operators can simplify the operating process, reduce manual intervention, and improve the operating efficiency and stability of equipment during the production process. However, although this technology has many advantages, there are still some problems in practical applications.

[0004] First of all, after a long period of operation, the equipment needs regular maintenance and parameter calibration. During this process, the calibration operation is prone to errors, especially in complex parameter settings. Due to improper human operation or insufficient system detection accuracy, the energy and distribution of the electron beam deviate. This deviation in accuracy will not only affect the effect of the irradiation process, but also have an adverse effect on the quality of the final product, and even cause production safety hazards. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent electron accelerator for industrial irradiation to solve the problem that the prior art is prone to errors in accurate operation, especially in complex parameter settings, due to improper human operation or insufficient system detection accuracy, resulting in deviations in the energy and distribution of the electron beam. This accuracy deviation not only affects the effect of the irradiation process, but also has an adverse effect on the quality of the final product and even causes production safety hazards.

[0006] To achieve the above object, the present invention provides the following technical solution: an intelligent electron accelerator for industrial irradiation, comprising:

[0007] A conveying unit, used for conveying the material to be irradiated into the irradiation chamber;

[0008] The pre-irradiation detection unit connected to the conveying unit is used to perform pre-irradiation detection on the material and record the initial parameters of the material.

[0009] Specifically, by collecting data on the initial parameters of the material and the corresponding irradiation effects, a multivariate linear regression model that can predict the irradiation effect is trained based on historical data. Based on the initial parameters of the material, the model is used to predict whether the effect after irradiation meets the standards, and the initial parameters of the material are transmitted to the accelerator unit to assist in subsequent irradiation control;

[0010] A calibration unit connected to the pre-irradiation detection unit is used to adjust the calibration parameters of the accelerator based on the pre-irradiation detection result, specifically to initialize the state estimation and error covariance matrix of the Kalman filter, use the result of the pre-irradiation detection unit as observation data, use the prediction formula of the Kalman filter to estimate the current calibration parameters according to the previous state, and use the Kalman gain to update the calibration parameters based on the difference between the measured value and the predicted value;

[0011] An accelerator unit connected to the calibration unit is used to start the accelerator after the calibration parameters are adjusted to accurately irradiate the material;

[0012] The maintenance calibration unit connected to the accelerator unit is used to perform parameter calibration again after maintenance to ensure that the accuracy deviation of the accelerator is within the allowable range. Specifically, the value range of the calibration parameters of the accelerator is set according to the initial state after maintenance, and a set of random calibration parameters is generated as the initial population. The calibration accuracy of each set of calibration parameters is evaluated by the fitness function, and an excellent parameter combination is selected according to the fitness. Crossover and mutation are performed to generate a new parameter combination. After multiple iterations, the calibration parameter combination with the highest accuracy is finally obtained.

[0013] Preferably, the conveying unit conveys the material to be irradiated into the irradiation chamber, comprising:

[0014] Establish a map of the working area of ​​the conveyor unit, set the starting point and target point of material conveyance, convey the material from the loading point to the irradiation chamber, and apply the A-star algorithm based on heuristic search to optimize the material conveyance path through the following evaluation function:

[0015] f(n)=g(n)+h(n),

[0016] Among them, f(n) represents the total cost of node n, g(n) represents the actual cost from the starting point to node n, h(n) represents the heuristic estimated cost from node n to the target node, and n represents a node in the path search.

[0017] Preferably, the pre-irradiation detection unit performs pre-irradiation detection on the material and records the initial parameters of the material further comprising:

[0018] The specific formula for multivariate linear regression is:

[0019] y=β 0 +β 1x 1 +β 2 x 2 +…+β n x n +∈,

[0020] Among them, y represents the expected irradiation result of the material, x 1 , x 2 , …, x n Indicates different initial parameters of the material during pre-irradiation testing, β 1 , β 2 , …, β n Represents the regression coefficient to be trained, and ∈ represents the error term.

[0021] Preferably, the calibration unit adjusts the calibration parameters of the accelerator based on the pre-irradiation detection result and further includes:

[0022] The prediction steps of the Kalman filter are:

[0023]

[0024] in, represents the estimated value of the predicted state at the kth moment, represents the updated state estimate at the kth moment, A represents the state transfer matrix, B represents the control input matrix, and u k Represents the control input value, P k|k-1 represents the prediction error covariance matrix, P k|k represents the updated error covariance matrix, Q represents the process noise covariance matrix;

[0025] The formula for the update step is:

[0026] K k =P k|k1 H T (HP k|k1 H T +R) -1 ,

[0027]

[0028] P k|k =(IK k H)P k|k-1 ,

[0029] Among them, K k represents the Kalman gain, P k|k-1 represents the prediction error covariance matrix, z k represents the measurement value at the kth moment, H represents the observation matrix, R represents the measurement noise covariance matrix, represents the updated state estimate at the kth moment, represents the predicted state estimate at the kth moment, and I represents the identity matrix.

[0030] Preferably, the accelerator unit starts the accelerator after the calibration parameters are adjusted, and accurately irradiates the material, including:

[0031] According to the calibration results, the target electron beam energy and distribution of the accelerator are set, and the electron beam energy and distribution of the accelerator are monitored in real time. By comparing the target value and the actual value, the error signal e(t) is obtained. According to the error value, the PID control algorithm is used to adjust the electron beam energy of the accelerator. The specific formula is:

[0032]

[0033] Among them, u(t) represents the control signal, e(t) represents the error value, K p , K i , K d Represents proportional, integral, and differential coefficients.

[0034] Preferably, the calibration unit further comprises:

[0035] Obtaining initial parameters for pre-irradiation testing;

[0036] Analyze the deviation between initial parameters and historical standard parameters;

[0037] Based on the deviation, calculating an adjustment amount for the calibration parameter;

[0038] Adjust the calibration parameters of the accelerator;

[0039] The adjustment amount of the calculated calibration parameter is specifically:

[0040] Calculate the deviation D between the initial parameters P of the pre-irradiation test and the historical standard parameters S, the formula is D = | PS |;

[0041] Wherein, P represents the initial parameters of pre-irradiation detection, S represents the historical standard parameters, and D represents the deviation;

[0042] Determine whether the deviation D exceeds a predetermined threshold T;

[0043] If the deviation D is greater than the threshold T, the calibration parameters need to be adjusted; otherwise, the calibration parameters do not need to be adjusted.

[0044] Preferably, if the deviation D is greater than a threshold value T, the calibration parameters need to be adjusted, including:

[0045] If D>T, proceed to the next step;

[0046] Calculate the adjustment amount A of the calibration parameter using the formula A = k × D;

[0047] Among them, k represents the proportionality coefficient, and D represents the deviation;

[0048] Based on the calculated adjustment amount A, the calibration parameters of the accelerator are updated.

[0049] Preferably, updating the calibration parameters of the accelerator based on the calculated adjustment amount A includes:

[0050] Get the current calibration parameter C;

[0051] Calculate the new calibration parameter N, the formula is N = C + A;

[0052] Where C represents the current calibration parameter and A represents the adjustment amount;

[0053] Based on the calculated new calibration parameter N, the calibration parameter of the accelerator is updated.

[0054] Preferably, updating the calibration parameters of the accelerator based on the calculated new calibration parameter N includes:

[0055] After each update of the calibration parameters, the new calibration parameters N and the current time T are recorded;

[0056] Determine whether the time interval ΔT between two updates is less than the predetermined time threshold t, the formula is ΔT <t;

[0057] Where ΔT represents the actual time interval between two updates, and t represents the predetermined time threshold;

[0058] If ΔT is less than t, continue monitoring, otherwise complete the update process.

[0059] It can be seen from the above technical solution that the present invention has the following beneficial effects:

[0060] The intelligent electron accelerator for industrial irradiation transmits the material to be irradiated to the irradiation room through the transmission unit, the pre-irradiation detection unit performs pre-irradiation detection on the material and records the initial parameters of the material, the calibration unit adjusts the calibration parameters of the accelerator based on the pre-irradiation detection result, the accelerator unit starts the accelerator after the calibration parameters are adjusted, and accurately irradiates the material, the maintenance calibration unit is used to perform parameter calibration again after maintenance to ensure that the accuracy deviation of the accelerator is within the allowable range, and can automatically detect and correct the energy and distribution of the electron beam, avoid the accuracy deviation caused by human operation, improve the stability and safety of the equipment, effectively reduce the maintenance frequency and ensure the quality consistency of the product, significantly reduce the need for human intervention, improve production efficiency, and reduce the safety hazards caused by human operation errors, and solve the problem that the existing technology is prone to errors in accurate operation, especially in complex parameter settings, due to improper human operation or insufficient system detection accuracy, the energy and distribution of the electron beam will deviate, this accuracy deviation will not only affect the effect of the irradiation process, but also have an adverse effect on the quality of the final product, and even cause production safety hazards. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a schematic diagram of the connection of each unit of the present invention. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. 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.

[0063] like Figure 1 As shown, an intelligent electron accelerator for industrial irradiation comprises:

[0064] A conveying unit, used for conveying the material to be irradiated into the irradiation chamber;

[0065] The pre-irradiation detection unit connected to the conveying unit is used to perform pre-irradiation detection on the material and record the initial parameters of the material.

[0066] Specifically, by collecting data on the initial parameters of the material and the corresponding irradiation effects, a multivariate linear regression model that can predict the irradiation effect is trained based on historical data. Based on the initial parameters of the material, the model is used to predict whether the effect after irradiation meets the standards, and the initial parameters of the material are transmitted to the accelerator unit to assist in subsequent irradiation control;

[0067] A calibration unit connected to the pre-irradiation detection unit is used to adjust the calibration parameters of the accelerator based on the pre-irradiation detection result, specifically to initialize the state estimation and error covariance matrix of the Kalman filter, use the result of the pre-irradiation detection unit as observation data, use the prediction formula of the Kalman filter to estimate the current calibration parameters according to the previous state, and use the Kalman gain to update the calibration parameters based on the difference between the measured value and the predicted value;

[0068] An accelerator unit connected to the calibration unit is used to start the accelerator after the calibration parameters are adjusted to accurately irradiate the material;

[0069] The maintenance calibration unit connected to the accelerator unit is used to perform parameter calibration again after maintenance to ensure that the accuracy deviation of the accelerator is within the allowable range. Specifically, the value range of the calibration parameters of the accelerator is set according to the initial state after maintenance, and a set of random calibration parameters is generated as the initial population. The calibration accuracy of each set of calibration parameters is evaluated by the fitness function, and an excellent parameter combination is selected according to the fitness. Crossover and mutation are performed to generate a new parameter combination. After multiple iterations, the calibration parameter combination with the highest accuracy is finally obtained.

[0070] After the material is delivered to the irradiation chamber by the conveying unit, the pre-irradiation detection unit detects the initial parameters of the material and predicts whether the irradiation effect meets the standard through a multivariate linear regression model. If the prediction result shows that the irradiation effect may not meet the requirements, the system will adjust the operating parameters of the accelerator in time. Based on the pre-irradiation detection, the calibration unit updates the calibration parameters of the accelerator through the Kalman filter technology to ensure the precise control of the accelerator. After the irradiation is completed, during equipment maintenance, the maintenance calibration unit will optimize the calibration parameters through genetic algorithms to ensure the long-term accuracy of the accelerator. The irradiation effect of the material is predicted through pre-irradiation detection and linear regression models, and the effect can be evaluated before irradiation, which greatly improves the accuracy and reliability of irradiation control and reduces the production of defective products. The calibration unit adopts the Kalman filter algorithm, which can adjust the calibration parameters of the accelerator according to real-time data to ensure that the accelerator always maintains high precision under different materials and operating conditions. The maintenance calibration unit optimizes the calibration parameters through genetic algorithms to avoid errors in manual calibration and improve the irradiation accuracy after maintenance. The automated calibration process is efficient and reliable, and the entire system is highly intelligent and can automatically adjust the irradiation parameters according to the characteristics of different materials, improve production efficiency, and reduce the burden on operators.

[0071] The conveying unit conveys the material to be irradiated into the irradiation room and includes:

[0072] Establish a map of the working area of ​​the conveyor unit, set the starting point and target point of material conveyance, convey the material from the loading point to the irradiation chamber, and apply the A-star algorithm based on heuristic search to optimize the material conveyance path through the following evaluation function:

[0073] f(n)=g(n)+h(n),

[0074] Among them, f(n) represents the total cost of node n, g(n) represents the actual cost from the starting point to node n, h(n) represents the heuristic estimated cost from node n to the target node, and n represents a node in the path search.

[0075] In this embodiment, the transmission unit first models the material transmission area of ​​the entire factory, generates a map and sets the starting point and target point of the material. The transmission system uses the A-star algorithm for path search. The A-star algorithm combines the cost of the actual path and heuristic estimation to plan the material transmission path as the best. The algorithm evaluates each path node by calculating the evaluation function f(n) at each step to ensure that the material can be safely and quickly delivered to the irradiation room along the optimal path. By using the A-star algorithm, the system can plan the optimal material transmission path in real time to ensure that the material is delivered to the destination in the shortest time and with the lowest energy consumption in a complex factory environment. The A-star algorithm can dynamically avoid obstacles and congestion in the working area, thereby improving the transmission efficiency and reducing the delay in material transmission. Through accurate evaluation function calculation and sensor data feedback, the transmission system can achieve accurate control of the material transmission process and reduce errors and deviations caused by improper path selection. This embodiment is highly intelligent and can autonomously plan the path, and adjust it according to the real-time situation during the transmission process, reducing human intervention and reducing the workload of operators.

[0076] The pre-irradiation detection unit performs pre-irradiation detection on the material and records the initial parameters of the material, including:

[0077] The specific formula for multivariate linear regression is: y = β 0 +β 1 x 1 +β 2 x 2 +...+β n x n +∈,

[0078] Among them, y represents the expected irradiation result of the material, x 1 , x 2 , ..., x n Indicates different initial parameters of the material during pre-irradiation testing, β 1 , β 2 , ..., β n Represents the regression coefficient to be trained, and ∈ represents the error term.

[0079] The pre-irradiation detection unit of the present invention predicts the irradiation effect of the material through a multivariate linear regression model. By collecting a large amount of historical data, the model can identify the relationship between the initial parameters of the material and its irradiation effect. The model uses the regression coefficient to predict the irradiation effect of the material, so that it can evaluate whether the material meets the irradiation requirements before the irradiation operation. The data after the pre-irradiation detection will be transmitted to the accelerator unit to help further optimize the irradiation parameters. The multivariate linear regression model can accurately predict the irradiation effect of the material through the training of historical data, reducing the uncertainty in the experiment. The pre-irradiation detection unit can obtain the initial parameters of the material in real time and quickly predict the irradiation effect, so that the entire system can operate efficiently. Through the real-time prediction results, the system can adjust the irradiation parameters of the accelerator before the actual irradiation operation, avoiding the occurrence of irradiation failures, improving production efficiency and product quality. The model can accumulate more data over time and retrain, thereby gradually improving the prediction ability of the system, so that it can cope with changes in different materials and environments.

[0080] The calibration unit adjusts the calibration parameters of the accelerator based on the pre-irradiation detection result and further includes:

[0081] The prediction steps of the Kalman filter are:

[0082]

[0083] in, represents the estimated value of the predicted state at the kth moment, represents the updated state estimate at the kth moment, A represents the state transfer matrix, B represents the control input matrix, and u k Represents the control input value, P k|k-1 represents the prediction error covariance matrix, P k|k represents the updated error covariance matrix, Q represents the process noise covariance matrix;

[0084] The formula for the update step is:

[0085] K k =P k|k-1 H T (HP k|k-1 H T +R) -1 ,

[0086]

[0087] P k|k =(IK k H)P k|k-1 ,

[0088] Among them, K k represents the Kalman gain, Pk|k-1 represents the prediction error covariance matrix, z k represents the measurement value at the kth moment, H represents the observation matrix, R represents the measurement noise covariance matrix, represents the updated state estimate at the kth moment, represents the predicted state estimate at the kth moment, and I represents the identity matrix.

[0089] In this embodiment, the Kalman filter is used to adjust the calibration parameters of the accelerator in real time. First, the system predicts the state of the accelerator according to the preset state transfer equation to obtain the predicted calibration parameters and error covariance matrix. Subsequently, the system obtains the actual observation data of the material through the pre-irradiation detection unit, and adjusts the predicted calibration parameters in combination with the Kalman gain. The updated calibration parameters ensure the accuracy of the accelerator, so that the irradiation effect can meet the expected standards. The Kalman filter can dynamically adjust the calibration parameters of the accelerator in the presence of noise, providing higher accuracy and ensuring the stability of the accelerator during the irradiation process. The algorithm has real-time adjustment capabilities and can quickly update the calibration parameters of the accelerator according to the pre-irradiation detection results, avoiding the lag in the traditional calibration method. The Kalman filter can handle random noise and uncertainty in the system, and can still maintain a good calibration effect in a more complex and unstable environment, reducing the impact of errors on the irradiation results. The calibration parameters can be continuously updated according to real-time observation data to adapt to changes in different materials or environments, thereby improving the flexibility of the system.

[0090] The accelerator unit starts the accelerator after adjusting the calibration parameters to accurately irradiate the material including:

[0091] According to the calibration results, the target electron beam energy and distribution of the accelerator are set, and the electron beam energy and distribution of the accelerator are monitored in real time. By comparing the target value and the actual value, the error signal e(t) is obtained. According to the error value, the PID control algorithm is used to adjust the electron beam energy of the accelerator. The specific formula is:

[0092]

[0093] Among them, u(t) represents the control signal, e(t) represents the error value, K p , K i , K d Represents proportional, integral, and differential coefficients.

[0094] This embodiment uses a PID control algorithm to dynamically adjust the electron beam energy of the accelerator in real time. The electron beam energy output by the accelerator unit is compared with the target value through real-time monitoring to generate an error signal. The PID controller calculates the adjustment amount based on the error signal, and dynamically adjusts the control parameters of the accelerator through the synergistic effect of the three parts of proportion, integration, and differentiation to ensure the accurate and stable output of the electron beam energy. Through the PID control algorithm, the accelerator can accurately adjust the output electron beam energy according to the real-time error to ensure that the irradiation effect meets the predetermined standard. The PID control algorithm can quickly respond to the change of the error and make corrections in advance through the differential term, so that the accelerator can adjust the target energy more quickly and accurately. The integral term in the PID control can accumulate error information, eliminate steady-state errors, and ensure that the electron beam energy of the accelerator is stable and does not deviate from the target value during long-term operation. The PID control has good anti-interference ability, can adapt to the noise and interference in the environment or system, and enables the accelerator to maintain stable output under different working conditions.

[0095] The calibration unit also includes:

[0096] Obtaining initial parameters for pre-irradiation testing;

[0097] Analyze the deviation between initial parameters and historical standard parameters;

[0098] Based on the deviation, calculating an adjustment amount for the calibration parameter;

[0099] Adjust the calibration parameters of the accelerator;

[0100] The adjustment amount of the calculated calibration parameter is specifically:

[0101] Calculate the deviation D between the initial parameter P of the pre-irradiation test and the historical standard parameter S, the formula is D = |P-S|;

[0102] Wherein, P represents the initial parameters of pre-irradiation detection, S represents the historical standard parameters, and D represents the deviation;

[0103] Determine whether the deviation D exceeds a predetermined threshold T;

[0104] If the deviation D is greater than the threshold T, the calibration parameters need to be adjusted; otherwise, the calibration parameters do not need to be adjusted.

[0105] If the deviation D is greater than the threshold T, the calibration parameters that need to be adjusted include:

[0106] If D>T, proceed to the next step;

[0107] Calculate the adjustment amount A of the calibration parameter using the formula A = k × D;

[0108] Among them, k represents the proportionality coefficient, and D represents the deviation;

[0109] Based on the calculated adjustment amount A, updating the calibration parameters of the accelerator;

[0110] Based on the calculated adjustment amount A, updating the calibration parameters of the accelerator includes:

[0111] Get the current calibration parameter C;

[0112] Calculate the new calibration parameter N, the formula is N = C + A;

[0113] Where C represents the current calibration parameter and A represents the adjustment amount;

[0114] Based on the calculated new calibration parameter N, the calibration parameter of the accelerator is updated;

[0115] Based on the calculated new calibration parameter N, updating the calibration parameters of the accelerator includes:

[0116] After each update of the calibration parameters, the new calibration parameters N and the current time T are recorded;

[0117] Determine whether the time interval ΔT between two updates is less than the predetermined time threshold t, the formula is ΔT <t;

[0118] Where ΔT represents the actual time interval between two updates, and t represents the predetermined time threshold;

[0119] If ΔT is less than t, continue monitoring, otherwise complete the update process.

[0120] The calibration unit analyzes the initial parameters obtained from the pre-irradiation test and calculates the deviation between them and the historical standard parameters. When the deviation exceeds the predetermined threshold, the system calculates the adjustment amount of the calibration parameters according to the size of the deviation, and dynamically updates the calibration parameters of the accelerator to ensure that it maintains the best irradiation state. The calibration parameters are recorded each time they are updated, and the adjustment frequency is further optimized through time interval monitoring. By analyzing the initial parameters of the material in real time and calculating the deviation, the calibration parameters of the accelerator can be accurately adjusted to ensure that the irradiation effect reaches the best state. The calibration parameters can be automatically adjusted according to the deviation, reducing manual intervention and improving the automation of the production process. The adjustment process of the calibration parameters is monitored at time intervals to ensure that the accelerator maintains the optimal calibration state in a short period of time, avoiding excessive frequent adjustments that lead to system instability. By referring to the historical standard parameters, the system can perform personalized calibration according to the characteristics of different materials, improving applicability and flexibility.

[0121] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent electron accelerator for industrial irradiation, characterized in that: include: A conveying unit, used for conveying the material to be irradiated into the irradiation chamber; The pre-irradiation detection unit connected to the conveying unit is used to perform pre-irradiation detection on the material and record the initial parameters of the material. Specifically, by collecting data on the initial parameters of the material and the corresponding irradiation effects, a multivariate linear regression model that can predict the irradiation effect is trained based on historical data. Based on the initial parameters of the material, the model is used to predict whether the effect after irradiation meets the standards, and the initial parameters of the material are transmitted to the accelerator unit to assist in subsequent irradiation control; A calibration unit connected to the pre-irradiation detection unit is used to adjust the calibration parameters of the accelerator based on the pre-irradiation detection result, specifically to initialize the state estimation and error covariance matrix of the Kalman filter, use the result of the pre-irradiation detection unit as observation data, use the prediction formula of the Kalman filter to estimate the current calibration parameters according to the previous state, and use the Kalman gain to update the calibration parameters based on the difference between the measured value and the predicted value; An accelerator unit connected to the calibration unit is used to start the accelerator after the calibration parameters are adjusted to accurately irradiate the material; The maintenance calibration unit connected to the accelerator unit is used to perform parameter calibration again after maintenance to ensure that the accuracy deviation of the accelerator is within the allowable range. Specifically, the value range of the calibration parameters of the accelerator is set according to the initial state after maintenance, and a set of random calibration parameters is generated as the initial population. The calibration accuracy of each set of calibration parameters is evaluated by the fitness function, and an excellent parameter combination is selected according to the fitness. Crossover and mutation are performed to generate a new parameter combination. After multiple iterations, the calibration parameter combination with the highest accuracy is finally obtained.

2. The intelligent electron accelerator for industrial irradiation according to claim 1, characterized in that: The conveying unit conveys the material to be irradiated into the irradiation chamber, comprising: Establish a map of the working area of ​​the conveyor unit, set the starting point and target point of material conveyance, convey the material from the loading point to the irradiation chamber, and apply the A-star algorithm based on heuristic search to optimize the material conveyance path through the following evaluation function: f(n)=g(n)+h(n), Among them, f(n) represents the total cost of node n, g(n) represents the actual cost from the starting point to node n, h(n) represents the heuristic estimated cost from node n to the target node, and n represents a node in the path search.

3. The intelligent electron accelerator for industrial irradiation according to claim 1, characterized in that: The pre-irradiation detection unit performs pre-irradiation detection on the material and records the initial parameters of the material, and further includes: The specific formula for multivariate linear regression is: y=β0+β1x1+β2x2+...+β n x n +∈, Where y represents the expected irradiation result of the material, x1, x2, ..., x n Indicates different initial parameters of the material during pre-irradiation testing, β1, β2, ..., β n Represents the regression coefficient to be trained, and ∈ represents the error term.

4. The intelligent electron accelerator for industrial irradiation according to claim 1, characterized in that: The calibration unit adjusting the calibration parameters of the accelerator based on the pre-irradiation detection result also includes: The prediction steps of the Kalman filter are: in, represents the estimated value of the predicted state at the kth moment, represents the updated state estimate at the kth moment, A represents the state transfer matrix, B represents the control input matrix, and u k Represents the control input value, P k|k-1 represents the prediction error covariance matrix, P k|k represents the updated error covariance matrix, Q represents the process noise covariance matrix; The formula for the update step is: K k =P k|k-1 H T =(HP k|k-1 H T +R) -1 , P k|k =(I-K k H)P k|k-1 , Among them, K k represents the Kalman gain, P k|k-1 represents the prediction error covariance matrix, z k represents the measurement value at the kth moment, H represents the observation matrix, R represents the measurement noise covariance matrix, represents the updated state estimate at the kth moment, represents the predicted state estimate at the kth moment, and I represents the identity matrix.

5. The intelligent electron accelerator for industrial irradiation according to claim 1, characterized in that: The accelerator unit starts the accelerator after the calibration parameters are adjusted, and the material is accurately irradiated, including: According to the calibration results, the target electron beam energy and distribution of the accelerator are set, and the electron beam energy and distribution of the accelerator are monitored in real time. By comparing the target value and the actual value, the error signal e(t) is obtained. According to the error value, the PID control algorithm is used to adjust the electron beam energy of the accelerator. The specific formula is: Among them, u(t) represents the control signal, e(t) represents the error value, K p , K i , k d Represents proportional, integral, and differential coefficients.

6. The intelligent electron accelerator for industrial irradiation according to claim 1, characterized in that: The calibration unit also includes: Obtaining initial parameters for pre-irradiation testing; Analyze the deviation between initial parameters and historical standard parameters; Based on the deviation, calculating an adjustment amount for the calibration parameter; Adjust the calibration parameters of the accelerator; The adjustment amount of the calculated calibration parameter is specifically: Calculate the deviation D between the initial parameter P of the pre-irradiation test and the historical standard parameter S, the formula is D = |P-S|; Wherein, P represents the initial parameters of pre-irradiation detection, S represents the historical standard parameters, and D represents the deviation; Determine whether the deviation D exceeds a predetermined threshold T; If the deviation D is greater than the threshold T, the calibration parameters need to be adjusted; otherwise, the calibration parameters do not need to be adjusted.

7. The intelligent electron accelerator for industrial irradiation according to claim 6, characterized in that: If the deviation D is greater than the threshold T, the calibration parameters need to be adjusted, including: If D>T, proceed to the next step; Calculate the adjustment amount A of the calibration parameter using the formula A = k × D; Among them, k represents the proportionality coefficient, and D represents the deviation; Based on the calculated adjustment amount A, the calibration parameters of the accelerator are updated.

8. The intelligent electron accelerator for industrial irradiation according to claim 7, characterized in that: The updating of the calibration parameters of the accelerator based on the calculated adjustment amount A comprises: Get the current calibration parameter C; Calculate the new calibration parameter N, the formula is N = C + A; Where C represents the current calibration parameter and A represents the adjustment amount; Based on the calculated new calibration parameter N, the calibration parameter of the accelerator is updated.

9. The intelligent electron accelerator for industrial irradiation according to claim 8, characterized in that: The updating of the calibration parameters of the accelerator based on the calculated new calibration parameters N includes: After each update of the calibration parameters, the new calibration parameters N and the current time T are recorded; Determine whether the time interval ΔT between two updates is less than the predetermined time threshold t, the formula is ΔT <t; Wherein, ΔT represents the actual time interval between two updates, and t represents the predetermined time threshold; If ΔT is less than t, continue monitoring, otherwise complete the update process.

Images