Intelligent management method and system for solder paste warehouse

Abstract

The invention discloses an intelligent management method and system for a solder paste library, and relates to the technical field of warehouse management, and the method comprises the steps: analyzing work order data of an MES system, synchronously collecting refrigerator temperature, charging bucket coordinates and weight data, constructing a dynamically mapped three-dimensional warehouse model, employing a quantum annealing algorithm to optimize the moving path of a manipulator, and calculating multi-target parameters. Outputting an optimal path instruction to the driving equipment; the tin taking state after distribution is recorded, secondary weighing verification is conducted on the returned material tank, and the material state of the three-dimensional storage model is updated according to the comparison result of the weight change rate and a preset scrap weight threshold value; and comparing the actual path of the manipulator with the optimization calculation of the manipulator, iteratively updating a path weight coefficient in the three-dimensional storage model, and synchronizing the three-dimensional storage state. A three-dimensional dynamic storage model is constructed based on refrigerator temperature gradient distribution and material tank position weight, the space utilization rate of a refrigeration area is optimized through coordinate mapping, and the positioning error of the manipulator is eliminated in combination with a space-time compensation algorithm.

Description

Technical Field

[0001] The present invention relates to the technical field of warehouse management, and particularly to an intelligent management method and system for a solder paste library. Background Art

[0002] With the significant improvement in the demand for automated solder paste warehouse management in the electronic manufacturing field, traditional solutions mainly focus on warehouse modeling, path planning, and material status monitoring. In terms of warehouse modeling, the existing technology has achieved a digital mapping of the static warehouse layout through 3D scanning and Internet of Things sensing technology; in the field of path optimization, genetic algorithms or A* algorithms are generally used, focusing on single-objective path planning; material status monitoring relies on the linkage of RFID tags and weighing sensors to achieve the early warning function of the basic inventory.

[0003] However, traditional solutions have obvious limitations in terms of adaptability to dynamic environments, multi-objective optimization, and full life cycle management of materials. Specifically, the traditional path planning algorithm is difficult to synchronously optimize multi-objective parameters such as the moving distance, energy consumption, and time of the manipulator in a dynamic warehouse environment, and the algorithm convergence speed is limited by the complexity of the high-dimensional solution space; the material status monitoring lacks a dynamic scrap determination mechanism based on the weight change rate, and the secondary verification of the returned material cans and the update of the warehouse model status do not form a closed loop, resulting in a relatively high scrap determination error rate; in addition, the replenishment strategy based on the FIFO rule cannot predict multi-can reflow conflicts, easily causing delays in work order execution; and the setting of the static torque threshold is difficult to adapt to the dynamic load changes during the operation of the manipulator, resulting in the accumulation of path execution deviations. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an intelligent management method for a solder paste library to solve the problem of low efficiency in dynamic multi-objective path optimization in solder paste warehouse management.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides an intelligent management method for a solder paste library, which includes parsing work order data of an MES system, synchronously collecting cold cabinet temperature, material tank coordinates and weight data, constructing a three-dimensional warehousing model with dynamic mapping, using a quantum annealing algorithm to optimize the moving path of a manipulator, calculating multi-objective parameters, and outputting an optimal path instruction to a driving device; obtaining batch information of incoming solder paste and weighing it, generating storage coordinates according to the remaining capacity in the refrigerated area in the three-dimensional warehousing model and the FIFO rule, and driving the manipulator to store according to the optimized path; when the inventory in the warming area is lower than the preset inventory lower limit threshold, inversely calculating the work order demand time window, simulating the multi-tank warming conflict probability to generate a replenishment queue, extracting the outgoing material tank according to the optimal path instruction, dynamically adjusting the revolution speed curve according to the simulation parameters in the three-dimensional warehousing model, and automatically correcting the operating parameters when the torque deviation exceeds the limit; recording the solder usage status after distribution, reweighing and verifying the returned material tank, and updating the material status of the three-dimensional warehousing model according to the comparison result between the weight change rate and the preset scrap weight threshold; comparing the actual path of the manipulator with the optimized calculation of the manipulator, iteratively updating the path weight coefficient in the three-dimensional warehousing model and synchronizing the three-dimensional warehousing status.

[0008] As a preferred solution of the intelligent management method for the solder paste library of the present invention, wherein: integrating the cold cabinet temperature, material tank coordinates and weight data, and outputting a device status data set;

[0009] Based on the warehousing layout, using a three-dimensional modeling tool to construct a space framework including a cold cabinet, a material tank and a manipulator, and outputting a three-dimensional model;

[0010] Dynamically mapping the device status data set to the three-dimensional model, marking the material tank status and the cold cabinet capacity, and outputting a three-dimensional warehousing model.

[0011] As a preferred solution of the intelligent management method for the solder paste library of the present invention, wherein: the specific steps of outputting the optimal path instruction to the driving device are as follows,

[0012] Using a quantum annealing algorithm, with the goal of minimizing the moving distance, energy consumption and time, constructing an objective function;

[0013] Generating an optimal path through annealing iteration, and decomposing it into a discrete optimal path instruction set;

[0014] Extracting a single instruction matching the current task from the optimal path instruction set, dynamically adjusting the instruction parameters and then transmitting them to the manipulator driving device for execution.

[0015] As a preferred solution of the intelligent management method for the solder paste library of the present invention, wherein: driving the manipulator to store according to the optimized path means controlling the manipulator to sequentially execute the moving instructions based on the optimal path instruction set generated by the quantum annealing algorithm, and transporting the solder paste material tank to the designated refrigerated area storage coordinates in the three-dimensional warehousing model.

[0016] As a preferred solution of the intelligent management method of the solder paste warehouse of the present invention, the preset inventory lower limit threshold refers to the minimum inventory critical value set based on the capacity of the cold storage area in the three-dimensional storage model, the reverse calculation result of the work order demand time window and the real-time status of the reheating area;

[0017] The specific steps of automatically correcting the operating parameters when the torque deviation exceeds the limit are as follows:

[0018] Collect the real-time torque, speed and position data of the robot and output the operation status monitoring data;

[0019] Based on the simulation parameters in the three-dimensional warehouse model and the historical operation status monitoring data, the deviation range of the current torque is dynamically calculated and the dynamic deviation range value is output;

[0020] When the real-time torque exceeds the dynamic deviation range, the correction logic is automatically triggered, the robot revolution speed is adjusted in real time based on the simulation parameters, and the corrected operating parameters are output.

[0021] As a preferred solution of the intelligent management method of the solder paste warehouse of the present invention, the preset scrap weight threshold refers to the minimum remaining weight critical value for determining whether the material tank is invalid, which is preset based on the secondary weighing result of the returned material tank, by calculating the weight change rate;

[0022] The specific steps of updating the material status of the three-dimensional warehouse model are as follows:

[0023] Record the status of tin retrieval after delivery and generate a retrieval log;

[0024] Perform secondary weighing verification on the returned material tanks and output the returned weight data;

[0025] Calculate the weight change rate and compare it with the preset scrap weight threshold. When the weight change rate is greater than the preset scrap weight threshold, output the material scrap status mark; otherwise, output the material available status mark;

[0026] Update the material status of the 3D warehouse model according to the material status identification.

[0027] As a preferred solution of the intelligent management method of the solder paste warehouse of the present invention, the iterative updating of the path weight coefficient in the three-dimensional storage model and the synchronization of the three-dimensional storage status are specifically performed as follows:

[0028] Collect the actual movement trajectory, time and energy consumption data of the robot in real time, and output the actual path data set;

[0029] Combine the actual path data set with the optimal path instruction set, calculate the distance deviation, time error and energy consumption difference through time-space alignment, and output the path deviation data set;

[0030] Based on the path deviation dataset, dynamically adjust the path weight coefficients in the three-dimensional warehousing model, output the updated path weight parameters, map them to the three-dimensional warehousing model, recalculate the dynamic path planning logic in the three-dimensional warehousing model, and output the synchronized three-dimensional warehousing status.

[0031] In a second aspect, the present invention provides an intelligent management system for a solder paste library, including a data parsing module, a path optimization module, a warehousing management module, a status update module, and a model iteration module; the data parsing module is used to parse the work order data of the MES system, synchronously collect the cold cabinet temperature, the coordinates and weight data of the material tanks, construct a dynamically mapped three-dimensional warehousing model, optimize the movement path of the manipulator using the quantum annealing algorithm, calculate multi-objective parameters, and output the optimal path instruction to the driving device; the path optimization module is used to obtain the batch information of the incoming solder paste and weigh it, generate storage coordinates according to the remaining capacity in the refrigerated area and the FIFO rule in the three-dimensional warehousing model, and drive the manipulator to store according to the optimized path; the warehousing management module is used to, when the inventory in the warming area is lower than the preset inventory lower limit threshold, inversely calculate the work order demand time window, generate a replenishment queue after simulating the multi-tank warming conflict probability, extract the outgoing material tank according to the optimal path instruction, dynamically adjust the revolution speed curve according to the simulation parameters in the three-dimensional warehousing model, and automatically correct it when the torque deviation exceeds the limit; the status update module is used to record the solder usage status after distribution, weigh the returned material tank again for verification, and update the material status of the three-dimensional warehousing model according to the comparison result between the weight change rate and the preset scrap weight threshold; the model iteration module is used to compare the actual path of the manipulator with the optimized calculation of the manipulator, iteratively update the path weight coefficients in the three-dimensional warehousing model, and synchronize the three-dimensional warehousing status.

[0032] In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the intelligent management method for the solder paste library as described in the first aspect of the present invention is implemented.

[0033] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the intelligent management method for the solder paste library as described in the first aspect of the present invention is implemented.

[0034] The beneficial effects of the present invention are as follows: The present invention reconstructs the path planning of the manipulator through the quantum annealing algorithm, establishes a multi-objective optimization function by integrating the moving distance, energy consumption, and time parameters, and utilizes the characteristics of quantum computing to break through the search bottleneck of traditional algorithms in high-dimensional solution spaces, realizing real-time path generation and correction in a dynamic warehousing environment. A three-dimensional dynamic warehousing model is constructed based on the temperature gradient distribution of the refrigerator and the position weight of the storage tank, the space utilization rate of the refrigerated area is optimized through coordinate mapping, and the positioning error of the manipulator is eliminated by combining the space-time compensation algorithm. For material management, a reverse parsing mechanism for the work order time sequence is designed to predict replenishment conflicts and generate a priority queue, and a torque dynamic compensation logic is synchronously integrated to cope with sudden load fluctuations. In quality control, a dynamic scrapping determination rule is established by comparing the secondary weighing of the returned storage tank with the historical data trend, and the self-adaptability of the warehousing model is optimized by using the iterative feedback of the path execution data, finally forming a closed-loop control system of environment perception-intelligent decision-making-autonomous correction. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0036] Figure 1 It is a flowchart of the intelligent management method for the solder paste library in Embodiment 1.

[0037] Figure 2 It is a flowchart of the construction of the three-dimensional warehousing model in Embodiment 1.

[0038] Figure 3 It is a flowchart of the path optimization of the quantum annealing algorithm in Embodiment 1.

[0039] Figure 4 It is a flowchart of the path optimization and correction of the manipulator in Embodiment 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe the specific embodiments of the present invention in detail with reference to the accompanying drawings of the specification.

[0041] Many specific details are set forth in the following description in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0042] Second, the "one embodiment" or "embodiment" referred to herein means a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that exclude each other with other embodiments.

[0043] Embodiment 1, referring to Figure 1 and Figure 2 , is the first embodiment of the present invention. This embodiment provides an intelligent management method for a solder paste library, including the following steps:

[0044] S1. Analyze the work order data of the MES system, synchronously collect the cold cabinet temperature, tank coordinates, and weight data, construct a dynamically mapped three-dimensional warehousing model, optimize the manipulator movement path using the quantum annealing algorithm, calculate multi-objective parameters, and output the optimal path instruction to the driving device.

[0045] Specifically, it includes the following steps:

[0046] Access the work order data through the MES system API, extract the work order number, material type, batch number, and priority, and output a standardized work order information table.

[0047] Connect the cold cabinet temperature sensor, tank RFID tag, and weighing device, and collect the temperature, coordinates, and weight data in real time, and output the device status data set.

[0048] Integrate the cold cabinet temperature, tank coordinates, and weight data, and output the device status data set.

[0049] Specifically, align the temperature, coordinate, and weight data according to the time stamp, generate a unified time series data set, generate a unique identifier (such as ID) for each tank, and associate it with the temperature, coordinate, and weight data.

[0050] Based on the warehousing layout, use a three-dimensional modeling tool to construct a space framework including cold cabinets, tanks, and manipulators, and output a three-dimensional model.

[0051] It should be noted that the warehousing layout refers to the planning and arrangement of the physical positions and spatial relationships of the internal equipment, materials, and functional areas in the warehouse.

[0052] Specifically, import the warehousing layout file (such as CAD drawing or JSON format) into a three-dimensional modeling tool (such as Blender or Unity 3D). Define the physical dimensions (length, width, height) and coordinate system of the warehouse in the tool. Initialize the 3D model templates of the cold cabinet, tank, and manipulator.

[0053] According to the layout file, place the cold cabinet model in the three-dimensional space and define its internal storage units and capacity.

[0054] Place the tank model, generate a unique identifier (such as ID) for each tank, and associate the initial coordinates and weight.

[0055] Place the robotic arm model, define its motion range and operation logic (such as grasping and placing the tank).

[0056] Output the 3D space frame model in a standard format (such as FBX or GLTF).

[0057] Dynamically map the device status dataset to the 3D model, label the tank status and the capacity of the freezer, and output the 3D warehousing model.

[0058] Specifically, read the device status dataset (including timestamp, tank ID, temperature, coordinates, weight).

[0059] Update the coordinates and weight of the tanks in the 3D model frame by frame according to the timestamp, and use interpolation method for smooth transition, expressed as:

[0060]

[0061] In the formula, P a and P b are the coordinates of the previous frame and the current frame respectively, t is the current time, t a represents the timestamp of the previous frame, and t b represents the timestamp of the current frame;

[0062] Update the freezer temperature data and dynamically adjust the color coding of the freezer model (such as green for normal and red for abnormal)

[0063] Furthermore, calculate the weight change rate of the tank and judge the status of the tank, expressed as:

[0064]

[0065] In the formula, W a is the previous weight, and W b is the current weight;

[0066] Label the normal and abnormal status of the tank according to the weight change rate: if the weight change rate is within the range of [-5%, +5%], it is judged as the normal status; if the weight change rate exceeds +5% or is lower than -5%, it is judged as the abnormal status.

[0067] Among them, the positive threshold: +5% (the maximum allowable weight increase ratio). The negative threshold: -5% (the maximum allowable weight decrease ratio). The positive threshold and the negative threshold are anti-misjudgment buffer intervals determined based on the solder paste volatilization characteristics, the accuracy of the weighing equipment (such as ±0.5%) and the production process tolerance.

[0068] Label the normal capacity and over-capacity of the freezer according to the freezer temperature: If the freezer temperature is within the closed interval of [2°C, 8°C], it is determined as the normal capacity; if the temperature is higher than 8°C or lower than 2°C, it is determined as the over-capacity.

[0069] Among them, the upper threshold: 8°C (the upper limit of the solder paste storage temperature, exceeding which will cause the activity of the flux to decrease). The lower threshold: 2°C (the lower limit of the solder paste storage temperature, below which crystallization may occur). The upper threshold and the lower threshold are determined by referring to the solder paste storage temperature requirements in the J-STD-004 standard and verified through accelerated aging experiments.

[0070] Export the dynamically mapped 3D model into an interactive format (such as WebGL or VR format) as a 3D warehousing model to support real-time monitoring and operation.

[0071] Adopt the quantum annealing algorithm to construct an objective function with the goal of minimizing the moving distance, energy consumption, and time.

[0072] Specifically, model the manipulator path optimization problem as a multi-objective optimization problem. By adjusting the path planning variable x, minimize the weighted sum of the path distance, energy consumption, and time. The objective function Z(x) is expressed as:

[0073] Z(x) = w 1 ·D(x) + w 2 ·E(x) + w 3 ·Q(x);

[0074] In the formula, D(x) is the total path distance, E(x) is the energy consumption, Q(x) is the total time, w 1 、w 2 and w 3 are weight coefficients, which are dynamically adjusted according to actual needs;

[0075] It should be noted that w 1 、w 2 and w 3 The weight coefficients are used to balance the priorities of the path distance, energy consumption, and time, and the value range is [0, 1], and they satisfy the normalization condition: W 1 + W 2 + W 3 = 1;

[0076] The path planning variable x is defined in the following way: Control point sequence: The manipulator movement path consists of a series of spatial coordinate points (such as (x, y, z)). Joint angle sequence: The manipulator movement path consists of the angle change sequence of each joint.

[0077] Generate the optimal path through annealing iteration and decompose it into a discrete set of movement instructions.

[0078] Map the manipulator path optimization problem to the Hamiltonian of the quantum annealing algorithm, and set the initial temperature T 0 and the annealing rate α, and initialize the quantum state;

[0079] Among them, the annealing rate α ranges from 0 < α < 1 and is used to control the speed of temperature decrease.

[0080] In each iteration, gradually reduce the temperature T, which is expressed as:

[0081] T k+1 = α·T k ;

[0082] In the formula, T k+1 represents the temperature value of the next iteration, and T k represents the temperature value of the current iteration;

[0083] Update the quantum state according to the quantum tunneling effect, generate a candidate path and calculate the objective function value f(x);

[0084] Use the Metropolis criterion to judge whether to accept the new path, which is expressed as:

[0085]

[0086] Δf = f(x new ) - f(x current );

[0087] In the formula, P(Δf) represents the probability of accepting the new path, which is used to judge whether to accept the current candidate path. Δf represents the change in the path objective function value, that is, the difference between the new path objective function value f(x new ) and the current path objective function value f(x current );

[0088] It should be noted that the Metropolis criterion is an acceptance probability criterion used in the simulated annealing algorithm (Simulated Annealing) to decide whether to accept a new solution that is worse than the current solution. Its core idea is: accept a worse solution with a certain probability to avoid the algorithm falling into a local optimal solution, thereby increasing the possibility of finding the global optimal solution.

[0089] Preferably, through the Metropolis criterion, the optimization algorithm can, while ensuring the search efficiency, significantly improve the success rate of finding the global optimal solution, providing a powerful solution for complex problems such as path planning and combinatorial optimization.

[0090] Record the current optimal path and its objective function value. When the temperature drops to the termination temperature T end , output the optimal path;

[0091] Decompose the optimal path into discrete movement instructions, including movement direction, speed, and target position, and use a dynamic programming algorithm to optimize the instruction set to ensure instruction coherence;

[0092] Extract a single instruction that matches the current task from the optimal path instruction set, dynamically adjust the instruction parameters, and then transmit them to the manipulator drive device for execution.

[0093] Extract a single optimal path instruction required for the current task, transmit it to the manipulator drive device, and the manipulator executes the instruction and feeds back the execution status to the system in real time;

[0094] Compare the actual path of the manipulator with the optimized path, and calculate the deviation value, which is expressed as:

[0095] ∈=||D 1 -D 2 ||

[0096] In the formula, ∈ represents the path deviation value, D 1 represents the total distance of the actual movement path of the manipulator, and D 2 represents the total distance of the optimal path calculated by the quantum annealing algorithm;

[0097] If the deviation value exceeds the path deviation threshold, trigger the iterative update of the path optimization based on the quantum annealing algorithm;

[0098] It should be noted that the path deviation threshold is 5% of the total path distance, and its calculation method is:

[0099] Path deviation threshold = total distance of the optimized path × 5%;

[0100] When ||actual total path distance - total distance of the optimized path|| > path deviation threshold, it is determined that the path deviation is excessive.

[0101] The basis for defining the path deviation threshold is as follows: based on the repeat positioning accuracy of the manipulator (usually ±0.1mm) and the storage layout size (assuming a 10m level); the 5% threshold can cover the transmission error of the manipulator (about 0.5%), the sensor error (about 0.2%), and the dynamic disturbance (about 4.3%); through Monte Carlo simulation verification, the 5% threshold can control the re-planning trigger frequency within a reasonable range of 3 - 5 times per day.

[0102] Update the path weight coefficient in the 3D storage model according to the actual path data, and extract a single optimal path instruction from the optimal path instruction set and transmit it to the manipulator drive device.

[0103] S2. Obtain the batch information of the incoming solder paste and weigh it, generate storage coordinates according to the remaining capacity in the refrigerated area in the 3D storage model and the FIFO rule, and drive the manipulator to store according to the optimized path.

[0104] Specifically, it includes the following steps:

[0105] Access the batch information of the incoming solder paste through the MES system API, extract the batch number, material type, and priority, combine with the weighing equipment to obtain the weight of the storage tank, and output a standardized incoming storage data table.

[0106] Based on the three-dimensional warehousing model, query the remaining capacity of the refrigerated area in real time, verify whether the current incoming storage tank meets the storage conditions, and output the capacity status of the refrigerated area.

[0107] According to the FIFO rule (first in, first out) and the remaining capacity of the refrigerated area, combine with the priority of the storage tank to calculate the optimal storage coordinates, and output the storage coordinate instruction.

[0108] Based on the optimal path instruction set generated by the quantum annealing algorithm, control the manipulator to execute the movement instructions in sequence, accurately transport the solder paste storage tank to the specified storage coordinates in the refrigerated area, and output the storage completion status.

[0109] Dynamically update the storage coordinates, weight, and status of the incoming storage tank to the three-dimensional warehousing model, mark the storage tank as "stored", and output the updated three-dimensional warehousing model.

[0110] S3. When the inventory in the rewarming area is lower than the preset lower inventory threshold, reverse calculate the work order demand time window, simulate the multi-tank rewarming conflict probability to generate a replenishment queue, extract the outgoing storage tank according to the optimal path instruction, dynamically adjust the revolution speed curve according to the simulation parameters in the three-dimensional warehousing model, and automatically correct the operating parameters when the torque deviation exceeds the limit.

[0111] Specifically, it includes the following steps:

[0112] The preset lower inventory threshold refers to the lowest inventory critical value set based on the capacity of the refrigerated area in the three-dimensional warehousing model, the reverse calculation result of the work order demand time window, and the real-time status of the rewarming area;

[0113] Specifically, obtain the total capacity of the refrigerated area (such as the number or weight of storage tanks that can be stored) from the three-dimensional warehousing model;

[0114] Use the total capacity minus the current inventory to calculate the remaining capacity of the refrigerated area in real time;

[0115] Extract the work order data from the MES system, including the work order demand time, demand quantity (such as the solder paste usage) and priority;

[0116] Based on the work order demand time, subtract the rewarming time from the work order demand time to reverse calculate the consumption time window of the rewarming area storage tank;

[0117] Statistically calculate the total demand within the estimated time window, and obtain the inventory in the warming area (such as the number or weight of storage tanks) in real time through sensors or system interfaces;

[0118] Calculate the consumption rate of the inventory in the warming area based on historical data. The formula is as follows:

[0119] Consumption rate = Historical consumption / Historical time interval;

[0120] Furthermore, calculate the preset inventory lower limit threshold. The formula is as follows:

[0121] Preset inventory lower limit threshold = Total demand × Safety factor;

[0122] When the work order demand, cold storage area capacity, or warming area status changes, dynamically update the preset inventory lower limit threshold.

[0123] Collect real-time torque, speed, and position data of the manipulator through sensors, and output operation status monitoring data.

[0124] Specifically, collect torque data, speed data, and position data in real time through sensors installed on the manipulator (such as torque sensors, encoders, position sensors, etc.);

[0125] Furthermore, filter the collected raw data to remove noise and outliers, and convert the data to standard units (such as torque unit is N·m, speed unit is m / s, position unit is mm);

[0126] Pack the processed data into operation status monitoring data and store it in the database.

[0127] Based on the simulation parameters in the three-dimensional warehousing model and historical operation status monitoring data, dynamically calculate the deviation range of the current torque, and output the dynamic deviation range value.

[0128] Specifically, collect the torque, speed, and position data of the manipulator through torque sensors, encoders, and position sensors in real time. After filtering and unit standardization processing, output the operation status monitoring data in JSON format and store it in the database.

[0129] Extract the load, path, and performance parameters of the manipulator from the three-dimensional warehousing model, combine with historical operation data (torque, speed, position), and dynamically calculate the reasonable deviation range based on the historical torque average value and standard deviation, output the dynamic deviation range value and synchronize it to the database.

[0130] When the real-time torque exceeds the dynamic deviation range value, automatically trigger the correction logic, and adjust the revolution speed of the manipulator in real time based on the simulation parameters, and output the corrected operation parameters.

[0131] Specifically, obtain the current torque value from the operation status monitoring data and compare it in real time with the dynamic deviation range value ([μ - 2σ, μ + 2σ]).

[0132] Extract the simulation parameters of the current task from the three-dimensional warehousing model;

[0133] Adjust the revolution speed V according to the torque deviation direction (too high or too low) new , and the formula is

[0134]

[0135] In the formula, ΔP is the torque deviation, that is, the difference between the current torque and the target torque, P max represents the maximum allowable value of the torque, P min represents the minimum allowable value of the torque, V b represents the current revolution speed;

[0136] If the corrected revolution speed V new exceeds the maximum allowable speed of the manipulator, it is forced to be set to the maximum speed;

[0137] If the speed is too low, resulting in a risk of task timeout, trigger path replanning (such as shortening the moving distance);

[0138] Among them, generating the corrected operation parameters includes the new revolution speed, the optimized path (if path replanning is triggered), and the expected torque value (the theoretical torque calculated based on the corrected speed);

[0139] Encapsulate the parameters into instructions, send them to the manipulator drive device in real time, and synchronously update the status of the manipulator in the three-dimensional warehousing model.

[0140] S4. Record the state of tin taking after distribution, verify the secondary weighing of the returned material tank, and update the material status in the three-dimensional warehousing model according to the comparison result between the weight change rate and the preset scrap weight threshold.

[0141] Specifically, it includes the following steps:

[0142] The preset scrap weight threshold refers to the lowest remaining weight critical value for determining whether the material tank fails, which is preset by calculating the weight change rate based on the secondary weighing result of the returned material tank;

[0143] Specifically, collect the secondary weighing data of historical returned material tanks, including the initial weight and the returned weight, and perform data cleaning and classification;

[0144] Calculate the weight change rate, which is expressed as:

[0145]

[0146] Combined with production demand and safety margin, a preset scrap weight threshold is set, expressed as:

[0147] Preset scrap weight threshold = initial weight × (failure critical percentage + safety margin percentage);

[0148] Verify the rationality of the preset scrap weight threshold and ensure its accuracy and reliability through historical data simulation and actual testing;

[0149] Dynamically adjust the preset scrap weight threshold, and regularly evaluate and optimize the preset scrap weight threshold according to changes in the production environment;

[0150] Compare the weight change rate with the preset scrap weight threshold, and update the tank labels, inventory parameters and location coordinates in the 3D warehouse model.

[0151] Record the status of tin retrieval after delivery and generate a retrieval log;

[0152] Perform secondary weighing verification on the returned material tanks and output the returned weight data;

[0153] Specifically, the returned material tanks are identified by barcode or RFID technology to obtain their unique identification information, and the weight of the returned material tanks is measured using weighing equipment to record the returned weight data;

[0154] The returned weight data is associated with the material tank identification information, the weight change rate is calculated, and compared with the preset scrap weight threshold. When the weight change rate is greater than the preset scrap weight threshold, the material scrap status mark is output; otherwise, the material available status mark is output;

[0155] Update the material status of the 3D warehouse model according to the material status identification.

[0156] Among them, the material status of the three-dimensional warehouse model includes tank labels, inventory parameters and location coordinates.

[0157] Specifically, the weight change rate is calculated based on the initial weight and the return weight, and the weight change rate is compared with the preset scrap weight threshold to determine whether the tank is invalid;

[0158] Based on the comparison results, a material status determination result (such as "normal" or "scrap") is generated.

[0159] S5. Compare the actual path of the robot with the robot optimization calculation, iteratively update the path weight coefficient in the three-dimensional warehousing model and synchronize the three-dimensional warehousing status.

[0160] Specifically, the following steps are included:

[0161] Through the interface between sensors and drive equipment, the actual movement trajectory, time and energy consumption data of the robot are collected in real time, and the actual path data set is output;

[0162] Specifically, communicate with the manipulator drive device interface through the EtherCAT bus protocol, synchronously collect the following trajectory data and energy consumption data, and synchronize the sensor clock using the IEEE 1588 Precision Time Protocol (PTP).

[0163] Combine the actual path data set and the optimal path instruction set, calculate the distance deviation, time error, and energy consumption difference through space-time alignment, and output the path deviation data set;

[0164] Specifically, establish the actual path timestamp t i and the optimized path instruction timestamp t' j The corresponding relationship is expressed as:

[0165] t′ j = t i + Δt latency ;

[0166] In the formula, Δt latency is the manipulator drive delay (measured and calibrated value), and t i is the actual path timestamp, representing the time point of a certain state of the manipulator in the actual motion path;

[0167] Perform cubic spline interpolation on the discrete coordinate points (x' j , y' j , z′ j ) in the optimized path instruction to generate a continuous trajectory function f opt (t);

[0168] The distance deviation is expressed as:

[0169]

[0170] In the formula, f act (t i ) represents the actual trajectory function, f opt (t i )| represents the optimized trajectory function, N represents the total number of sampling points, i represents the index variable of the sampling point, and |||| 2 represents the Euclidean distance calculation, unit: millimeter;

[0171] It should be noted that the actual trajectory function f act (t i) can be constructed by interpolating the sensor data collected by the robot during its actual operation. The specific steps are as follows: Data acquisition: Use sensors such as encoders, visual systems, or laser trackers to record the actual position coordinates (x, y, z) of the robot at different time points in real time. Data preprocessing: De-noise and filter the collected data to remove outliers and ensure the accuracy and smoothness of the data. Interpolation fitting: Use cubic spline interpolation (Cubic Spline Interpolation) or linear interpolation (Linear Interpolation) methods to fit the discrete sensor data into a continuous actual trajectory function f act (t i ). The interpolation function can accurately describe the actual position of the robot at any point in time. Verification and optimization: By comparing the interpolation results with the measured data, the accuracy of the function can be verified and optimized as needed.

[0172] Optimize trajectory function f opt (t i ) can be generated by a path planning algorithm. The specific steps are as follows: Path planning: Use path planning methods such as A* algorithm, RRT (rapidly expanding random tree) algorithm or Dijkstra algorithm to generate the optimal path from the starting point to the target point in the workspace of the manipulator. Path smoothing: Smooth the planned path (such as Bezier curve fitting or spline interpolation) to ensure the continuity and executableness of the path. Time parameterization: Associate the smoothed path with time to generate a time-parameterized optimized trajectory function f opt (t i ), describing the theoretical position of the manipulator at any point in time. Verification and optimization: Verify the feasibility of the optimized trajectory through simulation or experiment, and adjust it according to actual needs.

[0173] Time deviation, expressed as:

[0174]

[0175] In the formula, δ t is the mean time deviation, which represents the average difference between the actual execution time of the robot and the theoretical time of the optimized path, t act,i is the actual execution time of the robot, indicating the actual timestamp corresponding to the i-th sampling point, t opt,i is the theoretical time of the optimized path, which represents the theoretical timestamp corresponding to the i-th sampling point, |t act,i -t opt,i | represents the absolute deviation between the actual time and the theoretical time of the i-th sampling point;

[0176] Energy consumption deviation, expressed as:

[0177]

[0178] In the formula, δ e is the percentage of energy consumption deviation, representing the relative difference between the actual energy consumption of the manipulator and the theoretical energy consumption of the optimized path. E act is the actual energy consumption of the manipulator, representing the total energy consumed by the manipulator during the task execution. E opt is the theoretical energy consumption of the optimized path, representing the energy that the manipulator should consume when executing the task under ideal conditions. E act -E opt represents the absolute difference between the actual energy consumption and the theoretical energy consumption; is the relative value of the energy consumption deviation, representing the proportion of the difference between the actual energy consumption and the theoretical energy consumption in the theoretical energy consumption;

[0179] Based on the path deviation dataset, dynamically adjust the path weight coefficients (such as distance weight, energy consumption weight, priority weight) in the three-dimensional warehousing model, output the updated path weight parameters, and map them to the three-dimensional warehousing model;

[0180] Specifically, for the weight update rule, update the weight coefficient w according to the gradient of the objective function new , which is expressed as:

[0181]

[0182]

[0183] In the formula, w old represents the weight value before update, η is the learning rate (η = 0.01), represents the partial derivative (gradient) of the loss function L with respect to the weight w;

[0184] Output the updated weight parameters and inject them into the quantum annealing algorithm to reconstruct the path optimization objective;

[0185] Recalculate the dynamic path planning logic in the three-dimensional warehousing model and output the synchronized three-dimensional warehousing state.

[0186] Specifically, generate the Hamiltonian based on the new weight, re-solve the optimal path instruction set, and broadcast the updated warehousing model parameters to all subsystems through the OPC UA protocol;

[0187] The status parameters such as the cold cabinet capacity and the coordinates of the storage tank are refreshed in real time, with a delay < 50ms.

[0188] This embodiment also provides an intelligent management system for a solder paste library, including: a data parsing module, a path optimization module, a warehousing management module, a status update module, and a model iteration module; The data parsing module is used to parse the work order data of the MES system, synchronously collect the cold cabinet temperature, the coordinates and weight data of the material tanks, construct a dynamically mapped three-dimensional warehousing model, optimize the movement path of the manipulator using the quantum annealing algorithm, calculate multi-objective parameters, and output the optimal path instruction to the driving device; The path optimization module is used to obtain the batch information of the incoming solder paste and weigh it, generate storage coordinates according to the remaining capacity in the refrigerated area in the three-dimensional warehousing model and the FIFO rule, and drive the manipulator to store according to the optimized path; The warehousing management module is used to, when the inventory in the warming area is lower than the preset inventory lower limit threshold, inversely calculate the work order demand time window, generate a replenishment queue after simulating the multi-tank warming conflict probability, extract the outgoing material tank according to the optimal path instruction, dynamically adjust the revolution speed curve according to the simulation parameters in the three-dimensional warehousing model, and automatically correct it when the torque deviation exceeds the limit; The status update module is used to record the solder usage status after distribution, weigh the returned material tank again for verification, and update the material status of the three-dimensional warehousing model according to the comparison result between the weight change rate and the preset scrap weight threshold; The model iteration module is used to compare the actual path of the manipulator with the optimized calculation of the manipulator, iteratively update the path weight coefficient in the three-dimensional warehousing model, and synchronize the three-dimensional warehousing status.

[0189] This embodiment also provides a computer device, which is applicable to the situation of the intelligent management method of the solder paste library, including: a memory and a processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the intelligent management method of the solder paste library as proposed in the above embodiment.

[0190] This computer device can be a terminal. This computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of this computer device is used to provide computing and control capabilities. The memory of this computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of this computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of this computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of this computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the computer device housing, or an external keyboard, touchpad, or mouse, etc.

[0191] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the intelligent management method for the solder paste library as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM for short), electrically erasable programmable read-only memory (EEPROM for short), erasable programmable read-only memory (EPROM for short), programmable read-only memory (PROM for short), read-only memory (ROM for short), magnetic memory, flash memory, magnetic disk or optical disc.

[0192] In summary, the present invention reconstructs the manipulator path planning through the quantum annealing algorithm, fuses the moving distance, energy consumption and time parameters to establish a multi-objective optimization function, and utilizes the characteristics of quantum computing to break through the search bottleneck of traditional algorithms in the high-dimensional solution space, realizing real-time path generation and correction in a dynamic storage environment. Based on the temperature gradient distribution of the freezer and the position weight of the material tank, a three-dimensional dynamic storage model is constructed, the space utilization rate of the refrigerated area is optimized through coordinate mapping, and the manipulator positioning error is eliminated by combining the space-time compensation algorithm. For material management, a reverse parsing mechanism for the work order time sequence is designed to predict replenishment conflicts and generate a priority queue, and the torque dynamic compensation logic is synchronously integrated to cope with sudden load fluctuations. In quality control, a dynamic scrap determination rule is established by comparing the secondary weighing of the returned material tank with the historical data trend, and at the same time, the self-adaptability of the storage model is optimized by using the iterative feedback of the path execution data, and finally a closed-loop control system of environment perception-intelligent decision-making-autonomous correction is formed.

[0193] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. An intelligent management method for a solder paste warehouse, characterized in that: include, Analyze the work order data of the MES system, synchronously collect the freezer temperature, material tank coordinates and weight data, build a dynamic mapping three-dimensional warehouse model, use the quantum annealing algorithm to optimize the robot's movement path, calculate multi-objective parameters, and output the optimal path instructions to the drive equipment; Obtain the batch information of solder paste entering the warehouse and weigh it, generate storage coordinates based on the remaining capacity of the cold storage area and FIFO rules in the 3D storage model, and drive the robot to store according to the optimized path; When the inventory in the reheating zone is lower than the preset inventory lower limit, the work order demand time window is reversely calculated, and the replenishment queue is generated after simulating the probability of multi-tank reheating conflict. The outbound material tanks are extracted according to the optimal path instructions, and the revolution speed curve is dynamically adjusted according to the simulation parameters in the three-dimensional warehouse model. The operating parameters are automatically corrected when the torque deviation exceeds the limit. Record the status of tin use after delivery, re-weigh and verify the returned material tanks, and update the material status of the 3D storage model based on the comparison results of the weight change rate and the preset scrap weight threshold; Compare the actual path of the robot with the robot optimization calculation, iteratively update the path weight coefficient in the three-dimensional warehousing model and synchronize the three-dimensional warehousing status.

2. The intelligent management method of solder paste warehouse according to claim 1, characterized in that: Integrate freezer temperature, tank coordinates and weight data to output equipment status data set; Based on the warehouse layout, use 3D modeling tools to build a spatial framework including refrigerators, tanks and manipulators, and output a 3D model; Dynamically map the equipment status data set to the 3D model, annotate the tank status and freezer capacity, and output the 3D warehouse model.

3. The intelligent management method of solder paste warehouse according to claim 2, characterized in that: The specific steps of outputting the optimal path instruction to the driving device are as follows: The quantum annealing algorithm is used to construct the objective function with the goal of minimizing the moving distance, energy consumption and time; Generate the optimal path through annealing iteration and decompose it into a discrete optimal path instruction set; A single instruction matching the current task is extracted from the optimal path instruction set, and the instruction parameters are dynamically adjusted before being transmitted to the robot drive device for execution.

4. The intelligent management method of solder paste warehouse according to claim 3, characterized in that: Driving the manipulator to store according to the optimized path refers to controlling the manipulator to execute movement instructions in sequence based on the optimal path instruction set generated by the quantum annealing algorithm to transport the solder paste cans to the storage coordinates of the cold storage area specified in the three-dimensional storage model.

5. The intelligent management method of solder paste warehouse according to claim 4, characterized in that: The preset inventory lower limit threshold refers to the minimum inventory critical value set based on the capacity of the cold storage area in the three-dimensional warehouse model, the reverse calculation result of the work order demand time window and the real-time status of the warming area; The specific steps of automatically correcting the operating parameters when the torque deviation exceeds the limit are as follows: Collect the real-time torque, speed and position data of the robot and output the operation status monitoring data; Based on the simulation parameters in the three-dimensional warehouse model and the historical operation status monitoring data, the deviation range of the current torque is dynamically calculated and the dynamic deviation range value is output; When the real-time torque exceeds the dynamic deviation range, the correction logic is automatically triggered, the robot revolution speed is adjusted in real time based on the simulation parameters, and the corrected operating parameters are output.

6. The intelligent management method of solder paste warehouse according to claim 5, characterized in that: The preset scrap weight threshold refers to the minimum remaining weight critical value for determining whether the tank is invalid, which is preset based on the secondary weighing result of the returned tank by calculating the weight change rate; The specific steps of updating the material status of the three-dimensional warehouse model are as follows: Record the status of tin retrieval after delivery and generate a retrieval log; Perform secondary weighing verification on the returned material tanks and output the returned weight data; Calculate the weight change rate and compare it with the preset scrap weight threshold. When the weight change rate is greater than the preset scrap weight threshold, output the material scrap status mark; otherwise, output the material available status mark; Update the material status of the 3D warehouse model according to the material status identification.

7. The intelligent management method of solder paste warehouse according to claim 6, characterized in that: The specific steps of iteratively updating the path weight coefficient in the three-dimensional storage model and synchronizing the three-dimensional storage state are as follows: Collect the actual movement trajectory, time and energy consumption data of the robot in real time, and output the actual path data set; Combine the actual path data set with the optimal path instruction set, calculate the distance deviation, time error and energy consumption difference through time-space alignment, and output the path deviation data set; Based on the path deviation data set, the path weight coefficient in the three-dimensional warehousing model is dynamically adjusted, the updated path weight parameters are output, mapped to the three-dimensional warehousing model, the dynamic path planning logic in the three-dimensional warehousing model is recalculated, and the synchronized three-dimensional warehousing status is output.

8. An intelligent management system for a solder paste warehouse, based on the intelligent management method for a solder paste warehouse according to any one of claims 1 to 7, characterized in that: Including data analysis module, path optimization module, warehouse management module, status update module and model iteration module; The data analysis module is used to analyze the work order data of the MES system, synchronously collect the freezer temperature, material tank coordinates and weight data, build a dynamic mapping three-dimensional storage model, use the quantum annealing algorithm to optimize the manipulator movement path, calculate multi-objective parameters, and output the optimal path instruction to the drive device; The path optimization module is used to obtain batch information of solder paste entering the warehouse and weigh it, generate storage coordinates according to the remaining capacity of the cold storage area and the FIFO rule in the three-dimensional storage model, and drive the manipulator to store according to the optimized path; The warehouse management module is used to reversely calculate the work order demand time window when the inventory in the reheating zone is lower than the preset inventory lower limit threshold, generate a replenishment queue after simulating the probability of multi-tank reheating conflict, extract the outbound material tanks according to the optimal path instruction, dynamically adjust the revolution speed curve according to the simulation parameters in the three-dimensional warehouse model, and automatically correct it when the torque deviation exceeds the limit; The status update module is used to record the status of tin use after delivery, perform secondary weighing verification on the returned material tanks, and update the material status of the three-dimensional storage model based on the comparison result of the weight change rate and the preset scrap weight threshold; The model iteration module is used to compare the actual path of the manipulator with the manipulator optimization calculation, iteratively update the path weight coefficient in the three-dimensional warehousing model and synchronize the three-dimensional warehousing status.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the intelligent management method of the solder paste library described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent management method of the solder paste warehouse described in any one of claims 1 to 7 are implemented.

Images