Intelligent processing system and method for associated radioactive waste residues

Abstract

The application discloses an intelligent processing system and method for associated radioactive waste residues, and relates to the technical field of waste residue processing.The system comprises a bag opening module for performing a bag opening action;a monitoring module, which uses a material bag surface monitoring unit to extract wrinkle contours in a material bag surface image and simplify them into a line set;and a material bag weight monitoring unit for monitoring weight data of the material bag;a calculation module for calculating changes in physical characteristics of the line set and weight changes of the material bag along a time axis, and aligning and fusing the changes in physical characteristics of the line set with the weight changes of the material bag along the time axis.The application introduces an image visual disturbance path modeling mechanism on the basis of traditional weight monitoring, extracts the surface wrinkle evolution process of the material bag after bag opening into a "line set", and tracks the movement state of the path on a continuous time axis, thereby realizing structural identification of whether the material is continuously discharged and whether the discharge is smooth.

Description

Technical Field

[0001] This invention relates to the field of waste residue treatment technology, specifically to an intelligent treatment system and method for associated radioactive waste residue. Background Technology

[0002] Associated radioactive waste is inevitably generated during mineral resource extraction, rare earth smelting, and oil and gas processing. It is characterized by high radioactivity, complex composition, and high moisture content, and typically requires temporary storage in bags, centralized unpacking, and closed-loop transportation to enter a harmless treatment system. Intelligent solid waste treatment lines often include modules such as automated guided vehicles, intelligent unpacking, intelligent transfer, and waste storage linkage. Whether the material can be smoothly discharged after bagging is a critical factor affecting the system's stability, safety, and treatment efficiency.

[0003] In actual engineering processes, waste materials are transported to the integrated processing area by intelligent overhead cranes. The system performs steps such as automatic bag breaking, pretreatment, intelligent transfer, compaction, and warehousing. Existing systems generally use weighing modules to monitor the weight change trend of the material bags to determine whether the discharge is complete. However, this single weight judgment method has several failure risks, such as material accumulation, adhesion to the walls, differences in bag strength, and image stillness artifacts, all of which may lead to problems such as "false bag breaking," "false discharge," or "delayed emptying," thereby affecting the synchronous operation of subsequent systems and environmental safety.

[0004] Therefore, there is an urgent need to establish a multi-source fusion intelligent recognition mechanism that combines image perturbation structure and weight change trend. This mechanism should be able to sense the dynamic perturbation path during the discharge process and quantify whether the surface morphology changes after bag breaking are synchronized with weight changes, forming a more stable and interpretable discharge state discrimination logic. Especially in the unmanned treatment of associated radioactive waste, this application proposes an intelligent treatment system and method for associated radioactive waste residue. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent treatment system and method for associated radioactive waste residue, so as to solve the problems mentioned in the background art.

[0006] The present invention can be achieved through the following technical solution: an intelligent treatment system for associated radioactive waste residue, comprising an unpacking module, a monitoring module and a calculation module;

[0007] The unpacking module is used to perform the bag-breaking action and monitor the blade advancement status during the bag-breaking process, including:

[0008] The cutting tool assembly features a drive tool with both feed speed and the ability to cut material bags;

[0009] The thrust monitoring unit acquires the thrust data of the cutter assembly when cutting the material bag, and determines whether the bag has been effectively broken by comparing the thrust data with the set thrust data threshold.

[0010] The monitoring module includes a material bag surface monitoring unit and a material bag weight monitoring unit;

[0011] The material bag surface monitoring unit is used to extract the wrinkle contours from the material bag surface image, simplify them into a set of lines, and send them to the calculation module;

[0012] The material bag weight monitoring unit is used to monitor the weight data of the material bag and send it to the calculation module;

[0013] After receiving the weight data of the line set and the material bag, the calculation module performs the following calculations respectively:

[0014] Line set: Calculate the changes in its physical characteristics along the time axis, including quantity, distribution location, and frequency of change;

[0015] Weight data: Calculate the weight change of the material bag along the time axis;

[0016] The calculation module aligns and fuses the changes in the physical characteristics of the line set with the changes in the weight of the material bag over time to identify the state of the material bag. Specifically, the judgment logic includes, but is not limited to, the following methods:

[0017] a. When the line set changes significantly within the set time window, and the weight data shows a downward trend within the corresponding time period, the system determines that the material bag is in a normal discharge state, that is, the bag breaking action has been completed and the material is being continuously released.

[0018] The system maintains the current operating status and records the material outflow trend in real time for accumulating the material output.

[0019] If the weight drop of the material bag is monitored to be gradual, prepare for the next bag handling process in advance (such as hoisting, bag replacement, or moving).

[0020] b. When the line set remains stable and the weight data change value approaches zero over a period of time, the system determines that the material bag is stationary or not discharging, which may indicate that material blockage may occur.

[0021] System response: After a certain period of time, the system restarts the unpacking module to break the bag a second time and intervenes in the material bag by vibrating and shaking it to make the material bag move.

[0022] If intervention is ineffective, the system will enter a "card bag warning" state, prompting manual handling;

[0023] c. When the line set continues to change, but the weight data does not show an effective downward trend, the system determines that the material bag is in an abnormal movement state, that is, there may be abnormal behaviors such as jamming, clumping, and false discharge.

[0024] The system responds by first intervening in the material bag to determine whether the material structure inside the bag can be released. If there is no change, it executes the "deeper cut" or "extend cut time" operation.

[0025] If multiple interventions fail, the bag will be recorded as "material jamming abnormality," prompting manual handling. At the same time, subsequent bag filling operations will be suspended to prevent the system from malfunctioning due to the previous bag not being emptied.

[0026] d. When the weight data continues to decrease but the line set hardly changes, it is necessary to determine whether the image stability is an illusion caused by the strong rigidity of the bag support structure. This can be addressed by actively adjusting the movement state of the material bag to improve the accuracy of the judgment.

[0027] The system responds by intervening in the material bag and simultaneously capturing image frames to determine whether the lines have changed again.

[0028] If the changes are reversed, update the system background model and restore normal monitoring.

[0029] If there is still no image change after interference, a "Image acquisition abnormality" message will be displayed, which will be recorded as a special case. Alternatively, the image module restart / diagnostic process will be initiated. At the same time, the system can enter "fault-tolerant operation" mode, which will only rely on weight changes to control the next step of processing.

[0030] A further technical improvement of the present invention lies in: a method for calculating a line set, comprising the following steps:

[0031] S1. The material bag surface monitoring unit periodically collects grayscale images of the target material bag surface. Specifically, the collection frequency can be set based on the overall computational load, such as one frame every two seconds or one frame every three seconds.

[0032] Furthermore, after acquiring the grayscale image, the material bag surface monitoring unit preprocesses it, including:

[0033] Use Gaussian blur to reduce noise in grayscale images;

[0034] Enhance the grayscale contrast of the denoised grayscale image to improve the distinguishability of grayscale transitions;

[0035] S2. The material bag surface monitoring unit calculates the grayscale difference between the remaining adjacent pixels for each pixel in the grayscale image;

[0036] If the grayscale difference between a pixel and multiple surrounding pixels exceeds the set grayscale fluctuation threshold, then the pixel is determined to be in the grayscale fluctuation area and is a fluctuating pixel.

[0037] All pixels that meet the grayscale fluctuation conditions are marked in the form of a binary map to form the initial grayscale fluctuation map in the current image frame;

[0038] Subsequently, the material bag surface monitoring unit performs connected region analysis or image segmentation statistics on the initial grayscale fluctuation map to obtain the wrinkled area;

[0039] When using connected component analysis, all adjacent wavy pixel groups are extracted, and each group represents a folded region.

[0040] When using image segmentation and statistical methods, the grayscale image is divided into multiple grid blocks. The number of fluctuating pixels is counted in each grid block, and grid blocks that meet the quantity conditions are marked as wrinkled regions.

[0041] S3. Use the centroid-boundary centerline connection method to simplify the region structure into a 1-pixel wide main direction line by connecting the wrinkled regions extracted in S2.

[0042] S4. Encode the main direction lines in S3 into a unified "line set" for the current frame, where each line contains the following parameters:

[0043] Starting and ending points;

[0044] Line length;

[0045] The direction and angle of the lines.

[0046] A further technical improvement of the present invention is that the material bag surface monitoring unit simplifies the individual lines in the line collection:

[0047] A1. Merging adjacent lines:

[0048] Traverse the set of lines and determine whether any two lines satisfy any of the following conditions: the distance between their endpoints is less than the spatial threshold, the difference in their directional angles is less than the angular tolerance, or their image projections are connected or overlap.

[0049] When the conditions are met, a new line with the furthest start and end points is generated by straight line fitting to replace the original line until there are no more objects to merge.

[0050] A2. Removal of Noise Lines:

[0051] By using preset filtering conditions, lines that do not meet the filtering conditions are filtered out;

[0052] A3. Line normalization processing:

[0053] Extract the starting and ending coordinates, length, and direction angle of each line to form a standardized set of lines.

[0054] A further technical improvement of the present invention is that the method for calculating changes in physical characteristics by the calculation module includes the following steps:

[0055] Z1, Extraction of line quantity variations:

[0056] The total number of normalized lines in each frame is counted and compared with the number of lines in the previous frame. If the difference in the number of lines is greater than the preset difference threshold, the corresponding line is marked as "change in the number of wrinkles".

[0057] Z2. Extract the coordinates of the geometric midpoints of each line, obtain the set of midpoint coordinates, and calculate the overall geometric centroid.

[0058] Compare the centroid coordinates of the current frame with the centroid coordinates of the previous frame to obtain the spatial displacement value;

[0059] When the spatial displacement value is greater than the preset movement distance threshold, the wrinkle position in the current frame is marked as "position change".

[0060] Z3. Set a fixed time window, accumulate the number of frames marked as "change in the number of wrinkles" or "change in position" within the window to obtain the line change frequency, and compare it with the preset frequency threshold.

[0061] Image frames with line changes at a frequency not less than a frequency threshold are considered perturbation frames, while those with changes less than the frequency threshold are considered stable frames.

[0062] A further technical improvement of the present invention is that the method for the calculation module to perform fusion judgment includes the following steps:

[0063] H1. The calculation module establishes a sequence of weight change values ​​W(t) of the material bag in chronological order along the first time axis. When a downward trend is observed at multiple consecutive sampling points, the first time point T is recorded. w As the "starting point of weight change";

[0064] A perturbation characteristic change sequence L(t) of the line set is established through a second time axis. When the number of lines, centroid, or distribution changes abruptly, the starting time T of the change is recorded. l As the "starting point of line changes";

[0065] A complete perturbation time axis is constructed using frame sequences: L = {L(1), L(2), ..., L(T)};

[0066] H2. The calculation module sets a dynamic delay tolerance threshold ΔT. th And calculate the time difference ΔT between the starting point of weight change and the starting point of line change;

[0067] Furthermore, the calculation module is based on the time difference ΔT and the dynamic delay tolerance threshold ΔT. th The comparison result is matched with the preset judgment logic.

[0068] A further technical improvement of the present invention is that the calculation module is also used to monitor the path continuity of each line extracted from the image frame in the time series, and associate the path continuity with the batch identifier of the material bag, in order to construct a traceable mapping relationship between the bag discharge behavior and the batch quality, specifically including:

[0069] P1. Line set extraction and numbering:

[0070] For each frame, visualize it and extract its line set.

[0071] For each line in the line set, record its start and end coordinates, length, and direction angle;

[0072] P2, Cross-frame line path matching:

[0073] Line matching is performed in consecutive frames t and t+1 under the following conditions;

[0074] The distance difference between the start and end coordinates is less than the preset distance difference threshold;

[0075] The change in orientation angle is less than a preset angle threshold;

[0076] The length change is within a preset length change threshold range;

[0077] If the lines are successfully matched, a unique trajectory ID is assigned. k Recorded as path P k ;

[0078] P3, Path continuity construction:

[0079] For each path P k Record the consecutive frames N in which it exists k ;

[0080] Record the trajectory sequence of its center point: T k ={(x t ,x y ),(x t+1 ,x y+1 ),...,(x t+n ,x y+n )};

[0081] It also calculates the continuity indices of the trajectory, including the variance of direction changes, the number of jumps, and the smoothness of the trajectory;

[0082] P4. Define the continuous scoring function:

[0083] S k =f(N) k (variance of direction change, number of jumps, trajectory smoothness);

[0084] By merging the continuity scores of all paths, the overall path stability index S of the current set of lines on the surface of the material bag is obtained. total .

[0085] A further technical improvement of the present invention is that: after each material bag has finished discharging, the system records its S. total The value is archived together with the sequence of weight change values ​​for that material bag;

[0086] The overall path stability index S corresponding to multiple material bags total In this embodiment, RFID is used to associate the data with the batch.

[0087] If a batch of material bags generally exhibits severe path jumps and discontinuous trajectories, it will be marked as a batch with inconsistent quality.

[0088] This invention also discloses an intelligent treatment method for associated radioactive waste residue, comprising the following steps;

[0089] Step 1: Use a cutting tool assembly with both propulsion speed and the ability to cut the material bag to perform the bag-breaking operation;

[0090] The thrust data of the cutting tool assembly during the cutting of the material bag is collected synchronously. The thrust data is the axial thrust value of the cutting tool assembly during the bag breaking process.

[0091] Determine whether the axial thrust value is within the set axial thrust threshold range and whether its duration exceeds the duration threshold.

[0092] When all of the above conditions are met, the bag breaking is deemed to have been completed effectively;

[0093] Step 2: Obtain an image of the material bag surface using the material bag surface monitoring unit, and extract the wrinkle contours from the material bag surface image;

[0094] The extracted fold contours are simplified into a set of lines, and this set of lines is sent to the calculation module as image feature data;

[0095] Step 3: Monitor the weight data of the material bag through the material bag weight monitoring unit and send it to the calculation module;

[0096] Step 4: The calculation module calculates the physical feature changes of the line set along the time axis, including the number of lines, their distribution positions, and the frequency of change.

[0097] Step 5: The calculation module calculates the weight change data of the material bag along the time axis;

[0098] Step Six: The calculation module aligns the changes in the physical characteristics of the line set with the changes in the weight of the material bag on the time axis and performs a fusion judgment;

[0099] Based on the fusion results, identify the current status of the material bag, including whether the bag is broken, whether it is continuously discharging material, and whether there is a blockage or abnormal condition.

[0100] Compared with the prior art, the present invention has the following beneficial effects:

[0101] This invention introduces an image visual disturbance path modeling mechanism on the basis of traditional weight monitoring. It extracts the evolution process of surface wrinkles after the material bag is broken into a "set of lines" and tracks the motion state of its path on a continuous time axis. This enables structural identification of whether the material discharge is continuous and whether the material feeding is smooth, which significantly improves the accuracy of judging abnormalities such as material jamming, blockage, and false discharge.

[0102] Furthermore, the invention introduces a "dual time axis fusion judgment mechanism," which compares the starting point of weight change with the starting point of line disturbance over time, constructs a delay tolerance model ΔT, and accurately identifies various states such as normal material discharge, no material discharge, disturbance failure, or image artifacts through its difference and directional trend, and matches corresponding automatic control strategies (such as bag breaking again, active shaking, alarm prompts, etc.), significantly improving the system's automatic intervention and adaptability.

[0103] Meanwhile, this invention further proposes a trajectory stability index modeling method, which establishes a scoring function for the disturbance path of each line, comprehensively evaluates its jump number, directional continuity and trajectory slip, and establishes a correspondence between the scoring results of multiple material bags and their batch information. This can be used to determine the quality stability and structural consistency of a batch of bags in the feeding process, providing data support for source quality traceability and bag material optimization. Attached Figure Description

[0104] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0105] Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0106] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0107] Example 1

[0108] Please see Figure 1 As shown, the present invention provides an intelligent treatment system for associated radioactive waste residue, including an unpacking module, a monitoring module, and a calculation module;

[0109] The unpacking module is used to perform the bag-breaking action and monitor the blade advancement status during the bag-breaking process, including:

[0110] The cutting tool assembly features a drive tool with both feed speed and the ability to cut material bags;

[0111] The thrust monitoring unit acquires the thrust data of the cutting tool assembly when cutting the material bag, and determines whether the bag has been effectively broken by comparing the thrust data with a set thrust data threshold. Specifically, in this embodiment, the thrust data is the axial thrust value of the cutting tool assembly during the bag breaking process. When cutting the material bag, the axial thrust of the cutting tool assembly will change significantly, specifically, the thrust value increases for a short period of time and remains within a stable range. By monitoring the time of the change, it is determined whether the tool assembly is cutting the material bag. Therefore, the thrust data threshold includes the axial thrust threshold range and the duration threshold of the change.

[0112] Therefore, when the axial thrust of the thrust monitoring unit is within the axial thrust threshold range and the duration exceeds the duration threshold, the system determines that the bag breaking has been completed effectively.

[0113] The monitoring module includes a material bag surface monitoring unit and a material bag weight monitoring unit;

[0114] The material bag surface monitoring unit is used to extract the wrinkle contours from the material bag surface image, simplify them into a set of lines, and send them to the calculation module;

[0115] The method for calculating line sets includes the following steps:

[0116] S1. The material bag surface monitoring unit periodically collects grayscale images of the target material bag surface. Specifically, the collection frequency can be set based on the overall computational load, such as one frame every two seconds or one frame every three seconds.

[0117] Furthermore, after acquiring the grayscale image, the material bag surface monitoring unit preprocesses it, including:

[0118] Use Gaussian blur to reduce noise in grayscale images;

[0119] Enhance the grayscale contrast of the denoised grayscale image to improve the distinguishability of grayscale transitions;

[0120] S2. The material bag surface monitoring unit calculates the grayscale difference between the remaining adjacent pixels for each pixel in the grayscale image;

[0121] If the grayscale difference between a pixel and multiple surrounding pixels exceeds the set grayscale fluctuation threshold, then the pixel is determined to be in the grayscale fluctuation area and is a fluctuating pixel.

[0122] All pixels that meet the grayscale fluctuation conditions are marked in the form of a binary map to form the initial grayscale fluctuation map in the current image frame;

[0123] Subsequently, the material bag surface monitoring unit performs connected region analysis or image segmentation statistics on the initial grayscale fluctuation map to obtain the wrinkled area;

[0124] When using connected component analysis, all adjacent wavy pixel groups are extracted, and each group represents a folded region.

[0125] When using image segmentation and statistical methods, the grayscale image is divided into multiple grid blocks. The number of fluctuating pixels is counted in each grid block, and grid blocks that meet the quantity conditions are marked as wrinkled regions.

[0126] S3. Connect the wrinkled regions extracted in S2 using the centroid-boundary midline method to simplify the region structure into a 1-pixel wide main direction line. Specifically, this includes:

[0127] Contour region acquisition: Perform contour tracking operation on each wrinkle region to obtain its boundary contour coordinate set. The boundary is a closed or semi-closed curve, representing the two-dimensional projection range of the wrinkle in the image.

[0128] Centroid calculation: Calculate the geometric center (centroid) of each contour region. The centroid position can be obtained by averaging the x and y coordinates of all pixels within the contour, representing the centroid of pixel distribution within the region;

[0129] Boundary point selection: Select a set of representative boundary points on the contour area, specifically including:

[0130] The farthest boundary point, i.e., the contour point that is furthest from the centroid;

[0131] The principal direction boundary point pair is the boundary point pair that has the farthest projection distance in the horizontal, vertical, or principal direction angle.

[0132] Connect the centroid and boundary points to form a central axis: Starting from the centroid of the region, draw straight lines sequentially towards the aforementioned boundary points (or the midpoints of boundary point pairs) to obtain one or more main directional lines emanating from the centroid. If the region is approximately elongated, the farthest points can be selected to connect and construct the main axis direction lines;

[0133] The central axis is simplified into a line: if multiple centroid-boundary lines exist, a single main direction line can be generated through merging, fitting, etc., to represent the main extension direction of the region. This ultimately forms a simplified line represented at pixel-level resolution to represent the folded region.

[0134] S4. Encode the main direction lines in S3 into a unified "line set" for the current frame, where each line contains the following parameters:

[0135] Starting and ending points;

[0136] Line length;

[0137] Line direction and angle;

[0138] Furthermore, the material bag surface monitoring unit simplifies the lines in the line collection:

[0139] A1. Merging adjacent lines:

[0140] Iterate through the set of lines and analyze whether any two lines satisfy the following condition:

[0141] The distance between the endpoints of the two lines is less than a preset spatial threshold.

[0142] The difference in the directional angle between the two lines is less than the set angle tolerance;

[0143] The projections of the two lines onto the image are connected or overlap.

[0144] For any line pair that meets any of the conditions, a new merged line is generated by fitting a straight line. The starting point and ending point of the new line are the two farthest points of the original line endpoints. The original line that was merged is deleted, and a new merged line is added until there are no more objects to merge. This avoids the situation where a fold is mistakenly divided into multiple lines due to the breakage of the fold edge or image interference, thus improving the overall structural stability.

[0145] A2. Removal of Noise Lines:

[0146] By using preset filtering conditions, lines that do not meet the filtering conditions are filtered out;

[0147] In this embodiment, the filtering conditions include: setting a line length threshold (removing lines that are too short), setting the line direction change range (removing lines that have excessive direction changes, are scattered in relative direction distribution, or lack a main direction), and ensuring that the line segment does not constitute a structural correlation with other lines in its area (removing isolated or randomly distributed lines).

[0148] A3. Normalize each line:

[0149] Extract the following parameters for each line:

[0150] Extract the coordinates of the start and end points of each line within the same image coordinate system;

[0151] Line length;

[0152] Line direction and angle;

[0153] All line structures are packaged into a standard line set data in a uniform format for subsequent comparison and statistics on the timeline;

[0154] The material bag weight monitoring unit is used to monitor the weight data of the material bag and send it to the calculation module;

[0155] After receiving the weight data of the line set and the material bag, the calculation module performs the following calculations respectively:

[0156] Line set: Calculate the changes in its physical characteristics along the time axis, including quantity, distribution location, and frequency of change;

[0157] The calculation module calculates changes in physical characteristics using the following methods:

[0158] Z1, Extraction of line quantity variations:

[0159] In each image frame, the calculation module counts the total number of lines in the set of lines that have undergone normalization.

[0160] The number of lines in the current frame is compared with the number of lines in the previous frame to obtain the difference in the number of lines.

[0161] If the difference in the number of lines is greater than the preset difference threshold, the corresponding mark will be "change in the number of wrinkles";

[0162] Z2. Extraction of line distribution position changes:

[0163] For each line in the current image frame, extract its geometric midpoint coordinates (e.g., the average of the coordinates of the start and end points of the line segment);

[0164] The set of midpoint coordinates of all lines is used as the spatial distribution feature of the lines in this frame;

[0165] Calculate the overall geometric centroid (center position) of the midpoint coordinate set, compare the centroid coordinates of the current frame with the centroid coordinates of the previous frame, and obtain the spatial displacement value.

[0166] When the spatial displacement value is greater than the preset movement distance threshold, the wrinkle position in the current frame is marked as "position change".

[0167] Z3. The calculation module sets a fixed time window, records the line set of all continuous frames within the fixed time window, and accumulates the number of image frames marked as "change in the number of wrinkles" or "change in position" to obtain the line change frequency. It then compares the line change frequency with a preset frequency threshold. Image frames with a line change frequency not less than the frequency threshold are disturbed frames, and image frames with a frequency less than the frequency threshold are stable frames. This is used to determine whether the surface of the material bag is in a state of continuous disturbance.

[0168] Weight data: Calculate the weight change of the material bag along the time axis;

[0169] The calculation module aligns and fuses the changes in the physical characteristics of the line set with the changes in the weight of the material bag over time to identify the state of the material bag. The method for the calculation module to perform the fusion judgment includes the following steps:

[0170] H1. The calculation module establishes a sequence of weight change values ​​W(t) of the material bag in chronological order along the first time axis. When a downward trend is observed at multiple consecutive sampling points, the first time point T is recorded. w As the "starting point of weight change";

[0171] A perturbation characteristic change sequence L(t) of the line set is established through a second time axis. When the number of lines, centroid, or distribution changes abruptly, the starting time T of the change is recorded. l As the "starting point of line changes";

[0172] Specifically, for each frame t in the second time axis, when it is a perturbation frame, L(t) = 1 is recorded, and when it is a stable frame, L(t) = 0 is recorded.

[0173] A complete perturbation time axis is constructed by frame sequence: L={L(1),L(2),...,L(T)}, and the logical frame index of the perturbation time axis is consistent with the image sampling frequency to ensure one-to-one correspondence with the first time axis;

[0174] H2. The calculation module sets a dynamic delay tolerance threshold ΔT. th And calculate the time difference ΔT between the start of the weight change and the start of the line change, i.e., ΔT = T w -T l ;

[0175] Specifically, the judgment logic includes, but is not limited to, the following methods:

[0176] When ΔT≈0, the normal disturbance is consistent with the discharge, and the system determines that the material bag is in a normal discharge state, that is, the bag breaking action has been completed and the material is being continuously released.

[0177] The system maintains the current operating status and records the material outflow trend in real time for accumulating the material output.

[0178] If the weight drop of the material bag is monitored to be gradual, prepare for the next bag handling process in advance (such as hoisting, bag replacement, or moving).

[0179] When there is no ΔT, that is, no disturbance or no weight change, intervention is required, as material blockage may occur.

[0180] System response: After a certain period of time, the system restarts the unpacking module to break the bag a second time and intervenes in the material bag by vibrating and shaking it to make the material bag move.

[0181] If intervention is ineffective, the system will enter a "card bag warning" state, prompting manual handling;

[0182] When ΔT>ΔT th When the system determines that the material bag is in an abnormal movement state, it may be experiencing abnormal behaviors such as jamming, clumping, or false discharge.

[0183] The system responds by first intervening in the material bag to determine whether the material structure inside the bag can be released. If there is no change, it executes the "deeper cut" or "extend cut time" operation.

[0184] If multiple interventions fail, the bag will be recorded as "material jamming abnormality," prompting manual handling. At the same time, subsequent bag filling operations will be suspended to prevent the system from malfunctioning due to the previous bag not being emptied.

[0185] When ΔT < -ΔT th When material discharge occurs but no disturbance is detected, it can be determined whether the image instability is due to the strong rigidity of the bag support structure. This can be addressed by actively adjusting the movement of the material bag to improve the accuracy of the judgment.

[0186] The system responds by intervening in the material bag and simultaneously capturing image frames to determine whether the lines have changed again.

[0187] If the changes are reversed, update the system background model and restore normal monitoring.

[0188] If there is still no image change after interference, a "Image acquisition abnormality" message will be displayed, which will be recorded as a special case. Alternatively, the image module restart / diagnostic process will be initiated. At the same time, the system can enter "fault-tolerant operation" mode, which will only rely on weight changes to control the next step of processing.

[0189] Example 2

[0190] Compared to Example 1, the calculation module in Example 2 is further used to monitor the path continuity of each line extracted from the image frame in the time series, and associate the path continuity with the batch identifier of the material bag to construct a traceable mapping relationship between the bag's discharge behavior and batch quality, specifically including:

[0191] P1. Line set extraction and numbering:

[0192] For each frame, visualize it and extract its line set.

[0193] For each line in the line set, record its start and end coordinates, length, and direction angle;

[0194] P2, Cross-frame line path matching:

[0195] Line matching is performed in consecutive frames t and t+1 under the following conditions;

[0196] The distance difference between the start and end coordinates is less than the preset distance difference threshold;

[0197] The change in orientation angle is less than a preset angle threshold;

[0198] The length change is within a preset length change threshold range;

[0199] If the lines are successfully matched, a unique trajectory ID is assigned. k Recorded as path P k ;

[0200] P3, Path continuity construction:

[0201] For each path P k Record the consecutive frames N in which it exists k ;

[0202] Record the trajectory sequence of its center point: T k ={(x t ,x y ),(x t+1 ,x y+1 ),...,(x t+n ,x y+n )};

[0203] And calculate the continuity index of the trajectory:

[0204] Variance of directional change;

[0205] Number of jumps: The occurrence of matching interruptions, discontinuous trajectories, or the regeneration of new paths in the line's time series reflects interference, instability, or obstruction during the actual material output process. A higher number of jumps indicates poorer temporal stability of the structure. Specifically, this includes:

[0206] Path numbering establishment: For each frame t and the next frame t+1, determine path P. k Whether it can be continuously tracked (i.e., whether it can match lines with the same number);

[0207] If consecutive matches are successful, the trajectory is considered to have not changed.

[0208] If a new numbered path (spatially close) appears after an interruption, it is considered that a jump has occurred;

[0209] Track smoothness: reflects the continuity and stability of a line path's movement direction in a continuous time series, and can indirectly reflect whether the disturbance on the material surface is natural and orderly. If the path direction changes frequently or vibrates violently, the smoothness is poor. Specifically, this includes:

[0210] For path P k In each frame of the image, the direction angle of the corresponding line in that frame is calculated. In this embodiment, Arctangent is used to calculate the direction of the start and end points.

[0211] For path P k In a series of existing frames, calculate the difference in orientation angle between adjacent frames and then calculate the path P. k The sum of orientation changes over its continuous frame segment is calculated, and the reciprocal of the sum of orientation changes is calculated to obtain the trajectory smoothness;

[0212] P4. Define the continuous scoring function:

[0213] S k =f(N) k (variance of direction change, number of jumps, trajectory smoothness);

[0214] By merging the continuity scores of all paths, the overall path stability index S of the current set of lines on the surface of the material bag is obtained. total .

[0215] After each material bag has finished discharging, the system records its S. total The value is archived together with the sequence of weight change values ​​for that material bag;

[0216] The overall path stability index S corresponding to multiple material bags total In this embodiment, RFID is used to associate the data with the batch.

[0217] If a batch of material bags generally exhibits severe path jumps and discontinuous trajectories, it will be marked as a batch with inconsistent quality.

[0218] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0219] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. An intelligent treatment system for associated radioactive waste residue, characterized in that, include: The unpacking module is used to perform the bag-breaking action and to determine whether the bag breaking is effective by monitoring the blade advance status during the bag-breaking process. The monitoring module includes a material bag surface monitoring unit and a material bag weight monitoring unit; The material bag surface monitoring unit is used to extract the wrinkle contours from the material bag surface image, simplify them into a set of lines, and send them to the calculation module; The material bag weight monitoring unit is used to monitor the weight data of the material bag and send it to the calculation module; The calculation module calculates the changes in the physical characteristics of the line set and the weight changes of the material bag along the time axis. It then identifies the state of the material bag by aligning and fusing the changes in the physical characteristics of the line set with the weight changes of the material bag along the time axis.

2. The intelligent treatment system for associated radioactive waste residue according to claim 1, characterized in that, The method for calculating line sets includes the following steps: S1. The material bag surface monitoring unit periodically collects grayscale images of the target material bag surface; S2. The material bag surface monitoring unit calculates the grayscale difference between each pixel and its adjacent pixels in the grayscale image, and marks the pixels that exceed the grayscale fluctuation threshold as grayscale fluctuation areas. Generate an initial grayscale fluctuation map in binary form, and extract the wrinkled regions through connected component analysis or image segmentation statistics; S3. Use the centroid-boundary centerline connection method to simplify the region structure into a 1-pixel wide main direction line by connecting the wrinkled regions extracted in S2. S4. Encode the main direction lines in S3 into a unified "line set" for the current frame, where each line contains the following parameters: Starting and ending points; Line length; The direction and angle of the lines.

3. The intelligent treatment system for associated radioactive waste residue according to claim 2, characterized in that, The material bag surface monitoring unit simplifies the key lines in the line collection: A1. Merging adjacent lines: Traverse the set of lines and determine whether any two lines satisfy any of the following conditions: the distance between their endpoints is less than the spatial threshold, the difference in their directional angles is less than the angular tolerance, or their image projections are connected or overlap. When the conditions are met, a new line with the furthest start and end points is generated by straight line fitting to replace the original line until there are no more objects to merge. A2. Removal of Noise Lines: By using preset filtering conditions, lines that do not meet the filtering conditions are filtered out; A3. Line normalization processing: Extract the starting and ending coordinates, length, and direction angle of each line to form a standardized set of lines.

4. The intelligent treatment system for associated radioactive waste residue according to claim 3, characterized in that, The method for calculating changes in physical characteristics by the calculation module includes the following steps: Z1, Extraction of line quantity variations: The total number of normalized lines in each frame is counted and compared with the number of lines in the previous frame. If the difference in the number of lines is greater than the preset difference threshold, the corresponding line is marked as "change in the number of wrinkles". Z2. Extract the coordinates of the geometric midpoints of each line, obtain the set of midpoint coordinates, and calculate the overall geometric centroid. Compare the centroid coordinates of the current frame with the centroid coordinates of the previous frame to obtain the spatial displacement value; When the spatial displacement value is greater than the preset movement distance threshold, the wrinkle position in the current frame is marked as "position change"; Z3. Set a fixed time window, accumulate the number of frames marked as "change in the number of wrinkles" or "change in position" within the window to obtain the line change frequency, and compare it with the preset frequency threshold. Image frames with line changes at a frequency not less than a frequency threshold are considered perturbation frames, while those with changes less than the frequency threshold are considered stable frames.

5. The intelligent treatment system for associated radioactive waste residue according to claim 4, characterized in that, The method for the calculation module to perform fusion judgment includes the following steps: H1. The calculation module establishes a sequence of weight change values ​​W(t) of the material bag in chronological order along the first time axis. When a downward trend is observed at multiple consecutive sampling points, the first time point T is recorded. w As the "starting point of weight change"; A perturbation characteristic change sequence L(t) of the line set is established through a second time axis. When the number of lines, centroid, or distribution changes abruptly, the starting time T of the change is recorded. l As the "starting point of line variation"; A complete perturbation time axis is constructed using frame sequences: L = {L(1), L(2), ..., L(T)}; H2. The calculation module sets a dynamic delay tolerance threshold ΔT. th And calculate the time difference ΔT between the starting point of weight change and the starting point of line change; Furthermore, the calculation module is based on the time difference ΔT and the dynamic delay tolerance threshold ΔT. th The comparison result is matched with the preset judgment logic.

6. The intelligent treatment system for associated radioactive waste residue according to claim 5, characterized in that, The calculation module is also used to monitor the path continuity of each line extracted from the image frame in the time series, and associate the path continuity with the batch identifier of the material bag.

7. The intelligent treatment system for associated radioactive waste residue according to claim 6, characterized in that, Methods for monitoring path continuity include: P1. Line set extraction and numbering: For each frame, visualize it and extract its line set. For each line in the line set, record its start and end coordinates, length, and direction angle; P2, Cross-frame line path matching: Line matching is performed in consecutive frames t and t+1 under the following conditions; The distance difference between the start and end coordinates is less than the preset distance difference threshold; The change in orientation angle is less than a preset angle threshold; The length change is within a preset length change threshold range; If the lines are successfully matched, a unique trajectory ID is assigned. k Recorded as path P k ; P3, Path continuity construction: For each path P k Record the consecutive frames N in which it exists k ; Record the trajectory sequence of its center point: T k ={(x t ,x y ),(x t+1 ,x y+1 ),...,(x t+n ,x y+n )}; It also calculates the continuity indices of the trajectory, including the variance of direction changes, the number of jumps, and the smoothness of the trajectory; P4. Define the continuous scoring function: S k =f(N) k (variance of direction change, number of jumps, trajectory smoothness); By merging the continuity scores of all paths, the overall path stability index S of the current set of lines on the surface of the material bag is obtained. total .

8. The intelligent treatment system for associated radioactive waste residue according to claim 7, characterized in that, After each material bag has finished discharging, the system records its S. total The value is archived together with the sequence of weight change values ​​for that material bag; The overall path stability index S corresponding to multiple material bags total Associate with batches.

9. An intelligent treatment method for associated radioactive waste residue, characterized in that, The method employs the system of any one of claims 1-8, comprising: Step 1: Perform the bag-breaking action and obtain the axial thrust value of the tool assembly; Determine whether the thrust value is within the axial thrust threshold range and whether the duration exceeds the duration threshold; if both conditions are met, then the bag breaking is considered to have been completed effectively. Step 2: Extract the wrinkle contours from the image of the material bag surface, simplify them into a set of lines, and send them to the calculation module; Step 3: Obtain the weight data of the material bags and send it to the calculation module; Step 4: Calculate the changes in the physical characteristics of the line set and the weight changes of the material bag along the time axis; Step 5: Align and merge the changes in the physical characteristics and weight of the material bag over time to identify the state of the material bag.

Images