An automatic assembly system, assembly method and control method for detonator-primers

Abstract

The present application relates to the field of blasting automation assembly, in particular to a detonator-primacord automatic assembly system, assembly method and control method, comprising a second platform for carrying and clamping primacord, and a detonator clamp jaw corresponding to the vertical position of the primacord arranged above the second platform, the detonator clamp jaw is fixed on the driving end of the lifting cylinder and driven by the lifting cylinder to produce vertical lifting action to clamp the detonator vertically inserted into the set position of the primacord; the push cylinder, the second platform and the packing machine are arranged in sequence along the push direction of the push cylinder, and the completed detonator-primacord is linearly pushed to the packing machine by the push cylinder. The present application can realize high-precision and high-safety automatic assembly of detonator-primacord.

Description

Technical Field

[0001] This invention relates to the field of automated assembly in blasting, specifically an automated assembly system, assembly method, and control method for detonators and detonators. Background Technology

[0002] In blasting projects such as mining and tunneling, a complete blasting process mainly includes blasting design, drilling, charging, packing, and detonation. The charging stage, as the core stage connecting drilling and detonation, directly affects the blasting effect and operational safety. The reliable assembly of the detonator and detonating charge is a crucial step in determining the efficiency of energy transfer during detonation. This process requires selecting a suitable explosive charge as the detonating charge, accurately inserting the detonator into the center of the bottom of the charge, and securing it with tape or binding wire to ensure the stability and reliability of the detonating charge. However, this critical process currently still largely relies on manual operation, presenting several prominent problems.

[0003] Firstly, manual assembly is affected by the operator's skill level, making high-speed, standardized operations difficult and hindering overall blasting efficiency. Furthermore, both detonators and detonators are hazardous materials, and direct manual assembly poses a high safety risk; operational errors could lead to accidents. Secondly, manual insertion of detonators lacks precise control over the insertion force. Excessive force may damage the sensitive charge or structure inside the detonator; insufficient force may result in insecure fixation and poor contact, leading to detachment or misfires during subsequent loading or detonation, severely impacting the reliability and overall effectiveness of the blasting operation. In conclusion, the traditional manual assembly method of detonators and detonators is no longer suitable for the high efficiency, high precision, and high safety requirements of modern, large-scale blasting operations, and these problems urgently need to be addressed. Summary of the Invention

[0004] To avoid and overcome the technical problems existing in the prior art, this invention provides an automatic assembly system, assembly method, and control method for detonators and detonators. This invention can achieve high-precision and high-safety automated assembly of detonators and detonators.

[0005] To achieve the above objectives, the present invention provides the following technical solution: An automatic assembly system for detonators and detonators includes a second platform for carrying and clamping the detonator, and a detonator gripper positioned above the second platform corresponding to the vertical position of the detonator. The detonator gripper is fixed to the drive end of a lifting cylinder and, driven by the lifting cylinder, performs a vertical lifting motion to clamp the detonator and insert it vertically into the set position of the detonator. A push cylinder, the second platform, and a packing machine are arranged sequentially along the pushing direction of the push cylinder. The assembled detonator-detonator is linearly pushed by the push cylinder to the packing machine for strapping.

[0006] As a further aspect of the present invention: a clamping cylinder is provided above the baling machine. The clamping cylinder is configured to clamp and fix the detonator-detonating shell when the baling machine performs a winding and baling operation on the detonator-detonating shell. The clamping end of the clamping cylinder is set as an arc-shaped surface that matches the shape of the detonating shell.

[0007] As a further embodiment of the present invention, it also includes a detonator conveyor for conveying detonators, a slide rail is horizontally arranged above the second platform, a lifting cylinder is slidably arranged on the slide rail, and the detonators conveyed on the detonator conveyor correspond to the positions of the detonator grippers along the sliding path of the lifting cylinder.

[0008] As a further embodiment of the present invention: the second platform is fixed on the base and rotates with the base, with the rotation axis arranged horizontally; the base is also provided with an electric push rod that rotates with the second platform, and the electric push rod extends and retracts while driving the second platform to pitch and adjust the pitch angle of the detonating bomb; two sets of clamping cylinders are symmetrically arranged on the second platform along the horizontal direction, and the driving ends of the two sets of clamping cylinders extend and retract towards each other to clamp and position the detonating bomb; the driving end of the pushing cylinder is provided with a pushing claw that applies a clamping action to the detonating bomb.

[0009] As a further embodiment of the present invention: the push cylinder is fixed on the first platform of the base, the first platform and the second platform are arranged in sequence along a straight line, and the first platform and / or the second platform are also provided with a limiter for regulating and constraining the detonator lead wire.

[0010] As a further aspect of the present invention: a pressure sensor is provided between the detonator gripper and the drive end of the lifting cylinder, and the pressure signal received by the pressure sensor is transmitted to the PLC controller.

[0011] As a further aspect of the present invention: an electric proportional valve is installed in the air circuit of the lifting cylinder to adjust the cylinder speed according to PLC instructions; a photoelectric gate sensor for sensing the position of the detonating bomb is set on the second platform.

[0012] An assembly method for an automatic assembly system for detonators and detonators includes the following steps: S1. Place the detonating bomb on the second platform. Detect whether the detonating bomb has reached the set position through the photoelectric gate sensor. Once the set position is reached, activate the clamping cylinder to clamp the detonating bomb. Then, drive the electric push rod to adjust the second platform to the set pitch angle. S2. The lifting cylinder slides along the slide rail to above the detonator conveyor. After the detonator gripper moves down and clamps a set of detonators, the lifting cylinder slides along the slide rail to above the detonating shell. S3. The lifting cylinder drives the detonator gripper to insert the detonator into the detonating shell. Based on the pressure feedback from the pressure sensor, the PLC controller controls the electric proportional valve to dynamically adjust the insertion speed of the lifting cylinder so that the insertion force of the detonator is maintained within the set range. S4. After inserting the detonator to the set depth of the detonating bomb, the detonator gripper releases the detonator, and the lifting cylinder drives the detonator gripper to reset; the electric push rod drives the second platform to reset to the horizontal state. S5. After the pusher gripper has finished clamping the detonator-inserted explosive, it is pushed to the packing machine by the pusher cylinder. S6. After the clamping cylinder clamps the detonator, start the packing machine to tightly pack the detonator and the detonator's fuse bundle, completing the assembly of the detonator and the detonator. S7. After each mechanism is reset, repeat the above steps to form a cyclical operation.

[0013] A control method for an automatic assembly system of detonator-detonating bomb, characterized by comprising the following steps: S1. Set up the controller, which specifically includes: The main control loop is configured to run periodically based on an adaptive neuro-fuzzy inference system and optimize the parameters of the sub-control loop based on historical performance data. The secondary control loop is configured to activate when preset conditions are met. After activation, it performs real-time fuzzy inference on the input force deviation and the rate of change of force deviation, and outputs adaptive PID control parameters. The loss function module is configured to evaluate system performance and guide learning and decision-making; The stability assurance module is configured to constrain the learning process of the main control loop to ensure system stability; S2. Initialize the system by setting the initial membership function parameters, comprehensive loss function, domain transformation activation condition threshold, learning period, and stability constraint parameters of the controller. S3. Read the pressure sensor signal in real time during each control cycle and calculate the current insertion force of the detonator gripper. And calculate the current force deviation. Force deviation change rate and instantaneous loss function ; ; ; ; in, The preset insertion force; for The change in force deviation over time; for The control quantity output by the controller to the electro-proportional valve at any given moment; a, b All are weighting coefficients; aThe value range is 0.5 to 2.0; b The value range is 0.1 to 0.5; S4. Determine whether the preset domain transformation activation condition is met. If it is met, then... , Perform a nonlinear transformation to obtain the feature input. as well as Otherwise, use the raw input directly. ); ); ; ; ; ; in, This represents the force deviation value after nonlinear domain transformation. This represents the rate of change of force deviation after nonlinear domain transformation. This indicates the current mechanical state of the detonator gripper; The reference gain used in calculating the force deviation value; The reference gain used in calculating the rate of change of force deviation; This is the performance threshold; as well as For state S The corresponding fine-tuning coefficients; max This is a function to find the maximum value. 、 All are normalization factors; This represents the maximum allowable steady-state force deviation of the system. ΔT To control the cycle; C The control parameter is a constant, and its value ranges from 1.5 to 3.0. S5. Based on the current membership function parameter set, perform feature input... as well as Perform fuzzy inference and output an adaptive set of results. PID Control parameters KP(k), KI(k), KD(k) Calculate control quantity As The final control quantity output by the controller to the electro-proportional valve at any given time.

[0014] As a further aspect of this invention: Constructing an online evaluation loss function. To evaluate the overall performance of the system within a set time window: ; in, Indicates the instantaneous loss function within the set time window. The average value; λ This is the weighting coefficient, with a value ranging from 0.1 to 0.5; Indicates the control quantity within the set time window. The standard deviation of the output; At set intervals, the main control loop is activated, and it collects historical data to minimize... To achieve the target, calculate the parameter adjustment amount of the secondary control loop; Build a discriminant function ; in e This represents the force deviation value at the current moment; The learning rate of the main control loop is dynamically adjusted based on the changing trend of the discriminant function. : like Then let ; like and Then let ; like and Then let 1.2 ; This represents the preset base learning rate; This represents the stability threshold of the force deviation.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention achieves fully automated assembly of detonator-detonating bombs from feeding, positioning, clamping, detonator grabbing, insertion to pushing and packaging through the coordination of multiple cylinders, slide rail transfer and sensor triggering. It greatly reduces manual intervention and operation time, improves overall assembly efficiency, and can achieve high-precision and high-safety automated assembly of detonators-detonating bombs.

[0016] 2. This invention employs symmetrically arranged clamping cylinders to synchronously clamp the detonating bomb in the horizontal direction. Combined with the electric push rod of the second platform for pitch adjustment, this ensures the detonating bomb is adjusted to a suitable pitch angle to accommodate the detonator insertion angle requirements under different working conditions. The lifting cylinder and grippers, guided by a slide rail, achieve precise gripping and alignment of the detonator. A pressure sensor monitors the insertion force in real time, and a PLC-controlled proportional valve dynamically adjusts the cylinder pressure and downward speed, ensuring the insertion force remains within a set safe range. This prevents damage to the detonator or detonating bomb due to overload and also prevents incomplete insertion from affecting product quality.

[0017] 3. The present invention uses a push cylinder to move the assembly to the packaging station, and uses an arc-shaped pressing surface cylinder for contour fixing. Then, the strapping machine automatically completes the wrapping and tightening, ensuring accurate packaging position and consistent tightness, which is beneficial for subsequent storage and transportation.

[0018] 4. This invention establishes a real-time nonlinear domain variation mechanism by combining an adaptive neural fuzzy inference main loop with a fuzzy control sub-loop, achieving high-precision control of the insertion force in the automatic assembly of detonators and detonators. The sub-control loop is responsible for the rapid suppression and compensation of sudden disturbances and dynamic anomalies. The main control loop is responsible for continuously and progressively optimizing the parameters of the sub-control loop based on historical performance data. This achieves decoupling and coordination between the inherent parameter optimization of the controller and the deformation of runtime conditional behavior. Under normal processes, the controller operates in a main control learning + sub-controller conventional execution mode, continuously optimizing parameters to improve performance. Under abnormal processes, abnormal data is learned by the main control, further optimizing the inherent parameters, thereby enabling the controller to have emergency response and evolutionary capabilities, significantly improving the overall performance of the controller under complex, uncertain, and high-safety-requirement processes. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the structure of the present invention.

[0020] Figure 2 This is a top view of the present invention.

[0021] Figure 3 This is a schematic diagram of the fuzzy rule base used in fuzzy reasoning in this invention.

[0022] Figure 4 This is a comparison chart of the force tracking performance of the present invention.

[0023] Figure 5 This is a comparison chart of the tracking errors of the present invention.

[0024] Figure 6 This is a comparison diagram of the control signals of the present invention.

[0025] Figure 7 This is a comparison chart of energy consumption control according to the present invention.

[0026] In the picture: 100. Base; 110. First platform; 120. Second platform; 130. Electric actuator; 140. Limiting device; 150. Photoelectric gate sensor; 160. Detonator delivery device; 170. Packaging machine; 180. Clamping cylinder; 190. Detonator gripper; 200. Lifting cylinder; 210. Clamping cylinder; 220. Pushing gripper; 230. Pushing cylinder; 240. Electro-proportional valve; 250. Pressure sensor; 260. Slide rail. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] Please see Figures 1 - 2 In this embodiment of the invention, an automatic assembly system, assembly method, and control method for a detonator-detonating bomb are disclosed. The automatic assembly system is based on a base 100, on which a second platform 120 is mounted adjacent to a first platform 110. The second platform 120 is fixed to the base 100 and rotates with the base 100, with the rotation axis arranged horizontally. An electric push rod 130 is also provided on the base 100, rotating with the second platform 120. The electric push rod 130 extends and retracts, driving the second platform 120 to pitch, thereby adjusting the pitch angle of the detonating bomb. By adjusting the pitch angle, the detonating bomb adjusts its angle synchronously to adapt to the blasting requirements under different operating conditions.

[0029] Two sets of clamping cylinders 210 are symmetrically mounted on the second platform 120, with their drive ends facing each other. When the detonator is placed in the set position on the second platform 120, the two sets of clamping cylinders 210 operate synchronously, clamping the detonator from both sides in the horizontal direction to achieve positioning and fixation. To ensure that the detonator leads are neatly arranged during assembly, a wire limiter 140 is installed on the first platform 110 and / or the second platform 120. The wire limiter 140 is an existing mechanism, so its structure will not be described in detail. To detect whether the detonator has been correctly placed, a photoelectric gate sensor 150 is installed on the second platform 120. When the detonator is fed in and blocks the photoelectric beam, the sensor sends a positioning signal, and the system immediately starts clamping and subsequent processes. Each gripper is fixed to the drive end of the cylinder with bolts.

[0030] The slide rail 260 is suspended above the second platform 120 via a fixing bracket, used to fix the lifting cylinder 200. When a detonator needs to be picked up, the lifting cylinder 200 slides along the slide rail 260 above the detonator conveyor 160, driving the detonator gripper 190 to descend and pick up a detonator. Subsequently, the lifting cylinder 200, carrying the detonator, moves horizontally along the slide rail 260 to directly above the detonating shell already clamped and fixed on the second platform 120. The lifting cylinder 200 drives the detonator gripper 190 to descend vertically, precisely inserting the detonator into the predetermined position inside the detonating shell.

[0031] In this embodiment, the form of the detonator conveyor 160 is not limited. It can be a conveyor belt type conveyor or a robotic arm picking and delivering mechanism. The preferred embodiment is a turntable type mechanism. Before use, the detonators are inserted into the turntable. Each time the turntable is rotated at a set angle, one set of detonators is conveyed to the picking position.

[0032] To ensure controllable force during the insertion process and prevent damage to the detonator or detonating bomb due to excessive insertion force, or incomplete assembly due to insufficient insertion force, a pressure sensor 250 is installed between the detonator gripper 190 and the drive end of the lifting cylinder 200. The pressure sensor 250 monitors the pressure signal during the insertion process in real time and transmits it to the system's PLC controller. Simultaneously, an electro-proportional valve 240 is installed in the air circuit driving the lifting cylinder 200. The PLC controller compares the pressure value fed back by the pressure sensor 250 with a preset reasonable pressure range and dynamically adjusts the air pressure entering the cylinder by sending commands to the electro-proportional valve 240, thereby controlling the downward speed of the lifting cylinder 200 in real time and ensuring that the insertion force remains within the set safe range.

[0033] After the detonator is successfully inserted into the detonating shell and reaches the set depth, the detonator gripper 190 releases, and the lifting cylinder 200 drives the gripper to rise and reset. Simultaneously, the electric push rod 130 resets, causing the second platform 120 to return to a horizontal position, aligned with the first platform 110. Subsequently, the system enters the pushing and packaging stage. The pushing cylinder 230 is fixedly installed on the first platform 110, with its pushing direction pointing towards the packaging machine 170. The drive end of the pushing cylinder 230 is equipped with a pushing gripper 220, which holds the inserted detonator-detonating shell assembly. The pushing cylinder 230 then activates, pushing the assembly in a straight line to the packaging station of the packaging machine 170. Since the packaging machine 170 is existing technology, its structure will not be described in detail.

[0034] In this embodiment, the strapping machine 170 is preferably a strapping machine. To prevent the product from moving during the strapping process, a clamping cylinder 180 is installed above the strapping station of the strapping machine 170. The drive end of the clamping cylinder 180 is designed with an arc-shaped surface to apply contour-following clamping to the product, using the arc-shaped surface to press and clamp the detonator, fixing it on the strapping station. Subsequently, the strapping machine 170 starts, and the strapping tray of the strapping machine 170 rotates, wrapping the strapping around the junction of the detonator lead wire and the detonator, and tightening it. After strapping is completed, the clamping cylinder 180 is lifted, and all actuators reset.

[0035] The assembly method based on the automated assembly system of this embodiment includes the following steps: S1. The operator or automatic feeding device places the detonating bomb in the designated area of ​​the second platform 120. After the photoelectric gate sensor 150 detects that the detonating bomb is in place, the system controls the two sets of clamping cylinders 210 to operate simultaneously, clamping the detonating bomb from both sides. According to the preset program, the PLC controller starts the electric push rod 130 to raise and lower the second platform 120 along with the detonating bomb clamped on it to the set height, reserving the height for the subsequent insertion of the detonator.

[0036] S2. The lifting cylinder 200, carrying the detonator gripper 190, moves along the horizontal slide rail 260 to above the detonator conveyor 160. Then, the lifting cylinder 200 drives the detonator gripper 190 downwards to the gripping position, where the gripper grips a detonator. Next, the lifting cylinder 200 rises and moves in the opposite direction along the slide rail 260, transferring the detonator directly above the detonating shell that has been clamped and fixed on the second platform 120.

[0037] S3. The lifting cylinder 200 drives the detonator gripper 190 to move vertically downwards carrying the detonator, beginning to insert the detonator into the detonating shell. During this process, the pressure sensor 250 continuously monitors the insertion force. The PLC controller receives the pressure signal in real time, and obtains the actual insertion force value by acquiring the voltage signal from the pressure sensor 250 and performing A / D conversion; simultaneously, it outputs a 0-10V control signal to the electro-proportional valve 240 to precisely adjust the cylinder's movement speed.

[0038] S4. Once the detonator is inserted to the preset depth, the detonator gripper 190 releases, releasing the detonator. Subsequently, the lifting cylinder 200 drives the detonator gripper 190 to rise vertically to a safe height and move along the slide rail 260 back to the initial detonator gripping position, waiting for the next cycle.

[0039] S5. The clamping cylinder 210 on the second platform 120 releases its grip on the detonator, and the pusher cylinder 230 drives the pusher claw 220 to clamp the assembled detonator-detonator assembly. Then, the pusher cylinder 230 extends and pushes the assembly in a straight line to the packaging platform of the packaging machine 170.

[0040] S6. After the product is pushed onto the packaging platform of the packaging machine 170, the clamping cylinder 180 descends, and the arc-shaped clamping surface at its end compacts the detonator to prevent it from moving during the packaging process. After clamping, the strapping machine 170 automatically starts to complete the wrapping and tightening of the detonator lead wire and the connection between the detonator and the detonator.

[0041] S7. After packaging is completed, the clamping cylinder 180 lifts, the pusher gripper 220 releases and retracts back to its original position with the pusher cylinder 230. All mechanisms of the system return to their initial state. Subsequently, the system automatically begins the next assembly cycle.

[0042] The control method based on the automated assembly system specifically includes the following steps: Build a controller, which specifically includes: The main control loop is configured to run periodically based on an adaptive neuro-fuzzy inference system and optimize the parameters of the sub-control loop based on historical performance data. The main control loop is built based on an Adaptive Neural Fuzzy Inference System (ANFIS). The main control loop runs periodically, receiving historical system performance data. By minimizing a comprehensive loss function, it dynamically learns and outputs the Gaussian membership function center for the secondary controller. c ) and width ( σ The optimal parameter adjustment amount is obtained, thereby gradually optimizing the controller's intrinsic performance benchmark over a long period of time; The secondary control loop is activated when preset conditions are met. After activation, it performs real-time fuzzy inference on the input force deviation and the rate of change of force deviation, and outputs adaptive PID control parameters. The secondary control loop is based on a conventional fuzzy controller with a Gaussian membership function as its basis. It receives parameters from the primary control loop, performs real-time fuzzy inference of force deviation and its rate of change, and outputs control signals.

[0043] The secondary control loop embeds an intelligent conditional domain transformation module. This module is normally inactive, continuously monitoring the system status. It is activated when specific preset conditions are met (such as encountering sudden resistance, error plateauing, or a rapid increase in the loss function). When activated, this module performs a preset nonlinear mapping operation on the input force deviation and its rate of change. This operation does not directly adjust the structural parameter center of the membership function. c ) and width ( σ Instead of directly altering the input variables, this process applies a variable, nonlinear scaling transformation, effectively changing the mapping relationship of the original input values ​​in the membership calculation. Without disturbing the controller's inherent parameters, it changes the effective calculation form of the membership function within the current inference cycle in real time, enabling the fuzzy controller to generate more adaptive adjustment outputs for transient abnormal processes.

[0044] Using the transformed or untransformed feature input, based on the current membership function parameter center ( c ) and width ( σ ), respectively query the preset values ​​for proportional coefficient KP, integral coefficient KI, and differential coefficient KD, such as Figure 3 The fuzzy rule table shown is used to calculate and output a set of adaptive PID control parameters in real time. KP(k), KI(k), KD(k) .

[0045] The loss function module is used to evaluate system performance and guide learning and decision-making; it specifically includes: The instantaneous loss function, calculated in real time based on the current control deviation and output, serves as a health indicator of the system's instantaneous performance. When this indicator rises abnormally, it triggers rapid intervention from the domain transformation module of the secondary control loop.

[0046] A comprehensive evaluation loss function is calculated based on performance indicators such as the integral of absolute error and the integral of squared error over a period of time, serving as the optimization objective for parameter learning in the main control loop. The goal of the main control loop is to continuously adjust the parameters to minimize this comprehensive evaluation loss.

[0047] The stability discrimination function, constructed based on Lyapunov theory, is used to monitor the stability trend of the system in real time and dynamically constrain the learning rate of the main control loop accordingly, ensuring that the entire optimization process always remains in the stable region.

[0048] The stability assurance module is used to constrain the learning process of the main control loop to ensure system stability; The stability assurance module, based on Lyapunov stability theory, constructs the system energy function and derives the parameter update constraints that enable the system to achieve global asymptotic stability. These constraints are embedded in the learning algorithm of the main control loop, and the learning rate is dynamically adjusted according to changes in the loss function and energy function to ensure the safety and reliability of the entire adaptive process.

[0049] I. During the system initialization phase, the controller is powered on, each actuator returns to its origin, the main control loop ANFIS network parameters are initialized, and the initial center of the Gaussian membership function of the secondary control loop is set. c With width σ .

[0050] Define the comprehensive loss function L ,For example ; in, This is the integral of the absolute error; The integral of the squared error; To control the variance.

[0051] Accumulated weights for precision, with values ​​ranging from 0.2 to 0.5.

[0052] This is the weight for suppressing large errors, with a value range of 0.4 to 0.7.

[0053] To implement stationarity weights, the values ​​range from 0.05 to 0.2.

[0054] The detonator descent process is divided into four states (S) based on the force sensor readings: 1. No-load downstream: F < 1N; 2. Initial contact: 8N < F < 14N; 3. Critical puncture point: 15N < F < 20N; 4. Steady-state positioning: F = 15 N (long-term); II. Real-time data acquisition and feature extraction; In each control cycle, Read the signal from pressure sensor 250 and calculate the current insertion force. , Calculate the current force deviation Force deviation change rate and instantaneous loss function ; ; ; ; in, The preset insertion force; for The change in force deviation over time; for At any given moment, the controller outputs a control quantity to the electro-proportional valve 240. a, b All are weighting coefficients; a The value range is 0.5 to 2.0; b The value range is 0.1 to 0.5; III. Determine whether the activation condition set C for the domain transformation is satisfied, specifically including the following conditions: C1: The absolute value of the force deviation exceeds the threshold; C2: The absolute value of the force deviation exceeds the threshold and the absolute value of its rate of change is less than the threshold; C3: Instantaneous Loss Function Exceeding the threshold or instantaneous loss function The set period is continuously rising.

[0055] If any condition is satisfied, then for , Perform a nonlinear transformation to obtain the feature input. as well as Otherwise, use the raw input directly. ); ); ; ; ; ; in, This represents the force deviation value after nonlinear domain transformation. This represents the rate of change of force deviation after nonlinear domain transformation. This indicates the current mechanical state of the detonator gripper 190; The reference gain used in calculating the force deviation value; The reference gain used in calculating the rate of change of force deviation; This is the performance threshold; as well as The fine-tuning coefficients corresponding to state S; max This is a function to find the maximum value. 、 All are normalization factors; This represents the maximum allowable steady-state force deviation of the system. ΔT To control the cycle; C The control parameter is a constant, and its value ranges from 1.5 to 3.0. IV. The secondary control loop performs real-time fuzzy push and output.

[0056] Initial center of membership function of secondary control loop c With width σ The optimal parameter set for feature input as well as Perform fuzzy reasoning: During fuzzification, the Gaussian membership degree of the input to each linguistic variable is calculated.

[0057] When matching rules, according to Figure 3 The system uses a pre-defined fuzzy rule base to calculate the trigger strength of each rule.

[0058] During defuzzing, a weighted average method is used to calculate and output a set of adaptive values ​​in real time. PID Control parameters KP(k), KI (k), KD(k) .

[0059] 5. Perform PID calculations and generate the final control input.

[0060] The outer PID controller receives adaptive parameters from the real-time output of the secondary control loop. KP(k), KI(k), KD(k) and the current control cycle , .

[0061] The discrete position PID control algorithm is executed, and its control quantity is... The calculation formula is as follows: u(k) = KP(k) × + KI(k) × I(k) + KD(k) × D(k) ; KP(k) This is the proportionality coefficient; KI(k) The integral coefficient; KD(k) These are the differential coefficients; I(k)=I(k - 1)+e(k)ΔT ; I(k - 1) This is the integral value from the previous control cycle; ΔT To control the cycle.

[0062] D(k) For differential terms, D(k)=ec(k)=ΔTe(k)-e(k - 1) ; u(k) As an output, it is sent to the electro-proportional valve 240 to achieve precise and smooth adjustment of the speed of the cartridge cylinder.

[0063] VI. Update the comprehensive loss function and perform performance evaluation; The comprehensive loss function L Updated to online evaluation of loss function To evaluate the overall performance of the system within a set time window: ; in, Indicates the instantaneous loss function within the set time window. The average value; λ This is the weighting coefficient, with a value ranging from 0.1 to 0.5; Indicates the control quantity within the set time window. The standard deviation of the output; Comprehensive loss function L A global optimization objective for system performance is defined. However, because it includes an integral term, it cannot be directly computed at runtime for online learning. Therefore, an online evaluation loss function is defined. As a comprehensive loss function L The real-time approximation is obtained by the main control loop, which operates periodically and whose training objective is to minimize the current time step. Through this online optimization, the system can progressively approach the minimization of the ultimate goal. L The state.

[0064] VII. Perform parameter learning and optimization on the main control loop.

[0065] The main control circuit is activated at set intervals. The main control loop collects historical datasets from the past learning cycle, specifically including force deviation sequences. e}, the sequence of deviation change rates { ec Domain transformation activation flag sequence and corresponding .

[0066] To minimize As a training objective, the ANFIS network is trained using a hybrid learning algorithm (such as least squares estimation in forward propagation and gradient descent in backpropagation) to calculate the adjustment amount of the sub-control loop parameter set.

[0067] When performing stability constraint learning, a discriminant function is constructed. ; in e This represents the force deviation value at the current moment; The learning rate of the main control loop is dynamically adjusted based on the changing trend of the discriminant function. : like Then let Significantly reduce learning speed to prevent instability. like and Then let Normal learning.

[0068] like and Then let 1.2 Accelerate fine-tuning in areas with small errors.

[0069] This represents the preset base learning rate, with a preferred value of 0.01. This represents the stability threshold for force deviation; its value is related to the required control accuracy of the system and is typically set to 1.5 to 2 times the allowable steady-state error. For example, if the allowable steady-state error of the system is ±1N, then... 1.5N is acceptable.

[0070] The entire controller continuously runs the above steps, forming a complete process of closed-loop control and online learning.

[0071] like Figure 4 The figure shows a comparison of the force tracking performance of the control method of this application with that of traditional fixed ANFIS control and traditional PID control. It can be seen that the peak value of the control method of this application is lower than that of the traditional control method, and the target insertion force can be reached more quickly.

[0072] like Figure 5 The figure shown is a comparison of the tracking error of the control method of this application with that of traditional fixed ANFIS control and traditional PID control. It can be seen that the error of this application is smaller than that of traditional control methods. In particular, the error of traditional control methods is 100N, while the error of this application is only 20N.

[0073] like Figure 6 The diagram shows a comparison of the control signals of the control method of this application with those of traditional fixed ANFIS control and traditional PID control. It can be seen that the control method of this application has a faster response speed than the traditional control method. The control method of this application can reach the target insertion force in only 4 seconds, while the traditional control method requires 8 seconds to reach the target insertion force.

[0074] like Figure 7 The figure shows a comparison of the control energy consumption of the control method of this application with that of traditional fixed ANFIS control and traditional PID control. It can be seen that the control energy loss of the control method of this application is consistently lower than that of traditional control algorithms, and the control energy consumption is smaller compared to traditional control methods.

[0075] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.

[0076] The block diagrams of devices, apparatuses, devices, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

Claims

1. A system for automatic assembly of detonators and primers, characterized in that The system includes a second platform (120) for carrying and holding the detonator, and a detonator gripper (190) located above the second platform (120) corresponding to the vertical position of the detonator. The detonator gripper (190) is fixed to the drive end of the lifting cylinder (200) and is driven by the lifting cylinder (200) to generate a vertical lifting action to hold the detonator vertically inserted into the set position of the detonator. The push cylinder (230), the second platform (120) and the packing machine (170) are arranged in sequence along the push direction of the push cylinder (230). The assembled detonator-detonator is pushed linearly by the push cylinder (230) to the packing machine (170) for strapping.

2. The automatic assembly system for detonator-detonating bomb according to claim 1, characterized in that, A clamping cylinder (180) is provided above the baler (170). The clamping cylinder (180) is configured to clamp and fix the detonator-detonating shell when the baler (170) performs a winding and baling operation on the detonator-detonating shell. The clamping end of the clamping cylinder (180) is set to an arc-shaped surface that matches the shape of the detonating shell.

3. The automatic assembly system for detonator-detonating bomb according to claim 1, characterized in that, It also includes a detonator conveyor (160) for conveying detonators, a slide rail (260) is horizontally arranged above the second platform (120), a lifting cylinder (200) is slidably arranged on the slide rail (260), and the detonators conveyed on the detonator conveyor (160) and the detonator grippers (190) correspond to the sliding path of the lifting cylinder (200).

4. The automatic assembly system for detonator-detonating bomb according to claim 1, characterized in that, The second platform (120) is fixed on the base (100) and rotates with the base (100), with the rotation axis arranged horizontally. The base (100) is also provided with an electric push rod (130) that rotates with the second platform (120). The electric push rod (130) extends and retracts while driving the second platform (120) to pitch and adjust the pitch angle of the detonator. Two sets of clamping cylinders (210) are symmetrically arranged on the second platform (120) in the horizontal direction. The drive ends of the two sets of clamping cylinders (210) extend and retract towards each other to clamp and position the detonator. The drive end of the push cylinder (230) is provided with a pusher claw (220) that applies a clamping action to the detonator.

5. The automatic assembly system for detonator-detonating bomb according to claim 1, characterized in that, The push cylinder (230) is fixed on the first platform (110) of the base (100). The first platform (110) and the second platform (120) are arranged in a straight line. The first platform (110) and / or the second platform (120) are also equipped with a limiter (140) for regulating and constraining the detonator lead wire.

6. The automatic assembly system for detonator-detonating bomb according to claim 5, characterized in that, A pressure sensor (250) is provided between the detonator gripper (190) and the drive end of the lifting cylinder (200), and the pressure signal received by the pressure sensor (250) is transmitted to the PLC controller.

7. The automatic assembly system for detonator-detonating bomb according to claim 6, characterized in that, An electric proportional valve (240) is installed in the air circuit of the lifting cylinder (200) to adjust the cylinder speed according to the PLC command; a photoelectric gate sensor (150) for sensing the position of the detonating bomb is installed on the second platform (120).

8. An assembly method for an automatic assembly system for a detonator-detonating bomb according to any one of claims 1 to 7, characterized in that, Includes the following steps: S1. Place the detonating bomb on the second platform (120). Detect whether the detonating bomb has reached the set position through the photoelectric gate sensor (150). After reaching the set position, start the clamping cylinder (210) to clamp the detonating bomb. Then, drive the second platform (120) to adjust to the set pitch angle by the electric push rod (130). S2. The lifting cylinder (200) slides along the slide rail (260) to above the detonator conveyor (160). After the detonator gripper (190) moves down and clamps a set of detonators, the lifting cylinder (200) slides along the slide rail (260) to above the detonator. S3. The lifting cylinder (200) drives the detonator gripper (190) to insert the detonator into the detonating shell. Based on the pressure feedback from the pressure sensor (250), the PLC controller controls the electric proportional valve (240) to dynamically adjust the insertion speed of the lifting cylinder (200) so that the insertion force of the detonator is maintained within the set range. S4. After inserting the detonator to the set depth of the detonating bomb, the detonator gripper (190) releases the detonator, and the lifting cylinder (200) drives the detonator gripper (190) to reset; the electric push rod (130) drives the second platform (120) to reset to the horizontal state. S5. After the pusher gripper (220) has finished clamping the detonator-inserted explosive, it is pushed to the packer (170) by the pusher cylinder (230). S6. After the pressing cylinder (180) presses the detonator, the packing machine (170) is started to pack the detonator and the lead wire of the detonator tightly, thus completing the assembly of the detonator and the detonator. S7. After each mechanism is reset, repeat the above steps to form a cyclical operation.

9. A control method for an automatic assembly system of detonator-detonating bomb according to any one of claims 1 to 7, characterized in that, Includes the following steps: S1. Set up the controller, which specifically includes: The main control loop is configured to run periodically based on an adaptive neuro-fuzzy inference system and optimize the parameters of the sub-control loop based on historical performance data. The secondary control loop is configured to activate when preset conditions are met. After activation, it performs real-time fuzzy inference on the input force deviation and the rate of change of force deviation, and outputs adaptive PID control parameters. The loss function module is configured to evaluate system performance and guide learning and decision-making; The stability assurance module is configured to constrain the learning process of the main control loop to ensure system stability; S2. Initialize the system by setting the initial membership function parameters, comprehensive loss function, domain transformation activation condition threshold, learning period, and stability constraint parameters of the controller. S3. Read the signal from the pressure sensor (250) in real time during each control cycle and calculate the current insertion force of the detonator clamp (190). And calculate the current force deviation. Force deviation change rate and instantaneous loss function ; ； ； ； in, The preset insertion force; for The change in force deviation over time; for The control quantity output by the controller to the electric proportional valve (240) at this moment; a、b All are weighting coefficients; a The value range is 0.5 to 2.0; b The value range is 0.1 to 0.5; S4. Determine whether the preset domain transformation activation condition is met. If it is met, then... , Perform a nonlinear transformation to obtain the feature input. as well as Otherwise, use the raw input directly. )； )； ； ； ； ； in, This represents the force deviation value after nonlinear domain transformation. This represents the rate of change of force deviation after nonlinear domain transformation. This indicates the current mechanical state of the detonator gripper (190); The reference gain used in calculating the force deviation value; The reference gain used in calculating the rate of change of force deviation; This is the performance threshold; as well as For state S The corresponding fine-tuning coefficients; max This is a function to find the maximum value. 、 All are normalization factors; This represents the maximum allowable steady-state force deviation of the system. ΔT To control the cycle; C The control parameter is a constant, and its value ranges from 1.5 to 3.

0. S5. Based on the current membership function parameter set, perform feature input... as well as Perform fuzzy inference and output an adaptive set of results. PID Control parameters KP(k), KI(k), KD(k) Calculate control quantity As The final control quantity output by the controller to the electric proportional valve (240) at this moment.

10. The control method for an automatic assembly system of detonator-detonating bomb according to claim 9, characterized in that, Build an online evaluation loss function To evaluate the overall performance of the system within a set time window: ； in, Indicates the instantaneous loss function within the set time window. The average value; λ This is the weighting coefficient, with a value ranging from 0.1 to 0.5; Indicates the control quantity within the set time window. The standard deviation of the output; At set intervals, the main control loop is activated, and it collects historical data to minimize... To achieve the target, calculate the parameter adjustment amount of the secondary control loop; Build a discriminant function ; in e This represents the force deviation value at the current moment; The learning rate of the main control loop is dynamically adjusted based on the changing trend of the discriminant function. : like Then let ; like and Then let ; like and Then let 1.2 ; This represents the preset base learning rate; This represents the stability threshold of the force deviation.

Images