Comprehensive early warning method and system applied to starting gear ring automatic quenching equipment

Abstract

The invention relates to a comprehensive early warning method and system applied to starting gear ring automatic quenching equipment, and the method comprises the steps: M1, collecting the data information of the time spent by a transfer mechanism to transfer a gear ring in the quenching process of the starting gear ring to be machined by the automatic quenching equipment, the data information of the time spent on transferring the gear ring by the five-axis manipulator and the data information of the time spent on quenching the gear ring are obtained; and M2, inputting the data information of the time spent on moving the gear ring by the transfer mechanism, the data information of the time spent on transferring the gear ring by the five-axis manipulator and the data information of the time spent on quenching the gear ring into a mixed model based on a water circulation algorithm and an Elman neural network to predict the total time spent on completing gear ring machining. According to the invention, the automatic quenching process of the starting gear ring can be monitored in real time, early warning can be carried out in advance, and the production efficiency and quality of the starting gear ring are improved.

Description

Technical Field

[0001] This invention relates to the field of starting gear ring machining technology, and in particular to a comprehensive early warning method and system for automatic quenching equipment for starting gear rings. Background Technology

[0002] With the continuous development of science and technology, mechanical intelligent control technology is a rapidly developing emerging technology in recent years, mainly applied in fields such as industrial automation, intelligent manufacturing, and robotics.

[0003] In the automated machining process of starting gear rings, the starting gear rings to be processed are moved by a transfer mechanism and then transported to a quenching equipment for quenching by a five-axis robot. After quenching, they are transported to the unloading hopper for unloading. Although no human intervention is required in the entire process, if a problem occurs in any part of the process, it will lead to the failure of gear ring machining, thereby reducing the machining efficiency of the gear rings. How to monitor the entire process of automatic quenching of starting gear rings in real time and provide early warning has become an urgent problem for us to solve. Summary of the Invention

[0004] In view of the above problems, the present invention provides a comprehensive early warning method and system for automatic quenching equipment for starting gear rings. It can not only monitor the automatic quenching process of starting gear rings in real time and provide early warning, but also improve the production efficiency and quality of starting gear rings.

[0005] To achieve the above and other related objectives, the present invention provides the following technical solution: A comprehensive early warning method for automatic quenching equipment for starting gear rings, the method comprising: M1. During the quenching process of the starting gear ring to be processed by the automatic quenching equipment, data on the time spent by the transfer mechanism to transfer the gear ring is collected, and data on the time spent by the five-axis robot to transfer the gear ring and the time spent quenching the gear ring are also obtained. M2. Input the data information of the time spent by the transfer mechanism to move the gear ring, the data information of the time spent by the five-axis robot to transfer the gear ring, and the data information of the time spent by the gear ring quenching into the hybrid model based on the water circulation algorithm and the Elman neural network to predict the total time spent to complete the gear ring processing, and obtain the predicted data information of the total time spent to complete the gear ring processing. M3. Based on the predicted total time spent on completing the gear ring machining, an improved snake optimization algorithm is used to optimize the total time of gear ring machining, resulting in optimized data on the total time of gear ring machining. M4. Based on the data information of the total time of the optimized gear ring processing, a preset threshold is set. If the total time of the optimized gear ring processing is less than the preset threshold, the gear ring processing task is normal. If the total time of the optimized gear ring processing is greater than the preset threshold, the gear ring processing task is abnormal and an alarm is issued.

[0006] Furthermore, in step M2, inputting the data on the time spent by the transfer mechanism moving the gear ring, the data on the time spent by the five-axis robot transferring the gear ring, and the data on the time spent quenching the gear ring into a hybrid model based on the water circulation algorithm and the Elman neural network to predict the total time spent completing the gear ring machining includes: M21. Initialize the parameters required for the Elman neural network and create a new Elman neural network; M22. Initialize the parameters of the water cycle algorithm, including individual parameters, maximum number of iterations, upper bound of parameters, and lower bound of parameters; The M23 water cycle algorithm iteratively optimizes the Elman neural network by updating the position until the iteration ends and the optimal individual is formed. The optimal individual includes network weights and thresholds. M24. The Elman neural network is used to analyze the optimal individual, and the Elman neural network is trained using the analyzed network weights and thresholds to obtain a trained Elman neural network. M25. Input the data information of the time spent by the transfer mechanism to move the gear ring, the data information of the time spent by the five-axis robot to transfer the gear ring, and the data information of the time spent quenching the gear ring into the trained Elman neural network to predict the total time spent to complete the processing of the gear ring, and obtain the predicted data information of the total time spent to complete the processing of the gear ring.

[0007] Furthermore, the individual parameters include ocean-type individual parameters, river-type individual parameters, and stream-type individual parameters, with ocean-type individual parameters being superior to river-type individual parameters, and river-type individual parameters being superior to stream-type individual parameters.

[0008] Furthermore, the newly created Elman neural network is a net. net=newelm(minmax(input),[11,1],{“tansig”,“purelin”},“traingdx”), where input is the training set data, tansig is the activation function of the hidden layer, purelin is the activation function of the output layer, and traingdx is the gradient descent function.

[0009] Furthermore, in step M3, optimizing the total machining time of the gear ring using the improved snake optimization algorithm includes: M31. Based on the predicted total time taken to complete the gear ring machining, a chaotic sequence function P is constructed to represent the total time taken to complete the gear ring machining. , Where t is the data information of the total time spent in the predicted gear ring processing, r(t) is the normalization function of the total time spent in the gear ring processing, and h is any constant parameter between 0 and 1. The chaotic sequence of the total time spent in the gear ring processing is characterized to obtain the data information of the chaotic sequence of the total time spent in the gear ring processing. M32. Based on the chaotic sequence data information of the total time spent in processing the gear ring, initialize the snake population, determine the population parameters and the maximum number of iterations K, and obtain the data information of the initialized snake population; M33. Based on the initialized snake population data, establish a position update function S. , Where G represents the data of the snake population after initialization, ω is the correction factor, and ρ is the random adjustment parameter. The total processing time of the gear ring is optimized to obtain the data of the optimized total processing time of the gear ring.

[0010] Furthermore, the normalized function r(t) for the total time spent machining the gear ring is, , Where t represents the total time expected to be taken to complete the machining of the predicted gear ring.

[0011] Furthermore, the correction factor ω is any constant parameter between 0 and 1, and the constraint range of the random adjustment parameter ρ is any constant parameter between 1 and 2.

[0012] To achieve the above and other related objectives, the present invention also provides a comprehensive early warning system for an automatic quenching equipment for starting gear rings, used to implement the aforementioned comprehensive early warning method for the automatic quenching equipment for starting gear rings, the system comprising: The information collection module is used to collect data on the time spent by the transfer mechanism in transferring the gear ring, and to obtain data on the time spent by the five-axis robot in transporting the gear ring and the time spent in quenching the gear ring. The total time prediction module for completing the gear ring machining is connected to the information collection module. It is used to input the data information of the time spent by the transfer mechanism to move the gear ring, the data information of the time spent by the five-axis robot to transfer the gear ring, and the data information of the time spent by the gear ring quenching into a hybrid model based on water circulation algorithm and Elman neural network to predict the total time spent completing the gear ring machining, and obtain the predicted data information of the total time spent completing the gear ring machining. The total time optimization module for completing gear ring machining is connected to the total time prediction module for completing gear ring machining. It is used to optimize the total time of gear ring machining using an improved snake optimization algorithm to obtain the data information of the optimized total time of gear ring machining. The threshold judgment module is connected to the total time optimization module for completing the gear ring processing. It is used to set a preset threshold. If the optimized total time for gear ring processing is less than the preset threshold, the gear ring processing task is normal. If the optimized total time for gear ring processing is greater than the preset threshold, the gear ring processing task is abnormal and an alarm is issued.

[0013] Furthermore, the system also includes a voice broadcast module connected to the threshold judgment module, used to acquire the status of the gear ring machining task in real time and broadcast it.

[0014] Furthermore, the system also includes a human-computer interaction module connected to the threshold judgment module, used to turn the device on or off, and to display the working status of the gear ring machining.

[0015] The present invention has the following positive effects: 1. This invention predicts the total time required to complete gear ring processing using a hybrid model based on a water circulation algorithm and an Elman neural network, and further optimizes the total processing time using an improved snake optimization algorithm. This process not only predicts the total processing time to improve the anticipation of warnings and prevent the production of defective gear rings, but also optimizes the total processing time to improve the accuracy of warnings, thereby preventing false alarms.

[0016] 2. By setting a preset threshold, the gear ring processing task is normal if the total processing time of the optimized gear ring is less than the preset threshold, and abnormal if the total processing time of the optimized gear ring is greater than the preset threshold, and an alarm is issued. This not only enables real-time monitoring and early warning of the automatic quenching process of the starting gear ring, but also improves the production efficiency and quality of the starting gear ring. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the automatic quenching equipment for the starting gear ring of the present invention; Figure 2This is a schematic diagram of the method flow of the present invention; Figure 3 This is a schematic diagram illustrating the workflow of the hybrid model based on the water cycle algorithm and the Elman neural network of the present invention. Figure 4 This is a flowchart illustrating the improved snake optimization algorithm of the present invention; Figure 5 This is a schematic diagram of the system framework of the present invention.

[0018] The labels in the diagram are as follows: 1-Transfer mechanism, 2-Five-axis robot, 3-Quenching machine, 4-Transfer platform, 5-Unloading bin. Detailed Implementation

[0019] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0020] Example 1: As Figure 1 or Figure 2 As shown, a comprehensive early warning method for automatic quenching equipment for starting gear rings is provided, the method comprising: First, the transfer mechanism 1 transfers the starting gear ring to be processed onto the transfer platform 4. Then, the five-axis robot 2 transfers the starting gear ring to be processed from the transfer platform 4 to the quenching machine 3 for quenching. Finally, the five-axis robot 2 transfers the quenched gear ring to the unloading bin 5 to complete the processing of the gear ring.

[0021] M1. During the quenching process of the starting gear ring to be processed by the automatic quenching equipment, data information on the time spent by the transfer mechanism 1 to transfer the gear ring is collected, and data information on the time spent by the five-axis robot to transfer the gear ring 2 and the time spent by the gear ring to quench in the quenching machine 3 are also obtained. M2. Input the data information of the time spent by the transfer mechanism to move the gear ring, the data information of the time spent by the five-axis robot to transfer the gear ring, and the data information of the time spent by the gear ring quenching into the hybrid model based on the water circulation algorithm and the Elman neural network to predict the total time spent to complete the gear ring processing, and obtain the predicted data information of the total time spent to complete the gear ring processing. M3. Based on the predicted total time spent on completing the gear ring machining, an improved snake optimization algorithm is used to optimize the total time of gear ring machining, resulting in optimized data on the total time of gear ring machining. M4. Based on the data information of the total time of the optimized gear ring processing, a preset threshold is set. If the total time of the optimized gear ring processing is less than the preset threshold, the gear ring processing task is normal. If the total time of the optimized gear ring processing is greater than the preset threshold, the gear ring processing task is abnormal and an alarm is issued.

[0022] In this embodiment, as Figure 3 As shown, in step M2, the process of inputting the data on the time spent by the transfer mechanism moving the gear ring, the data on the time spent by the five-axis robot transferring the gear ring, and the data on the time spent quenching the gear ring into a hybrid model based on a water circulation algorithm and an Elman neural network to predict the total time spent on completing the gear ring machining includes: M21. Initialize the parameters required for the Elman neural network and create a new Elman neural network; M22. Initialize the parameters of the water cycle algorithm, including individual parameters, maximum number of iterations, upper bound of parameters, and lower bound of parameters; The M23 water cycle algorithm iteratively optimizes the Elman neural network by updating the position until the iteration ends and the optimal individual is formed. The optimal individual includes network weights and thresholds. M24. The Elman neural network is used to analyze the optimal individual, and the Elman neural network is trained using the analyzed network weights and thresholds to obtain a trained Elman neural network. M25. Input the data information of the time spent by the transfer mechanism to move the gear ring, the data information of the time spent by the five-axis robot to transfer the gear ring, and the data information of the time spent quenching the gear ring into the trained Elman neural network to predict the total time spent to complete the processing of the gear ring, and obtain the predicted data information of the total time spent to complete the processing of the gear ring.

[0023] In this embodiment, the individual parameters include ocean-type individual parameters, river-type individual parameters, and stream-type individual parameters. The ocean-type individual parameters are superior to the river-type individual parameters, and the river-type individual parameters are superior to the stream-type individual parameters.

[0024] In this embodiment, the newly created Elman neural network is a net. net=newelm(minmax(input),[11,1],{“tansig”,“purelin”},“traingdx”), where input is the training set data, tansig is the activation function of the hidden layer, purelin is the activation function of the output layer, and traingdx is the gradient descent function.

[0025] In this embodiment, as Figure 4As shown, in step M3, optimizing the total machining time of the gear ring using the improved snake optimization algorithm includes: M31. Based on the predicted total time taken to complete the gear ring machining, a chaotic sequence function P is constructed to represent the total time taken to complete the gear ring machining. , Where t is the data information of the total time spent in the predicted gear ring processing, r(t) is the normalization function of the total time spent in the gear ring processing, and h is any constant parameter between 0 and 1. The chaotic sequence of the total time spent in the gear ring processing is characterized to obtain the data information of the chaotic sequence of the total time spent in the gear ring processing. M32. Based on the chaotic sequence data information of the total time spent in processing the gear ring, initialize the snake population, determine the population parameters and the maximum number of iterations K, and obtain the data information of the initialized snake population; M33. Based on the initialized snake population data, establish a position update function S. , Where G represents the data of the snake population after initialization, ω is the correction factor, and ρ is the random adjustment parameter. The total processing time of the gear ring is optimized to obtain the data of the optimized total processing time of the gear ring.

[0026] In this embodiment, the normalized function r(t) for the total time spent machining the gear ring is, , Where t represents the total time expected to be taken to complete the machining of the predicted gear ring.

[0027] In this embodiment, the correction factor ω is any constant parameter between 0 and 1, and the constraint range of the random adjustment parameter ρ is any constant parameter between 1 and 2.

[0028] In this embodiment, during the quenching process of the starting gear ring to be processed by the automatic quenching equipment, the time t1 spent by the transfer mechanism to complete the gripping and placement of the gear ring is collected in real time, the total time t2 spent by the five-axis robot from the loading position to the heating position and then to the quenching position is obtained, and the actual time t3 spent by the gear ring in the sensor to complete the entire process of heating and water spraying cooling is recorded. All time data are sampled with millisecond precision and stored in the local database.

[0029] The three types of time series data (t1, t2, and t3) are input into a hybrid prediction model based on the Water Cycle Algorithm (WCA) optimized by an Elman neural network. The model structure includes an input layer (3 nodes corresponding to the three time inputs), a hidden layer (the optimal number of nodes is determined by WCA optimization), a context layer (receiving historical state information), and an output layer (1 node outputting the predicted total processing time T_pred). WCA is used to globally search for the optimal combination of initial weights and thresholds, avoiding the traditional gradient descent method from getting stuck in local minima; the Elman network, due to its dynamic memory characteristics, is suitable for processing time-dependent industrial process data. After training, the model can receive a new batch of t1, t2, and t3 inputs in real time and output the predicted result T of the total time T required to complete the gear ring processing in the next cycle. pred .

[0030] Based on the predicted T pred An improved Snake Optimization Algorithm (ISOA) is employed to perform reverse optimization of the time allocation for each sub-process. ISOA introduces a population diversity preservation mechanism and an adaptive step-size strategy on top of the original SOA, making it converge faster and more robust when solving multivariate constrained optimization problems. The objective function is set to minimize the total processing time, with constraints including: t1 ≥ t 1min (Mechanical limit), t2∈[t 2ower , t 2upper (Safe exercise range), t3≥t 3critical (Ensure sufficient hardening). Through iterative optimization, obtain a theoretically optimal time configuration (t). 1opt , t 2opt , t 3opt Then, the optimized total processing time T is calculated. opt .

[0031] M4. Set the preset threshold T threshold This value can be determined by adding three times the standard deviation to the average processing time under historical normal operating conditions, i.e., T. threshold =μ+3σ. Compare T opt With T threshold If T opt ≤T threshold If T is normal, the system continues to run; otherwise, it is determined that the current processing task is in a normal state. opt >T threshold If the alarm is triggered, a warning mechanism will be established, generating a three-level alarm signal. The first level is a pop-up notification on the visual interface, the second level is the activation of the audible and visual alarm, and the third level is the sending of a shutdown suggestion instruction to the MES system for the operator to confirm whether to suspend subsequent batches of equipment.

[0032] The method further includes: establishing a time deviation tracing module, which automatically analyzes (t) when the early warning is activated. 1opt -t1), (t 2opt -t2), (t 3opt The three differences (-t3) are used to pinpoint the main source of the delay and push maintenance suggestions to the corresponding workstations.

[0033] The technical inspiration for this invention comes from the natural water cycle and the information transmission mechanism of biological nervous systems. WCA simulates the process of a river flowing to the sea to optimize parameters, giving the model a stronger learning starting point; the Elman network mimics the short-term memory function of the human brain, effectively capturing dynamic features in time series; and ISOA, inspired by the predatory behavior of snakes, efficiently explores the solution space through a meandering approach, achieving precise control over the processing rhythm. The combination of these three technologies forms an intelligent closed loop of "prediction-optimization-judgment."

[0034] Example 2: Based on the comprehensive early warning method applied to the automatic quenching equipment for starting gear rings in Example 1, the present invention will be further explained and described below.

[0035] like Figure 1 or Figure 2 As shown, a comprehensive early warning method for an automatic quenching equipment for starting gear rings is provided, the method comprising: M1. During the quenching process of the starting gear ring to be processed by the automatic quenching equipment, data information on the time spent by the transfer mechanism 1 to transfer the gear ring is collected, and data information on the time spent by the five-axis robot to transfer the gear ring 2 and the time spent by the gear ring to quench in the quenching machine 3 are also obtained. M2. Input the data information of the time spent by the transfer mechanism to move the gear ring, the data information of the time spent by the five-axis robot to transfer the gear ring, and the data information of the time spent by the gear ring quenching into the hybrid model based on the water circulation algorithm and the Elman neural network to predict the total time spent to complete the gear ring processing, and obtain the predicted data information of the total time spent to complete the gear ring processing. M3. Based on the predicted total time spent on completing the gear ring machining, an improved snake optimization algorithm is used to optimize the total time of gear ring machining, resulting in optimized data on the total time of gear ring machining. M4. Based on the data information of the total time of the optimized gear ring processing, a preset threshold is set. If the total time of the optimized gear ring processing is less than the preset threshold, the gear ring processing task is normal. If the total time of the optimized gear ring processing is greater than the preset threshold, the gear ring processing task is abnormal and an alarm is issued.

[0036] In this embodiment, the production line is configured as follows: a servo-driven transfer mechanism is responsible for feeding the gear ring from the conveyor belt into the positioning fixture, with a repeatability of ±0.02mm; a five-axis robot is equipped with a torque sensor, with a maximum load of 10kg, and an average time of 18.5s to complete one complete transfer path; a medium-frequency induction heating device with a power of 100kW, a set heating time of 8s, a cooling water pressure of 0.6MPa, and a total quenching time of approximately 12.3s for a single quenching.

[0037] The system ran continuously for 30 days, collecting a total of 2160 sets of valid processing cycle data. The raw data showed that t1 had a mean of 5.2s and a standard deviation of 0.4s; t2 had a mean of 18.7s and a standard deviation of 1.1s; t3 had a mean of 12.5s and a standard deviation of 0.9s; and the total processing time had a mean of 36.4s and a standard deviation of 1.8s. Based on this, T was set... threshold =36.4 + 3 × 1.8 = 41.8 s.

[0038] A WCA-Elman hybrid model was constructed with a population size of 50, a maximum number of iterations of 200, a learning rate of 0.01, and a momentum factor of 0.9. After 1000 training iterations, the model's mean absolute error (MAE) on the test set was 0.63 s, and the coefficient of determination R0 was [missing value]. 2 The performance reached 0.94, significantly outperforming the use of a BP neural network alone (MAE=1.37s, R0). 2 =0.82) or ARIMA model (MAE=1.89s, R 2 =0.71).

[0039] Next, ISOA was applied for cycle time optimization. The population size was set to 30, the maximum iterations to 150 rounds, and Cauchy mutation was introduced to enhance the escape capability. Optimization results showed that the theoretically optimal total time was 34.1s, a reduction of 2.3s compared to the current average. This was mainly due to the robot's path replanning reducing t2 to 17.2s, and the fine-tuning of cooling parameters stabilizing t3 at 11.8s.

[0040] During subsequent operation, the system detected that the t2 of a certain batch suddenly rose to 20.5s. Although it did not exceed the limit, the predicted T pred It rose to 38.9s, close to the threshold. ISOA recalculated T. opt =42.6s> T threshold An early warning was immediately issued. Investigation revealed insufficient lubrication in the sixth axis bearing of the robotic arm; timely maintenance restored normal operation. This early warning successfully prevented a production line backlog caused by subsequent consecutive timeouts.

[0041] In this embodiment, as Figure 5As shown, this invention provides a comprehensive early warning system for an automatic quenching equipment for starting gear rings, used to implement the aforementioned comprehensive early warning method for the automatic quenching equipment for starting gear rings. The system includes: The information collection module is used to collect data on the time spent by the transfer mechanism in transferring the gear ring, and to obtain data on the time spent by the five-axis robot in transporting the gear ring and the time spent in quenching the gear ring. The total time prediction module for completing the gear ring machining is connected to the information collection module. It is used to input the data information of the time spent by the transfer mechanism to move the gear ring, the data information of the time spent by the five-axis robot to transfer the gear ring, and the data information of the time spent by the gear ring quenching into a hybrid model based on water circulation algorithm and Elman neural network to predict the total time spent completing the gear ring machining, and obtain the predicted data information of the total time spent completing the gear ring machining. The total time optimization module for completing gear ring machining is connected to the total time prediction module for completing gear ring machining. It is used to optimize the total time of gear ring machining using an improved snake optimization algorithm to obtain the data information of the optimized total time of gear ring machining. The threshold judgment module is connected to the total time optimization module for completing the gear ring processing. It is used to set a preset threshold. If the optimized total time for gear ring processing is less than the preset threshold, the gear ring processing task is normal. If the optimized total time for gear ring processing is greater than the preset threshold, the gear ring processing task is abnormal and an alarm is issued.

[0042] In this embodiment, the system further includes a voice broadcast module connected to the threshold judgment module, used to acquire the status of the gear ring machining task in real time and broadcast it.

[0043] In this embodiment, the system further includes a human-computer interaction module connected to the threshold judgment module, used to turn the device on or off, and to display the working status of the gear ring machining.

[0044] In this embodiment, a shift from "post-event alarm" to "pre-event warning" is achieved. Traditional systems mostly rely on real-time data exceeding limits to trigger alarms, which is essentially reactive control; while this invention, by predicting future processing time and combining optimization results for risk assessment, can identify potential hazards before anomalies occur, truly achieving preventative maintenance.

[0045] In this embodiment, prediction accuracy and system robustness are improved. WCA is used to initialize the Elman network parameters, effectively overcoming the training instability caused by random parameter assignment, enabling the model to maintain high prediction accuracy even in complex noisy environments. Experimental data shows that its MAE is less than 1 second, fully meeting the requirements of industrial field control.

[0046] In this embodiment, the system's adaptive adjustment capability is enhanced. The introduction of ISOA not only assists in judgment but also provides quantitative basis for adjusting process parameters, driving the system towards "autonomy." In particular, it exhibits good fault tolerance and recovery capabilities when facing disturbances such as raw material differences and changes in environmental temperature and humidity.

[0047] This embodiment demonstrates good engineering applicability and scalability. The method does not rely on expensive sensors; data acquisition can be completed using only the device's built-in encoder and PLC timestamps, resulting in low deployment costs. Furthermore, its framework is not limited to gear ring quenching scenarios; with minor modifications, it can be migrated to other automated assembly, welding, painting, and other process-oriented operations.

[0048] The present invention provides a computer-readable storage medium storing a computer program programmed or configured to perform the comprehensive early warning method applied to an automatic quenching device for starting gear rings.

[0049] Any references to memory, storage, database, or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0050] In summary, this invention not only enables real-time monitoring and early warning of the automatic quenching process of the starting gear ring, but also improves the production efficiency and quality of the starting gear ring.

[0051] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A comprehensive early warning method for automatic quenching equipment for starting gear rings, characterized in that, The method includes: M1. During the quenching process of the starting gear ring to be processed by the automatic quenching equipment, data on the time spent by the transfer mechanism to transfer the gear ring is collected, and data on the time spent by the five-axis robot to transfer the gear ring and the time spent quenching the gear ring are also obtained. M2. Input the data information of the time spent by the transfer mechanism to move the gear ring, the data information of the time spent by the five-axis robot to transfer the gear ring, and the data information of the time spent by the gear ring quenching into the hybrid model based on the water circulation algorithm and the Elman neural network to predict the total time spent to complete the gear ring processing, and obtain the predicted data information of the total time spent to complete the gear ring processing. M3. Based on the predicted total time spent on completing the gear ring machining, an improved snake optimization algorithm is used to optimize the total time of gear ring machining, resulting in optimized data on the total time of gear ring machining. M4. Based on the data information of the total time of the optimized gear ring processing, a preset threshold is set. If the total time of the optimized gear ring processing is less than the preset threshold, the gear ring processing task is normal. If the total time of the optimized gear ring processing is greater than the preset threshold, the gear ring processing task is abnormal and an alarm is issued.

2. The comprehensive early warning method for automatic quenching equipment for starting gear rings according to claim 1, characterized in that, In step M2, the process of inputting the data on the time spent by the transfer mechanism moving the gear ring, the data on the time spent by the five-axis robot transferring the gear ring, and the data on the time spent quenching the gear ring into a hybrid model based on a water circulation algorithm and an Elman neural network to predict the total time spent on completing the gear ring machining includes: M21. Initialize the parameters required for the Elman neural network and create a new Elman neural network; M22. Initialize the parameters of the water cycle algorithm, including individual parameters, maximum number of iterations, upper bound of parameters, and lower bound of parameters; The M23 water cycle algorithm iteratively optimizes the Elman neural network by updating the position until the iteration ends and the optimal individual is formed. The optimal individual includes network weights and thresholds. M24. The Elman neural network is used to analyze the optimal individual, and the Elman neural network is trained using the analyzed network weights and thresholds to obtain a trained Elman neural network. M25. Input the data information of the time spent by the transfer mechanism to move the gear ring, the data information of the time spent by the five-axis robot to transfer the gear ring, and the data information of the time spent quenching the gear ring into the trained Elman neural network to predict the total time spent to complete the processing of the gear ring, and obtain the predicted data information of the total time spent to complete the processing of the gear ring.

3. The comprehensive early warning method for automatic quenching equipment for starting gear rings according to claim 2, characterized in that: The individual parameters include ocean-type individual parameters, river-type individual parameters, and stream-type individual parameters. The ocean-type individual parameters are superior to the river-type individual parameters, and the river-type individual parameters are superior to the stream-type individual parameters.

4. The comprehensive early warning method for automatic quenching equipment for starting gear rings according to claim 2, characterized in that: The newly created Elman neural network is named net. net=newelm(minmax(input),[11,1],{"tansig”,"purelin”},"traingdx”), where input is the training set data, tansig is the activation function of the hidden layer, purelin is the activation function of the output layer, and traingdx is the gradient descent function.

5. The comprehensive early warning method for automatic quenching equipment for starting gear rings according to claim 1, characterized in that, In step M3, optimizing the total machining time of the gear ring using the improved snake optimization algorithm includes: M31. Based on the predicted total time taken to complete the gear ring machining, a chaotic sequence function P is constructed to represent the total time taken to complete the gear ring machining. ， Where t is the data information of the total time spent in the predicted gear ring processing, r(t) is the normalization function of the total time spent in the gear ring processing, and h is any constant parameter between 0 and 1. The chaotic sequence of the total time spent in the gear ring processing is characterized to obtain the data information of the chaotic sequence of the total time spent in the gear ring processing. M32. Based on the chaotic sequence data information of the total time spent in processing the gear ring, initialize the snake population, determine the population parameters and the maximum number of iterations K, and obtain the data information of the initialized snake population; M33. Based on the initialized snake population data, establish a position update function S. ， Where G represents the data of the snake population after initialization, ω is the correction factor, and ρ is the random adjustment parameter. The total processing time of the gear ring is optimized to obtain the data of the optimized total processing time of the gear ring.

6. The comprehensive early warning method for automatic quenching equipment for starting gear rings according to claim 5, characterized in that: The normalized function r(t) for the total time taken to complete the machining of the gear ring is: ， Where t represents the total time expected to be taken to complete the machining of the predicted gear ring.

7. The comprehensive early warning method for automatic quenching equipment for starting gear rings according to claim 5, characterized in that: The correction factor ω is any constant parameter between 0 and 1, and the random adjustment parameter ρ is constrained to any constant parameter between 1 and 2.

8. A comprehensive early warning system for automatic quenching equipment for starting gear rings, characterized in that, For implementing the comprehensive early warning method for an automatic quenching equipment for starting gear rings as described in any one of claims 1-7, the system comprises: The information collection module is used to collect data on the time spent by the transfer mechanism in transferring the gear ring, and to obtain data on the time spent by the five-axis robot in transporting the gear ring and the time spent in quenching the gear ring. The total time prediction module for completing the gear ring machining is connected to the information collection module. It is used to input the data information of the time spent by the transfer mechanism to move the gear ring, the data information of the time spent by the five-axis robot to transfer the gear ring, and the data information of the time spent by the gear ring quenching into a hybrid model based on water circulation algorithm and Elman neural network to predict the total time spent completing the gear ring machining, and obtain the predicted data information of the total time spent completing the gear ring machining. The total time optimization module for completing gear ring machining is connected to the total time prediction module for completing gear ring machining. It is used to optimize the total time for gear ring machining using an improved snake optimization algorithm to obtain the data information of the optimized total time for gear ring machining. The threshold judgment module is connected to the total time optimization module for completing the gear ring processing. It is used to set a preset threshold. If the optimized total time for gear ring processing is less than the preset threshold, the gear ring processing task is normal. If the optimized total time for gear ring processing is greater than the preset threshold, the gear ring processing task is abnormal and an alarm is issued.

9. The comprehensive early warning system for automatic quenching equipment for starting gear rings according to claim 8, characterized in that, The system also includes a voice broadcast module, which is connected to the threshold judgment module, for real-time acquisition of the status of the gear ring machining task and broadcasting it.

10. The comprehensive early warning system for automatic quenching equipment for starting gear rings according to claim 8, characterized in that, The system also includes a human-computer interaction module connected to the threshold judgment module, used to turn the device on or off and to display the working status of the gear ring machining.

Images