Method and system for intelligent storage of electroplating hangers

Abstract

The embodiment provides a method and system for intelligent storage of electroplating hangers, the method comprising: establishing a three-dimensional space right-angle coordinate positioning system, and positioning and calibrating the electroplating hangers in the positioning system; and controlling a mechanical hand to take and place the electroplating hangers, comprising: adopting a linear control system with state feedback, when the distance between the mechanical hand and a target position is greater than a preset distance threshold, controlling the mechanical hand to adopt a first control gain to perform rapid movement control, when the distance between the mechanical hand and the target position is less than or equal to the preset distance threshold, controlling the mechanical hand to adopt a second control gain to perform accurate positioning control, and compensating inertia error and mechanical friction in the movement process of the mechanical hand in real time; and adopting a distributed data storage architecture to store position information, movement states and work records of the electroplating hangers, so as to realize multidimensional query and statistical analysis of the storage information of the electroplating hangers.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent manufacturing, in particular to the field of electroplating production line automation, Internet of Things and intelligent control, and specifically relates to a method and system for intelligent storage of electroplating hangers. BACKGROUND

[0002] The hanger storage management in the electroplating production line is an important link in the electroplating manufacturing process. In the process of implementing the present application, it is found that the existing technology has the following problems in the storage management of electroplating hangers:

[0003] Traditional electroplating hanger storage management mainly relies on manual operation, which is not only inefficient but also prone to errors, affecting production efficiency. Although some automated storage systems can achieve basic storage and retrieval functions, they lack real-time positioning and intelligent scheduling functions, and cannot accurately manage electroplating hangers, affecting the efficiency of the storage system. The existing storage systems generally have low system integration and difficulty in real-time data tracing, making it difficult to achieve intelligent management of electroplating hangers throughout the entire process.

[0004] The existing technology has the technical problems of low automation and intelligence, insufficient positioning and scheduling capabilities, and low system integration of electroplating hanger storage systems. SUMMARY

[0005] In view of this, the present application provides a method and system for intelligent storage of electroplating hangers, which solves the technical problems of low automation and intelligence, insufficient positioning and scheduling capabilities, and low system integration of electroplating hanger storage systems in the prior art.

[0006] The present application provides a method for intelligent storage of electroplating hangers, comprising:

[0007] Establishing a three-dimensional space rectangular coordinate positioning system, and positioning and calibrating the electroplating hangers in the positioning system;

[0008] Controlling the robot to take and place the electroplating hangers, comprising: using a linear control system with state feedback, when the distance between the robot and the target position is greater than a preset distance threshold, controlling the robot to use a first control gain for fast movement control, and when the distance between the robot and the target position is less than or equal to the preset distance threshold, controlling the robot to use a second control gain for precise positioning control, and real-time compensating for inertia errors and mechanical friction in the movement of the robot; wherein, during the electroplating hanger taking and placing operation, a state transition model based on Markov chain is used to optimize the path planning of the electroplating hanger taking and placing;

[0009] The position information, motion state and work record of the electroplating hanger are stored by using a distributed data storage architecture to realize multi-dimensional query and statistical analysis of the electroplating hanger storage information.

[0010] Optionally, the path planning of the electroplating hanger taking and placing is optimized using a state transition model based on Markov chain, including:

[0011] Collecting historical work data of the electroplating hanger to train a state transition probability matrix;

[0012] Predicting future taking and placing requirements according to the state transition probability matrix;

[0013] Using a dynamic programming algorithm to calculate a scheduling path that satisfies the shortest path and shortest waiting time constraints.

[0014] Optionally, before optimizing the path planning of the electroplating hanger taking and placing using a state transition model based on Markov chain, the method further includes:

[0015] Establishing a state space describing the position and state change of the electroplating hanger;

[0016] Calculating the transition probability of the state space based on the historical work data;

[0017] Introducing robustness constraints in the dynamic programming algorithm to ensure scheduling reliability under uncertain conditions;

[0018] Distributing the path planning task to multiple computing nodes for parallel processing;

[0019] Real-time updating the state transition probability matrix.

[0020] Optionally, the method further includes:

[0021] Establishing a digital demand management system, including:

[0022] Establishing a structural view, a behavioral view, a demand view and a parameter view describing the electroplating hanger intelligent storage system;

[0023] Decomposing the functional requirements and performance requirements of the electroplating hanger intelligent storage system into multiple technical indicators through a demand decomposition matrix.

[0024] Optionally, the digital demand management system is constructed based on the SysML meta-model derived from INCOSE, and the decomposition of the functional requirements and performance requirements of the electroplating hanger intelligent storage system into multiple technical indicators through the demand decomposition matrix includes:

[0025] Using a SysML requirement diagram to describe the functional and performance requirements of the electroplating hanger intelligent storage system;

[0026] A module definition diagram is used to establish a structural model of the electroplating rack intelligent storage system, and structural technical indexes are determined;

[0027] The internal block diagram is used to define the connection relationship inside the electroplating rack intelligent storage system, and connection technical indexes are determined;

[0028] The activity diagram and the state machine diagram are used to describe the dynamic behavior of the electroplating rack intelligent storage system, and dynamic behavior technical indexes are determined;

[0029] A bidirectional traceability relationship between requirements and design elements is established.

[0030] Optionally, a three-dimensional space rectangular coordinate positioning system is established, and the electroplating rack is positioned and calibrated in the positioning system, including:

[0031] A laser ranging sensor with an accuracy of ±1 mm is installed at a key node position of the storage rack, and the laser ranging sensor forms a sensor network matrix;

[0032] Each electroplating rack is equipped with a source RFID tag, and the RFID tag stores identification information including a unique identification code of the electroplating rack, a type of the electroplating rack, and a load parameter;

[0033] A plurality of parallel laser beam array emitters are used to collect reflection signals of the electroplating rack in real time;

[0034] The distance between the electroplating rack and a fixed reference point is calculated based on a time-of-flight ranging principle, and the distance is calibrated in position with a preset fixed calibration point.

[0035] Optionally, the linear control system is a predictor-feedback control system with a transient switch, and the method further includes:

[0036] The transient switch is set in the state feedback link, and is used to dynamically switch the first control gain and the second control gain;

[0037] The predictor model is used to calculate the motion trajectory of the mechanical hand in advance;

[0038] A feedback correction mechanism is established according to the actual position of the mechanical hand;

[0039] Based on the weighting of the predictor calculation result and the feedback signal, a control instruction of the mechanical hand is generated.

[0040] Optionally, the method further includes:

[0041] A smooth transition operation is performed during the switching process of the transient switch to avoid a jump in control quantity;

[0042] Adaptively adjusting the predictor model parameters according to the robot load;

[0043] Adaptive weight distribution of predictive control and feedback control is realized, and control quantity limiting and change rate constraints are set.

[0044] Optionally, the position information, motion state and job record of the electroplating rack are stored by using a distributed data storage architecture, including:

[0045] A master data node and a plurality of slave data nodes of the distributed data storage architecture are established, the master data node is responsible for data writing and distribution, and the slave data nodes are responsible for data backup and query;

[0046] A time series database is used to store the position information and motion state of the rack in the master data node and the slave data nodes;

[0047] A relational database is used to store the job record of the rack in the master data node and the slave data nodes;

[0048] Data is synchronized between the master data node and the slave data nodes based on a data consistency protocol;

[0049] A data sharding strategy is set, and the rack information is stored in a data partition according to time dimension and space dimension.

[0050] The embodiment of the application also provides an electroplating rack intelligent warehousing system for realizing the above method, including:

[0051] The multi-layer three-dimensional warehousing frame includes a plurality of shelf units, and each shelf unit is connected through a standard interface;

[0052] The automated transmission system includes a six-degree-of-freedom robot and a conveyor belt with an encoder, the robot is connected to the control system through a control bus, and the conveyor belt is connected to the control system through an industrial Ethernet;

[0053] The identification system includes a laser scanner with an accuracy of ±1mm and an RFID reader and writer supporting EPC Gen2 protocol;

[0054] The distributed control system realizes multi-task cooperative control by using a master-slave architecture.

[0055] The embodiment of the application has the following technical effects:

[0056] The electroplating rack is accurately positioned by establishing a three-dimensional space right-angle coordinate positioning system, and the robot is controlled with high precision by using a predictor-feedback control system with transient switching, thereby significantly improving the automation degree and positioning accuracy of the warehousing system.

[0057] The path planning is realized by a state transition model based on Markov chain, the intelligent scheduling of the warehouse system is realized, and the efficiency of taking and placing the electroplating hanger is greatly improved.

[0058] The collection and management of the electroplating hanger life cycle data are realized by a distributed data storage architecture, and the integration and reliability of the system are improved. BRIEF DESCRIPTION OF DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are only embodiments of the present application, and other drawings can also be obtained according to the provided drawings without creative labor for those skilled in the art.

[0060] Figure 1 is a flowchart of the electroplating hanger intelligent warehouse method provided by the embodiments of the present application;

[0061] Figure 2 is a structural diagram of a three-dimensional space rectangular coordinate positioning system provided by the embodiments of the present application;

[0062] Figure 3 is a principle block diagram of a predictor-feedback control system provided by the embodiments of the present application;

[0063] Figure 4 is a state transition model based on Markov chain provided by the embodiments of the present application;

[0064] Figure 5 is a distributed data storage architecture provided by the embodiments of the present application;

[0065] Figure 6 is a structural diagram of an electroplating hanger intelligent warehouse system provided by the embodiments of the present application;

[0066] Figure 7 is a practical structure diagram of a factory containing an intelligent warehouse system provided by the embodiments of the present application. DETAILED DESCRIPTION

[0067] In order to facilitate the description, the terms in the embodiments of the present application can be explained here. "Electroplating hanger" refers to a tooling fixture used to carry and fix the workpiece to be electroplated in the electroplating production line. "Predictor-feedback control" refers to a control system with both feedforward prediction and feedback correction functions. "Instantaneous switch" refers to a mechanism capable of realizing rapid switching of control parameters. "Markov chain" refers to a probability model based on predicting future states from current states. "SysML" refers to system modeling language, which is a modeling language for system engineering.

[0068] Referring toFigure 1 The electroplating hanger intelligent storage method provided by the embodiments of the present application comprises three main steps: establishing a three-dimensional space rectangular coordinate positioning system and performing positioning calibration, controlling a mechanical hand to take and place an electroplating hanger, and using a distributed data storage architecture to perform data management.

[0069] Specifically, the method comprises:

[0070] S1. Establishing a three-dimensional space rectangular coordinate positioning system, and performing positioning calibration on an electroplating hanger in the positioning system;

[0071] S2. Controlling a mechanical hand to take and place an electroplating hanger, comprising: using a linear control system with state feedback, when the distance between the mechanical hand and the target position is greater than a preset distance threshold, controlling the mechanical hand to use a first control gain for rapid movement control, when the distance between the mechanical hand and the target position is less than or equal to the preset distance threshold, controlling the mechanical hand to use a second control gain for precise positioning control, and compensating for inertia error and mechanical friction in the movement of the mechanical hand in real time; wherein, during the taking and placing operation of the electroplating hanger, a state transition model based on Markov chain is used to optimize the path planning of the taking and placing of the electroplating hanger;

[0072] S3. Using a distributed data storage architecture to store the position information, motion state and operation record of the electroplating hanger, so as to realize multidimensional query and statistical analysis of the storage information of the electroplating hanger.

[0073] The specific implementation process and mutual relationship of the three main steps are described in detail below.

[0074] In the first main step, the embodiments of the present application first establish a three-dimensional space rectangular coordinate positioning system, which is used to realize the precise positioning of the electroplating hanger in the storage space. Specifically, the system installs laser ranging sensors with an accuracy of ±1mm at key node positions such as the four corners and the center of the storage rack, and these sensors are connected through an industrial bus to form a sensor network matrix. At the same time, the system is equipped with a source RFID tag on each electroplating hanger, which is used to store identification information including the unique identification code of the electroplating hanger, the type of the electroplating hanger and the load parameter. The RFID tag is used in cooperation with the laser ranging sensor to realize the unique identification and spatial positioning of the electroplating hanger.

[0075] In the second main step, the embodiments of the present application control the robot to carry out the pick-and-place operation of the electroplating hanger. The system adopts a linear control system with state feedback, dynamically adjusting the control strategy according to the real-time position during the movement of the robot. When the robot is more than a preset distance threshold away from the target position, a first control gain is used to achieve rapid approach; when the distance is less than or equal to the preset threshold, a second control gain is switched to achieve precise positioning. In this process, the system uses a state transition model based on Markov chain to optimize the pick-and-place path of the electroplating hanger, predicts future pick-and-place requirements by collecting and analyzing historical operation data, and calculates the optimal scheduling path.

[0076] In the third main step, the embodiments of the present application use a distributed data storage architecture to manage the information of the electroplating hanger. The architecture includes a master data node and multiple slave data nodes, which are responsible for data writing and distribution and backup query, respectively. Among them, the system uses a time series database to store the position information and motion state of the electroplating hanger, and uses a relational database to store the operation record. Through this distributed architecture, the system realizes reliable storage and efficient query of the electroplating hanger's full life cycle data.

[0077] It should be noted that there is a close logical relationship between the above three main steps. First, the three-dimensional space rectangular coordinate positioning system provides position feedback information for the precise control of the robot. Second, the motion control process of the robot generates a large amount of real-time data, which is managed through the distributed storage architecture and provides data support for subsequent path optimization. In addition, the historical data accumulated in the distributed data storage architecture can be used to optimize the calibration parameters of the positioning system, thus forming a closed-loop optimization process.

[0078] Referring to Figure 2 , the embodiments of the present application detail the specific steps of establishing a three-dimensional space rectangular coordinate positioning system. The system installs laser ranging sensors with an accuracy of ±1mm at key node positions of the storage rack, and these sensors form a sensor network matrix. In addition, the system is equipped with a source RFID tag on each electroplating hanger, which is used to store the unique identification code, electroplating hanger type, and load parameter identification information of the electroplating hanger.

[0079] During the positioning calibration process, the system uses multiple parallel laser beam array emitters to collect the reflection signals of the electroplating hanger in real time. It should be noted that the system calculates the distance between the electroplating hanger and the fixed reference point based on the time-of-flight ranging principle. In addition, the system will calibrate the position of the calculated distance with the preset fixed calibration point, thereby achieving precise positioning of the electroplating hanger.

[0080] The embodiment of the application describes the process of controlling the robot to take and place the plating hanger. The system adopts a linear control system with state feedback. The core feature of the control system is to dynamically adjust the control strategy according to the distance between the robot and the target position. Specifically, when the distance between the robot and the target position is greater than a preset distance threshold, the system controls the robot to use a first control gain for fast movement control; when the distance is less than or equal to the preset distance threshold, the second control gain is switched for precise positioning control. In this process, the system compensates for the inertia error and mechanical friction in the movement of the robot in real time.

[0081] Referring to Figure 3 , the embodiment of the application describes in detail the process of controlling the robot to take and place the plating hanger. The control process of the robot mainly involves three aspects: the basic architecture of the control system, the dynamic adjustment of the control gain, and the real-time compensation mechanism.

[0082] First, the system adopts a linear control system with state feedback as the basic control architecture. As shown in Figure 3 , the control system includes a state feedback link, a feedforward control link, and a compensation control link. The state feedback link collects the position, speed, and acceleration information of the robot in real time through a high-precision encoder to form a complete state vector. The feedforward control link calculates the theoretical control quantity based on the kinematics model, and the compensation control link corrects it according to the actual motion state. The outputs of the three control links are combined after weighting to generate the final control instruction.

[0083] Second, the core feature of the control system is to dynamically adjust the control strategy according to the distance between the robot and the target position. The system first obtains the Euclidean distance between the current position of the robot and the target position, and compares it with the preset distance threshold (usually set to 100mm). According to the comparison result, the system adopts different control strategies:

[0084] When the distance between the robot and the target position is greater than the preset distance threshold, the system controls the robot to use a first control gain for fast movement control. The first control gain is usually set between 0.8-0.9. In this stage, the control system mainly focuses on the fast response performance of the robot, so that it can approach the target position at a faster speed. At the same time, the system sets the corresponding acceleration limit according to the motion speed, and the typical value is 2m / s 2 , to prevent excessive inertia impact.

[0085] When the distance is less than or equal to the preset distance threshold, the system switches to the second control gain for precise positioning control. The second control gain is usually set between 0.1-0.2, and in this stage, the control system mainly focuses on the positioning accuracy of the manipulator, reduces the response speed of the system through a smaller control gain, and reduces the overshoot. At this time, the system will also correspondingly reduce the speed and acceleration limit, and the typical values are 0.1 m / s and 0.2 m / s respectively 2 .

[0086] In addition, during the control process, the system compensates for the inertia error and mechanical friction during the movement of the manipulator in real time. The inertia error compensation mainly aims at the dynamic error generated by the manipulator during high-speed movement and sharp speed change. The system calculates the inertia moment of each joint by establishing the dynamics model of the manipulator, and adds the corresponding compensation term in the control quantity. Typical compensation strategies include:

[0087] 1) Acceleration feedforward compensation: calculate inertia compensation based on the acceleration of the desired trajectory

[0088] F_inertia=M*a_desired

[0089] Where M is the equivalent mass matrix of the manipulator, and a_desired is the desired acceleration.

[0090] 2) Coriolis force and centrifugal force compensation: for additional dynamic forces generated by multi-joint coupled motion

[0091]

[0092] Where C is the Coriolis force coefficient matrix, q and q· are the joint position and velocity vectors respectively.

[0093] In addition, the mechanical friction compensation mainly includes Coulomb friction compensation and viscous friction compensation:

[0094] 1) Coulomb friction compensation uses a piecewise continuous function:

[0095] F_coulomb=Fc*sign(v),|v|>v_threshold

[0096] F_coulomb=Fc*(v / v_threshold),|v|≤v_threshold

[0097] Where Fc is the Coulomb friction coefficient, v is the movement speed, and v_threshold is the speed threshold.

[0098] 2) Viscous friction compensation is proportional to speed:

[0099] F_viscous=Fv*v

[0100] where Fvis the viscous friction coefficient.

[0101] It should be noted that, in order to prevent the control amount jump generated when the control gain is switched, the system designs a smooth transition mechanism. Specifically, the sigmoid function is used to realize the gradual change of the control gain:

[0102] K = K1 + (K2 - K1) * (1 / (1 + e^(-a (d - d_threshold)))

[0103] where K1 is the first control gain, K2 is the second control gain, d is the current distance, d_threshold is the distance threshold, and a is the adjustment coefficient.

[0104] Through the above control strategy, the embodiment of the application realizes the unification of fast response and precise positioning in the movement process of the manipulator, while ensuring the smoothness of the movement process. Tests show that the maximum speed of the control system in the fast moving stage can reach 1 m / s, and the positioning accuracy in the precise positioning stage is better than ±0.1 mm.

[0105] It should be noted that the linear control system adopts a predictor-feedback control structure. Referring to Figure 3 , a transient switch is provided in the state feedback link, which is used to dynamically switch the first control gain and the second control gain. The system uses a predictor model to calculate the motion trajectory of the manipulator in advance, and at the same time establishes a feedback correction mechanism according to the actual position of the manipulator. Finally, the system generates the control command of the manipulator based on the weighted sum of the predictor calculation result and the feedback signal.

[0106] Specifically, the predictor model adopts a feedforward predictive control method based on the dynamics equation of the manipulator, which realizes accurate prediction of the motion trajectory of the manipulator by establishing the kinematics and dynamics models of the manipulator. The predictor model includes the following key components:

[0107] Firstly, the system establishes a kinematics model of the manipulator based on Denavit-Hartenberg parameters. Through this model, the Cartesian space position of the end effector of the manipulator can be corresponded to the joint angle:

[0108] X = f (θ)

[0109] where X is the position and attitude vector of the end effector in the Cartesian space, θ is the joint angle vector, and f(·) is the forward kinematics equation.

[0110] Secondly, the system establishes a dynamics model considering the influence of joint inertia, Coriolis force and gravity:

[0111]

[0112] where M(0) is the inertia matrix, C(0, 0) is the Coriolis and centrifugal force term, G(0) is the gravity term, and T is the joint driving torque.

[0113] Based on the above model, the predictor adopts a recursive algorithm to predict the manipulator state in the next N sampling periods:

[0114] 1. First, according to the current position p(k) and the target position p_target, a quintic polynomial interpolation is used to generate the expected trajectory:

[0115] p_d(t) = a0 + a1t + a2t 2 + a3t 3 + a4t 4 + a5t 5

[0116] where the coefficients a0-a5 are solved by the boundary conditions of position, velocity, and acceleration.

[0117] 2. Then, for each prediction time k+i (i = 1, 2,..., N), the following steps are used to calculate the predicted state:

[0118] Solve the expected joint angle 0_d(k+i) through inverse kinematics;

[0119] Substitute into the dynamics equation to calculate the required driving torque T(k+i);

[0120] Consider the actuator dynamic characteristics, predict the actual joint motion response 0(k+i);

[0121] Calculate the predicted position p(k+i) through forward kinematics;

[0122] 3. In addition, the predictor also considers the following factors to improve the prediction accuracy:

[0123] Introduce state estimation based on Kalman filtering, online identification and update of model parameters;

[0124] Consider the nonlinear characteristics of mechanical friction, use LuGre friction model for compensation;

[0125] Add a load mass adaptive estimation mechanism to adjust the dynamics model parameters in real time;

[0126] 4. In order to improve the calculation efficiency, the system adopts a rolling horizon optimization strategy:

[0127] Only calculate the prediction value of the next N sampling points in each control period;

[0128] Use parallel computing architecture to calculate the state of multiple prediction times simultaneously;

[0129] An adaptive adjustment mechanism of prediction length N and control period is set up;

[0130] Through the above predictor model, the system can predict the motion trajectory of the manipulator 0.1 seconds in advance, with a prediction error within ±0.5 mm, providing effective support for high-precision motion control. The predictor model works in cooperation with the state feedback controller, significantly improving the dynamic tracking performance and positioning accuracy of the system.

[0131] In the specific implementation of transient switching, the embodiments of the present application perform smooth transition operation during switching to avoid jumps in control variables. At the same time, the system adaptively adjusts the predictor model parameters according to the load of the manipulator. In addition, the system realizes adaptive weight distribution of predictive control and feedback control, and sets control variable amplitude limiting and rate of change constraints to ensure the stability of the control system.

[0132] Under the cooperative control framework of predictive control and feedback control, the system uses an adaptive weight distribution mechanism to balance the effects of the two control strategies. The specific implementation scheme includes the following key aspects:

[0133] Firstly, the system establishes a hybrid control law of predictive control and feedback control:

[0134] u(k) = λ(k)u_pred(k) + (1 - λ(k))u_fb(k)

[0135] where u(k) is the final control variable, u_pred(k) is the predictive control variable, u_fb(k) is the feedback control variable, and λ(k) is the adaptive weight coefficient.

[0136] Secondly, the system designs an adaptive weight adjustment mechanism based on multi-objective evaluation:

[0137] 1. The weight adjustment uses the following dynamic equation:

[0138] λ(k) = λ(k-1) + η[J_pred(k) - J_fb(k)]

[0139] where η is the learning rate (typical value is 0.01), and J_pred and J_fb are the performance indicators of predictive control and feedback control, respectively.

[0140] 2. The calculation of the performance indicators takes the following factors into account:

[0141] Tracking error: e(k) = ||p_d(k) - p(k)||

[0142] Control energy consumption: E(k) = ||u(k)||2

[0143] Overshoot: σ(k) = max(0, p(k) - p_d(k))

[0144] settling time: t_s(k)

[0145] 3. The calculation formula of the comprehensive performance index is:

[0146] J(k) = w1·e(k) + w2·E(k) + w3·σ(k) + w4·t_s(k)

[0147] where w1-w4 are weight coefficients, which are determined by experimental optimization.

[0148] In order to ensure the stability of the control system, the system sets strict control quantity amplitude and rate constraints:

[0149] 1. Control quantity amplitude constraint:

[0150] |u(k)|≤u_max

[0151] where u_max is determined according to the rated torque of each joint of the manipulator, and the typical value is:

[0152] Large joint: ±100N·m

[0153] Medium joint: ±50N·m

[0154] Small joint: ±20N·m

[0155] 2. Control quantity rate constraint:

[0156] |u(k)-u(k-1)|≤Δu_max

[0157] where Δu_max is the maximum allowed rate, which is determined by the following adaptive mechanism:

[0158] Δu_max(k) = min(α·|e(k)|, Δu_limit)

[0159] where α is the proportional coefficient, and Δu_limit is the hard limit value.

[0160] 3. The system also introduces a constraint adjustment mechanism based on Lyapunov stability theory:

[0161] Define Lyapunov function:

[0162] V(k) = e(k) T Pe(k) + λ(k) T Qλ(k)

[0163] where P and Q are positive definite weight matrices.

[0164] The system is required to satisfy the stability condition:

[0165] AV(k) = V(k+1) - V(k) < 0

[0166] 4. To further improve the robustness of the system, adaptive dead-zone control is introduced:

[0167] u_dead(k) = {

[0168] u(k), |e(k)| > ε

[0169] 0, |e(k)| < ε

[0170] }

[0171] where ε is the dead-zone threshold, determined by online identification.

[0172] In addition, the system also implements the following optimization mechanisms:

[0173] 1. Soft constraint handling method is adopted to convert control constraints into penalty terms:

[0174] J_constraint(k) = μ1 max(0, |u(k)| - u_max) 2 + μ2 max(0, |u(k) - u(k-1)| - Δu_max) 2

[0175] where μ1 and μ2 are penalty factors.

[0176] 2. An online parameter optimization algorithm based on gradient descent is designed:

[0177]

[0178] where θ is the controller parameter vector, and γ is the learning step size.

[0179] 3. Anti-saturation mechanism for control parameters is implemented:

[0180] Integral separation strategy is adopted;

[0181] Upper and lower limits for the integral term are set;

[0182] Feedback anti-integral saturation link is introduced;

[0183] Through the above design, the system realizes the coordination of predictive control and feedback control, ensuring the control performance while ensuring the stability of the system. Experimental results show that the control strategy can achieve a good balance between fast response and precise positioning for the manipulator, with positioning accuracy better than ±0.1mm, overshoot less than 1%, and setting time not more than 0.5s.

[0184] See Figure 4In the process of electroplating hanger taking and placing operation, the embodiment of the application uses a state transition model based on Markov chain to optimize path planning. First, the system collects historical operation data of the electroplating hanger for training the state transition probability matrix. Then, the system predicts future taking and placing requirements according to the probability matrix. Finally, the system calculates the scheduling path that meets the constraints of the shortest path and the shortest waiting time by using a dynamic programming algorithm.

[0185] Firstly, the system establishes a state transition model based on unbounded Markov chain to describe the dynamic scheduling process of the hanger in the warehouse system. Secondly, by collecting historical operation data, the system trains the state transition probability matrix by using a machine learning algorithm to realize intelligent prediction of future taking and placing requirements. Thirdly, based on the prediction results, the system calculates the optimal scheduling path by using a dynamic programming algorithm, while considering multiple optimization objectives, including the shortest path, the shortest waiting time, and the optimal energy consumption. In order to handle random interference factors, the algorithm introduces robustness constraints to ensure good scheduling performance under uncertain conditions. In addition, the system designs a priority-based task scheduling mechanism that can dynamically adjust the operation order according to production requirements. It should be noted that in order to improve the parallel processing capability of the system, the scheduling algorithm uses a distributed computing architecture, and multiple scheduling units work cooperatively. Finally, the system continuously optimizes the scheduling strategy through real-time monitoring and feedback mechanism to adapt to the dynamically changing production environment.

[0186] Before using the Markov chain state transition model, the system needs to first establish a state space that describes the position and state change of the electroplating hanger. Then, the system calculates the transition probability of the state space based on historical operation data. It is worth noting that the system introduces robustness constraints in the dynamic programming algorithm to ensure the reliability of scheduling under uncertain conditions. In addition, in order to improve the calculation efficiency, the system distributes the path planning task to multiple computing nodes for parallel processing, and updates the state transition probability matrix in real time.

[0187] The specific process of state space construction, state transition probability matrix training, and path optimization is described in detail below.

[0188] Firstly, the embodiment of the application establishes a state space that describes the position and state change of the electroplating hanger. The state space is a multi-dimensional vector space, and each state vector contains the following dimension information:

[0189] 1) Spatial position information: including the position coordinates (x, y, z) of the electroplating hanger in the three-dimensional coordinate system;

[0190] 2) Motion state information: including the motion speed v and acceleration a of the electroplating hanger;

[0191] 3) Job status information: including the occupancy status of the electroplating hanger (free / occupied), task type (pick / place), and task priority (high / medium / low);

[0192] 4) Time status information: including the job start time t_start and the expected completion time t_end.

[0193] Secondly, the system trains the state transition probability matrix based on historical job data. The specific training process includes:

[0194] 1) Data preprocessing: collect the electroplating hanger job data of the last 30 days, and perform cleaning and standardization processing on the data. Mainly includes:

[0195] Removing outliers and noise data;

[0196] Unifying timestamp format;

[0197] Standardizing spatial coordinate values;

[0198] 2) State sequence generation: generate state transition sequences in chronological order based on the preprocessed data, each sequence contains:

[0199] Initial state s_t;

[0200] Subsequent state s_(t+1);

[0201] Transition time interval Δt;

[0202] Environmental conditions c_t;

[0203] 3) Probability matrix calculation: calculate the transition probability based on the state transition sequence:

[0204] P(s_(t+1)|s_t)=N(s_t→s_(t+1)) / N(s_t)

[0205] Where N(s_t→s_(t+1)) represents the number of samples from state s_t to s_(t+1), and N(s_t) represents the total number of samples of state s_t.

[0206] Thirdly, the system predicts future pick and place requirements according to the trained state transition probability matrix P. The prediction process uses an iterative approach:

[0207] 1) Given the current state s_t, calculate the most likely state at the next time:

[0208] s_(t+1)=argmax_s P(s|s_t)

[0209] 2) Prediction confidence evaluation:

[0210] conf(s_(t+1)) = P(s_(t+1)|s_t) / max(P(s|s_t))

[0211] When the confidence is below a threshold (usually set to 0.6), the system considers multiple possible subsequent states simultaneously.

[0212] In addition, the system uses a dynamic programming algorithm to calculate the optimal scheduling path that satisfies the constraint conditions. The specific implementation process includes:

[0213] 1) Define the optimization objective function:

[0214] J = w1*L + w2*T + w3*E

[0215] Where L is the path length, T is the waiting time, E is the energy consumption, and w1, w2, w3 are weight coefficients.

[0216] 2) Establish the state transition equation:

[0217] s_(t+1) = f(s_t, a_t)

[0218] Where a_t is the action decision at time t.

[0219] 3) Introduce robustness constraints:

[0220] Path feasibility constraint: Ensure that the planned path meets the kinematic constraints of the robot;

[0221] Collision avoidance constraint: Ensure a safety distance of at least 200mm from other electroplating hangers and fixed obstacles;

[0222] Time window constraint: Ensure that the task completion time meets the production rhythm requirements;

[0223] Load constraint: Ensure that the maximum load capacity of the robot is not exceeded.

[0224] Finally, to improve computational efficiency, the system uses a parallel computing architecture to handle path planning tasks:

[0225] 1) Task decomposition: Decompose the global path planning into multiple local planning subtasks, each responsible for path optimization in a specific area or time period.

[0226] 2) Parallel processing:

[0227] Use the Map-Reduce mode for parallel computing;

[0228] Map phase: Multiple computing nodes simultaneously process different subtasks;

[0229] Reduce phase: Combine the results of each subtask to obtain the global optimal solution;

[0230] 3) Real-time update mechanism:

[0231] The state transition probability matrix is updated every 5 minutes;

[0232] When an abnormal state transition is detected (probability change exceeds 20%), an update is triggered immediately;

[0233] Incremental update method is adopted, only the changed probability values are updated;

[0234] It should be noted that the system also designs an adaptive adjustment mechanism to dynamically adjust the optimization parameters according to the actual running effect:

[0235] When the actual running time exceeds the expected time by 20%, the time weight w2 is automatically adjusted;

[0236] When the energy consumption exceeds the expected energy consumption by 15%, the energy consumption weight w3 is automatically adjusted;

[0237] When the prediction accuracy is less than 85%, the historical data sampling window is increased;

[0238] Through the above design, the embodiment of the application realizes the intelligent planning of the electroplating hanger taking and placing path. Through testing, the scheme can shorten the average taking and placing time by 30%, and the accuracy rate of the planned path reaches 92%.

[0239] Referring to Figure 5 , the embodiment of the application uses a distributed data storage architecture to store and manage various information of the electroplating hanger. Specifically, the system establishes a master data node and multiple slave data nodes, wherein the master data node is responsible for data writing and distribution, and the slave data node is responsible for data backup and query. In actual application, the system uses a time series database to store the position information and motion state of the electroplating hanger in the master data node and the slave data node, and uses a relational database to store the job record.

[0240] In the data storage process, the system synchronizes data between the master data node and the slave data node based on a data consistency protocol. In addition, the system sets a data sharding strategy to store the electroplating hanger information in time dimension and space dimension. This distributed architecture not only improves the reliability of data storage, but also realizes multi-dimensional query and statistical analysis of electroplating hanger storage information.

[0241] In addition, the system establishes a multi-level data acquisition network, including the device layer, control layer and management layer, to achieve comprehensive data collection and real-time transmission. Secondly, each hanger is equipped with an intelligent sensor with a unique identity, which can record full life cycle data including usage times, maintenance records, and working status. Thirdly, the system adopts a distributed data storage architecture combined with blockchain technology to ensure data integrity and tamper resistance. Among them, by setting up multiple data verification mechanisms, the accuracy and reliability of data acquisition are guaranteed. In addition, the system develops an intelligent data analysis module that can monitor the running status of the equipment in real time and predict potential failures. It should be noted that the data traceability system supports multi-dimensional query and statistical analysis, which can quickly locate the root cause of any abnormal situation. However, considering the data security requirements, the system implements strict access control and encryption measures. Finally, through the establishment of a perfect data backup and recovery mechanism, the security and availability of critical data are ensured.

[0242] The embodiments of the present application also establish a digital demand management system. The system first establishes a structural view, a behavior view, a demand view and a parameter view describing the electroplating hanger intelligent warehouse system. Then, through the demand decomposition matrix, the functional requirements and performance requirements of the system are decomposed into multiple technical indicators.

[0243] It should be noted that the digital demand management system is based on the SysML meta-model derived from INCOSE. Specifically, the system uses SysML demand diagrams to describe the functions and performance requirements of the electroplating hanger intelligent warehouse system, and adopts module definition diagrams to establish system structure models and determine structural technical indicators. At the same time, the system defines the connection relationship within the system through internal block diagrams and determines the connection technical indicators, uses activity diagrams and state machine diagrams to describe the dynamic behavior of the system and determines the dynamic behavior technical indicators. Finally, the system establishes a bidirectional traceability relationship between requirements and design elements.

[0244] Specifically, the system adopts the SysML modeling language of INCOSE (International Council on Systems Engineering) standards to build a complete meta-model of the electroplating hanger storage system. Secondly, the meta-model contains four key views: structural view, behavioral view, requirement view, and parameter view, which are used to comprehensively describe the functional and performance requirements of the system. Thirdly, through the requirement decomposition matrix, the system decomposes high-level business requirements into specific technical indicators and implementation schemes step by step. Each requirement contains explicit verification criteria and traceability links to ensure the integrity and consistency of the requirements. In addition, the system integrates a requirement change management mechanism to track and evaluate the impact of requirement changes on the system in real time. It should be noted that in order to improve the reusability of the requirement model, the system establishes a standardized requirement template library. However, considering the individualized requirements of different enterprises, the model design retains sufficient flexibility and extensibility. Finally, through continuous requirement verification and confirmation processes, the system ensures that its functions are consistent with user expectations.

[0245] Reference Figure 6 The embodiments of the present application also provide an electroplating hanger intelligent storage system, which is used in cooperation with a production workshop to implement the above method. The system includes a multi-layer three-dimensional storage rack, which contains a plurality of shelf units, and each shelf unit is connected through a standard interface. The automatic transmission system of the system includes a six-degree-of-freedom manipulator and a conveyor belt with an encoder, wherein the manipulator is connected with the control system through a control bus, and the conveyor belt is connected with the control system through an industrial Ethernet.

[0246] The identification system of the system includes a laser scanner with a precision of ±1mm and an RFID reader supporting the EPC Gen2 protocol. In addition, the system adopts a distributed control system to realize multi-task collaborative control through a master-slave architecture. The hardware system architecture provides reliable infrastructure support for the implementation of the above method.

[0247] The electroplating hanger intelligent storage method provided by the embodiments of the present application realizes accurate positioning, intelligent scheduling, and full-process management of electroplating hangers by establishing a three-dimensional space rectangular coordinate positioning system, using a predictor-feedback control system, introducing a Markov chain state transition model, and using a distributed data storage architecture, which significantly improves the automation level and operation efficiency of the storage system.

[0248] It should be noted that the technical features not described in detail in the embodiments of the present application can be reasonably inferred by those skilled in the art based on the above content. For example, the specific parameter settings of each module in the system can be adjusted according to the actual application scenario. In addition, the specific implementation manner of the embodiments of the present application is not limited to the above description, and those skilled in the art can make equivalent substitutions or improvements according to actual needs.

[0249] Figure 7This is a schematic diagram of an actual factory including smart warehousing, as described in another embodiment of this application. Figure 7 The factory includes an intelligent warehouse 200 and a main production line 100. The intelligent warehouse 200 works in conjunction with the main production line 100. By establishing a three-dimensional rectangular coordinate positioning system, adopting a predictor-feedback control system, introducing a Markov chain state transition model, and using a distributed data storage architecture, the intelligent warehouse 200 achieves precise positioning, intelligent scheduling, and full-process management of electroplating racks, significantly improving the automation level and operating efficiency of the warehousing system.

[0250] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0251] This invention also provides a computer system suitable for implementing a dual-starter switching and control method for artificial intelligence-based intelligent devices. The computer system includes a Central Processing Unit (CPU), which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) or loaded from storage into random access memory (RAM), such as executing the methods described in the above embodiments. The RAM also stores various programs and data required for system operation. The CPU, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0252] The following components are connected to the I / O interface: input components including a keyboard, mouse, etc.; output components including cathode ray tubes (CRTs), liquid crystal displays (LCDs), and speakers; storage components including hard disks; and communication components including network interface cards such as LAN (Local Area Network) cards and modems. The communication components perform communication processing via a network such as the Internet. Drive 310 is also connected to the I / O interface as needed. Removable media, such as disks, optical disks, magneto-optical disks, semiconductor memories, etc., are installed on the drive as needed so that computer programs read from them can be installed into the storage components as required.

[0253] In particular, the processes described above with reference to the flow charts can be implemented as computer software programs in accordance with embodiments of the present application. For example, embodiments of the present application include a computer program product comprising a computer program carried on a computer readable medium, the computer program comprising computer programs for executing the methods illustrated by the flow charts. In such embodiments, the computer program can be downloaded and installed from a network via a communication section, and / or installed from a detachable medium. When the computer program is executed by a central processing unit (CPU), various functions defined in the system of the present application are executed.

[0254] It should be noted that the computer readable medium shown in embodiments of the present application can be a computer readable signal medium or a computer readable storage medium or any combination thereof. The computer readable storage medium may, for example, be, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or apparatus, or any suitable combination thereof. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0255] In the present application, the computer readable storage medium can be any tangible medium that contains or stores a program used by or in connection with an instruction execution system, apparatus or device. In the present application, the computer readable signal medium can include a data signal carried in a baseband or as part of a carrier wave, in which a computer readable program is carried. Such a propagated data signal can take a variety of forms, including but not limited to an electromagnetic signal, an optical signal or any suitable combination thereof. The computer readable signal medium can also be any computer readable medium that can send, propagate or transfer a program for use by or in connection with an instruction execution system, apparatus or device. The computer program contained in the computer readable medium can be transmitted in any suitable medium, including but not limited to wireless, wired, or the like, or any suitable combination thereof.

[0256] The computer readable storage medium can be a tangible storage medium (e.g., any type of disk including floppy disks, optical disks, CD- ROMs, and magnetic- or optical storage disks) encoded with the computer program, docket # 30699.2 PCT

[0257] The units described in the embodiments of the present application can be implemented by software, or by hardware, or by a combination of software and hardware. The units described may

[0258] According to an aspect of the present application, a computer program product or computer program is provided, which comprises computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device performs the method provided in the various possible implementation manners described above.

[0259] As another aspect, the present application also provides a computer readable medium, which can be included in the electronic device described in the above embodiments, or can exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to implement the method described in the above embodiments.

[0260] It should be noted that although several modules or units for performing actions are mentioned in the above detailed description, the division into such modules or units is not mandatory. In fact, according to the embodiments of the present application, features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, features and functions of one module or unit described above can be further split into multiple modules or units.

[0261] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solutions according to the embodiments of the present application can be embodied in the form of a software product. The software product can be stored in a non-volatile storage medium (which can be a CD-ROM, a U disk, a mobile hard disk, etc.) or on a network, and includes a number of instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to perform the method according to the embodiments of the present application.

[0262] Although the embodiments of the present specification disclose as above, the present application is not limited thereto. Any person skilled in the art, without departing from the spirit and scope of the present application, can make various changes and modifications, therefore the protection scope of the present application should be subject to the scope defined by the claims.

Claims

1. A method for intelligent storage of electroplating hangers, characterized in that , including: establishing a three-dimensional space rectangular coordinate positioning system, in which the electroplating hanger is positioned and calibrated; controlling the mechanical hand to take and place the electroplating hanger, including: using a linear control system with state feedback, when the distance between the mechanical hand and the target position is greater than a preset distance threshold, controlling the mechanical hand to move quickly using a first control gain, when the distance between the mechanical hand and the target position is less than or equal to the preset distance threshold, controlling the mechanical hand to accurately position using a second control gain, and real-time compensating for inertia error and mechanical friction in the movement of the mechanical hand; wherein, during the electroplating hanger taking and placing operation, a state transition model based on Markov chain is used to optimize the path planning of the electroplating hanger taking and placing, including: collecting historical operation data of the electroplating hanger to train a state transition probability matrix; predicting future taking and placing requirements according to the state transition probability matrix; using a dynamic programming algorithm to calculate a scheduling path that meets the shortest path and least waiting time constraints; wherein, before using the state transition model based on Markov chain to optimize the path planning of the electroplating hanger taking and placing, the method further includes: establishing a state space describing the position and state change of the electroplating hanger; calculating the transition probability of the state space based on the historical operation data; introducing robustness constraints in the dynamic programming algorithm to ensure scheduling reliability under uncertain conditions; distributing the path planning task to multiple computing nodes for parallel processing; real-time updating the state transition probability matrix; using a distributed data storage architecture to store the position information, motion state and operation record of the electroplating hanger, to realize multi-dimensional query and statistical analysis of the electroplating hanger storage information; the method further includes: establishing a digital demand management system, including: establishing a structure view, a behavior view, a demand view and a parameter view of the electroplating hanger intelligent storage system; decomposing the functional requirements and performance requirements of the electroplating hanger intelligent storage system into multiple technical indicators through a demand decomposition matrix.

2. The method of claim 1, wherein the digital demand management system is constructed based on the SysML meta-model derived from INCOSE, and the decomposition of the functional requirements and performance requirements of the electroplating hanger intelligent storage system into multiple technical indicators through the demand decomposition matrix includes: using a SysML requirement diagram to describe the function and performance requirements of the electroplating hanger intelligent storage system; using a module definition diagram to establish a structure model of the electroplating hanger intelligent storage system and determine structure technical indicators; defining the connection relationship inside the electroplating hanger intelligent storage system through an internal block diagram and determining connection technical indicators; using an activity diagram and a state machine diagram to describe the dynamic behavior of the electroplating hanger intelligent storage system and determine dynamic behavior technical indicators; establishing a bidirectional traceability relationship between requirements and design elements.

3. The method of claim 1, wherein establishing a three-dimensional space rectangular coordinate positioning system, in which the electroplating hanger is positioned and calibrated, including: installing laser ranging sensors with an accuracy of ±1mm at key node positions of the storage rack, the laser ranging sensors forming a sensor network matrix; Each of the electroplating hangers is equipped with a source RFID tag, which stores identification information including a unique identification code, a type and a load parameter of the electroplating hanger; A plurality of parallel laser beam array emitters are used to collect reflection signals of the electroplating hangers in real time; The distance between the electroplating hanger and a fixed reference point is calculated based on a time-of-flight ranging principle, and the distance is calibrated with a preset fixed calibration point.

4. The method of claim 1, wherein The linear control system is a predictor-feedback control system with a transient switch, and the method further comprises: The transient switch is arranged in the state feedback link, and is used to dynamically switch the first control gain and the second control gain; A predictor model is used to calculate a motion trajectory of the manipulator in advance; A feedback correction mechanism is established according to an actual position of the manipulator; Based on a weighted combination of a calculation result of the predictor model and a feedback signal, a control instruction of the manipulator is generated.

5. The method of claim 4, wherein The method further comprises: A smooth transition operation is performed during the switching process of the transient switch to avoid a jump of a control quantity; Parameters of the predictor model are adaptively adjusted according to a load of the manipulator; An adaptive weight distribution of the predictive control and the feedback control is realized, and a control quantity amplitude limit and a change rate constraint are set.

6. The method of claim 1, wherein, The position information, the motion state and the work record of the electroplating hangers are stored by using a distributed data storage architecture, comprising: A master data node and a plurality of slave data nodes of the distributed data storage architecture are established, the master data node is responsible for data writing and distribution, and the slave data nodes are responsible for data backup and query; A time series database is used to store the position information and the motion state of the electroplating hangers in the master data node and the slave data nodes; A relational database is used to store the work record of the electroplating hangers in the master data node and the slave data nodes; Data is synchronized between the master data node and the slave data nodes based on a data consistency protocol; A data sharding strategy is set, and the electroplating hanger information is stored in a data partition according to a time dimension and a space dimension.

7. A system for intelligent warehousing of electroplating racks, characterized in that The method is used to realize the method of any one of claims 1-6, comprising: A multi-layer three-dimensional warehouse frame comprises a plurality of layers of shelf units, and each layer of shelf unit is connected through a standard interface; An automatic transmission system comprises a six-degree-of-freedom manipulator and a conveyor belt with an encoder, the manipulator is connected to a control system through a control bus, and the conveyor belt is connected to the control system through an industrial Ethernet; An identification system comprises a laser scanner with an accuracy of ±1mm and an RFID reader supporting an EPC Gen2 protocol; A distributed control system uses a master-slave architecture to realize multi-task cooperative control.

Images