A dynamic scheduling method and system for impeller processing production line resources

Abstract

The present invention relates to a method and system for dynamic scheduling of resources for an impeller processing production line, and relates to the field of scheduling management. By utilizing the models and service times of the five-axis processing, blade polishing processing, heat treatment processing, and dynamic balancing debugging equipment on the target production line, a service time difference weight distribution is performed on the impeller processing log, and then the process times of the processing log are sorted and the centralized value is evaluated according to the weight distribution, and the centralized time of each type of processing is obtained. Based on this, the resource scheduling of the impeller processing production line is performed, and the technical problem that the process time of the impeller processing production line cannot adapt to the actual operating status and dynamic changes of the production line equipment for resource scheduling, resulting in low resource scheduling accuracy and affecting the overall production efficiency and product quality is solved. The technical effect of being able to adapt the resource scheduling adjustment in combination with the actual operating status and dynamic changes of the production line equipment is achieved, thereby improving the accuracy of resource scheduling and production efficiency.

Description

Technical Field

[0001] The present invention relates to the field of scheduling management, and in particular to a method and system for dynamically scheduling resources of an impeller processing production line. Background Art

[0002] In traditional impeller processing lines, process durations are typically set as static data. These data are often based on historical experience or preliminary estimates, and fail to fully consider the actual operating status and dynamic changes of the production line equipment. For example, factors such as the equipment's service life and processing quality stability are not considered in their impact on processing time. This static scheduling approach results in low resource scheduling accuracy, making it ineffective in responding to unexpected situations and equipment performance changes during the production process, thereby affecting overall production efficiency and product quality.

[0003] As the manufacturing industry evolves toward intelligent and sophisticated processes, higher requirements are placed on the control and management of production processes. Traditional static scheduling methods are no longer able to meet the needs of modern impeller processing lines. A more intelligent production line management approach is urgently needed to improve resource utilization efficiency, reduce waste, and ensure the stability and controllability of the production process. Summary of the Invention

[0004] The present invention aims to solve the technical problem in the prior art that the time taken by the impeller processing production line process is too long to adapt to the actual operating status and dynamic changes of the production line equipment for resource scheduling, resulting in low resource scheduling accuracy and affecting the overall production efficiency and product quality. A method and system for dynamic scheduling of impeller processing production line resources are provided to solve the problem.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] In the first aspect, the present invention provides a method for dynamic scheduling of impeller processing production line resources, including: based on the five-axis processing equipment model and the first service time, the blade polishing processing equipment model and the second service time, the heat treatment processing equipment model and the third service time, and the dynamic balancing and debugging equipment model and the fourth service time of the target production line, performing service time difference weight distribution on several impeller processing logs to obtain a processing log weight distribution, wherein the service time difference weight distribution rule is that the greater the service time difference, the smaller the processing log weight; based on the processing log weight distribution, sorting the process time of several impeller processing logs, and performing centralized value evaluation on the selected process time to obtain the five-axis processing centralized time, blade polishing processing centralized time, heat treatment processing centralized time and dynamic balancing and debugging centralized time, and executing impeller processing production line resource scheduling.

[0007] In a second aspect, the present invention provides a dynamic scheduling system for resources of an impeller processing production line, the system comprising: a weight distribution module for performing service time difference weight distribution on a number of impeller processing logs based on the five-axis processing equipment model and the first service time, the blade polishing processing equipment model and the second service time, the heat treatment processing equipment model and the third service time, and the dynamic balancing and debugging equipment model and the fourth service time of the target production line, to obtain a processing log weight distribution, wherein the service time difference weight distribution rule is that the greater the service time difference, the smaller the processing log weight; a resource scheduling module for sorting the process durations of a number of impeller processing logs based on the processing log weight distribution, and performing centralized value evaluation on the selected process durations, to obtain the concentrated time of five-axis processing, the concentrated time of blade polishing processing, the concentrated time of heat treatment processing, and the concentrated time of dynamic balancing and debugging, and to perform impeller processing production line resource scheduling.

[0008] The beneficial effects of the present invention are as follows: by utilizing the models and service times of the five-axis machining, blade polishing, heat treatment and dynamic balancing equipment on the target production line, a service time difference weight distribution is performed on the impeller machining log, and then the process times of the machining log are sorted and the centralized value is evaluated according to the weight distribution, and the centralized time of each type of machining is obtained, and the impeller machining production line resource scheduling is executed based on this, achieving the technical effect of being able to adapt the resource scheduling and adjustment in combination with the actual operating status and dynamic changes of the production line equipment, thereby improving the accuracy of resource scheduling and production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A schematic flow chart of a method for dynamically scheduling resources for an impeller processing production line provided by the present invention.

[0010] Figure 2 This is a structural schematic diagram of a dynamic resource scheduling system for an impeller processing production line provided by the present invention.

[0011] Description of the reference numerals: weight distribution module 11, resource scheduling module 12. DETAILED DESCRIPTION

[0012] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0013] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the specified features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0014] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or illustration". Any embodiment of the present invention described as "for example" is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed herein.

[0015] Example 1:

[0016] like Figure 1 As shown, an embodiment of the present invention provides a method for dynamically scheduling resources of an impeller processing production line, including:

[0017] S10: Based on the five-axis machining equipment model and the first service time, the blade polishing machining equipment model and the second service time, the heat treatment machining equipment model and the third service time, and the dynamic balancing and debugging equipment model and the fourth service time of the target production line, a service time difference weight distribution is performed on several impeller machining logs to obtain a machining log weight distribution, wherein the service time difference weight distribution rule is that the larger the service time difference, the smaller the machining log weight.

[0018] For example, the impeller is a key component in rotating machinery such as centrifugal pumps, fans, and turbines. It is mainly composed of blades and a hub, and is used to convert mechanical energy into kinetic energy and pressure energy of the fluid to achieve fluid transportation or pressurization functions.

[0019] The impeller processing production line generally includes: five-axis machining equipment, which is used to perform high-precision machining of complex curved surfaces of impellers to ensure that the geometric shape and dimensional accuracy of the impeller meet the design requirements; blade polishing equipment, which polishes the impeller blades to remove machining marks, improve the smoothness of the blade surface, reduce the resistance during fluid flow, and improve equipment efficiency; heat treatment equipment improves the organizational structure and performance of the impeller material through heating, insulation and cooling processes, and improves the hardness, wear resistance and corrosion resistance of the impeller; finally, dynamic balancing debugging equipment is used to perform dynamic balancing test and debugging on the processed impeller to ensure that the impeller has small vibration and smooth operation during rotation, thereby extending the service life of the equipment and improving operational safety.

[0020] In this solution's dynamic resource scheduling method for the impeller machining production line, the core equipment involved and its characteristic parameters must first be clearly defined. This includes the five-axis machining equipment model and its corresponding first service life, the blade polishing equipment model and its second service life, the heat treatment equipment model and its third service life, and the dynamic balancing and debugging equipment model and its fourth service life. These equipment models reflect the hardware configuration level of the production line, while service life directly correlates to the degree of performance degradation of the equipment. For example, if a five-axis machining center configured on a production line has been in operation for 36 months, its first service life data will serve as an important input for weight calculation.

[0021] The weight distribution rule for service time differences is clearly defined as follows: the greater the equipment service time difference, the smaller the weight assigned to the corresponding processing log. In practice, a baseline for equipment service time needs to be established. For example, the service time of new equipment is set to 0 months, and a time difference gradient is formed as the service time increases. Assuming that the blade polishing equipment has a service time of 24 months, forming a 24-month time difference with the baseline, while the dynamic balancing and debugging equipment of the same period has only been in service for 6 months, with a time difference of 6 months, the weight of the processing log of the former will be lower than that of the latter. This weight distribution mechanism is implemented through a mathematical model, and the negative correlation between the time difference and weight can be quantified using an exponential decay function or a piecewise linear function.

[0022] The process of obtaining a weighted distribution for processing logs involves three steps: data collection, time difference calculation, and weight assignment. For example, for heat treatment equipment, if a certain type of box furnace (with a third-term service life of 18 months) generates N processing logs within a period T, each log must be annotated with metadata such as the equipment model, start and end times, and processing parameters. The system calculates the precise time difference by comparing the equipment's commissioning date with the current date and then generates a weighted value within the range [0, 1] based on a preset algorithm. When multiple pieces of the same type of equipment are operating simultaneously on a production line, the time difference must be calculated and assigned independently for each piece of equipment.

[0023] The adaptability of the above steps is reflected in the precise capture of the dynamic characteristics of the production line. Assuming that in a batch of impeller processing tasks, the spindle accuracy of the five-axis processing equipment has decreased due to long-term service, the reduction in the weight of its processing log will prompt the system to give priority to the processing data of new equipment or equipment with low time difference as the basis for scheduling. In the blade polishing process, if it is detected that the processing log weight of a certain equipment (the second service time is 24 months) is continuously lower than that of new equipment of the same type, the system can automatically trigger the preventive maintenance process. Compared with the traditional static scheduling method, this dynamic scheduling mechanism based on the health status of the equipment not only improves the overall efficiency of the production line and optimizes equipment utilization, but also reduces the product defect rate.

[0024] S20: Based on the weight distribution of the processing logs, several impeller processing logs are sorted for process duration, and concentrated value evaluation is performed on the selected process duration to obtain the concentrated duration of five-axis processing, blade polishing processing, heat treatment processing and dynamic balancing debugging, and perform impeller processing production line resource scheduling.

[0025] Specifically, sorting the impeller processing logs based on their weight distribution and performing centralized value evaluation on the selected process durations are the core of obtaining accurate processing duration parameters and guiding resource scheduling.

[0026] The process duration sorting process first implements differentiated screening of the massive impeller machining logs based on the log weight distribution obtained in the previous step. For example, for five-axis machining processes, the system automatically filters out machining logs with weights below a threshold (e.g., 0.3), retaining only high-confidence data with weights between 0.3 and 1. This sorting mechanism effectively eliminates interference data generated by outdated equipment or abnormal operating conditions. For example, if a five-axis machining center (with an initial service life of 36 months) generates 1,000 machining logs within a specific period, only 650 valid data points will be retained after weighted filtering, ensuring that subsequent analysis is based on a high-quality dataset.

[0027] The concentration value assessment uses a weighted median algorithm to further process the sorted process durations. Taking the blade polishing process as an example, the retained processing logs are sorted by processing duration, and the weighted median is calculated based on the weight of each log entry. For example, if the processing logs for a polishing machine (second service life of 24 months) have weights of 0.4, 0.6, and 0.8, corresponding to processing times of 120, 130, and 115 minutes, respectively, the weighted median is calculated as (0.4 × 120 + 0.6 × 130 + 0.8 × 115) / (0.4 + 0.6 + 0.8) = 121.67 minutes, which is the concentrated duration of the blade polishing process. Similarly, the same algorithm can be used to calculate the concentrated duration of five-axis machining, heat treatment, and dynamic balancing.

[0028] During the resource scheduling execution phase, the system inputs four types of concentrated duration parameters into the dynamic scheduling model. When a new batch of impellers enters the production line, the scheduling model intelligently matches the optimal processing path based on the current equipment status matrix (including parameters such as each equipment model, real-time service time, and health index). For example, if it is detected that the concentrated duration of heat treatment processing of a box furnace (the third service time is 18 months) is 8% lower than the production line average, the system will prioritize the allocation of impeller orders with high-precision requirements to this equipment, while adjusting the parallelism of the dynamic balancing and debugging process to increase the overall efficiency of the production line by 18%. This dynamic scheduling mechanism based on concentrated duration can effectively improve equipment utilization, shorten order delivery cycles, and improve production efficiency compared to traditional fixed-beat scheduling.

[0029] In a preferred embodiment, based on the weight distribution of the processing logs, several impeller processing logs are sorted by process duration, and concentrated value evaluation is performed on the selected process durations to obtain the concentrated duration of five-axis processing, blade polishing processing, heat treatment processing, and dynamic balancing debugging, and perform impeller processing production line resource scheduling, which also includes:

[0030] Based on the position numbers of the five-axis machining equipment, blade polishing equipment and heat treatment equipment of the target production line, the defective rates of five-axis machining, blade polishing and heat treatment are calculated.

[0031] According to the five-axis machining defect rate, the blade polishing defect rate and the heat treatment defect rate, the five-axis machining concentrated time, the blade polishing machining concentrated time and the heat treatment machining concentrated time are corrected for invalid time to obtain the five-axis machining corrected time, the blade polishing corrected time and the heat treatment machining corrected time.

[0032] Based on the five-axis machining correction time, the blade polishing correction time, the heat treatment machining correction time and the dynamic balancing debugging concentrated time, impeller machining production line resource scheduling is performed.

[0033] Preferably, the defect rate for each process is calculated based on the equipment location of the target production line, such as the location of the five-axis machining equipment, the blade polishing equipment, and the heat treatment equipment. For example, if a five-axis machining equipment (location: M-01) on a production line processes 1,000 impellers in a month, and 20 of them are scrapped due to dimensional deviations, the five-axis machining defect rate is 2%. Similarly, if the blade polishing equipment (location: P-05) produces 15 surface defects, the defect rate is 1.5%; if the heat treatment equipment (location: H-10) produces 10 parts that do not meet hardness standards, the defect rate is 1%. These defect rate data reflect the potential impact of equipment processing quality stability on delivery cycle time.

[0034] Given the indirect relationship between the defect rate and machining time, direct prediction is difficult. Therefore, the system employs an invalid time correction strategy. Taking five-axis machining as an example, if the original concentrated machining time is 120 minutes and the defect rate is 2%, the system converts the defect rate into a correction factor using a preset algorithm model. Assuming the correction factor calculation formula is: Correction factor = 1 / (1-defect rate)k, where k is an empirical constant (e.g., k = 1.2), the five-axis machining correction factor is 1 / (1-0.02)1.2≈1.024, and the corrected five-axis machining time is 120×1.024≈122.88 minutes. Similarly, the blade polishing process originally had a concentrated duration of 80 minutes and a defect rate of 1.5%. After correction, the duration was approximately 80 × 1 / (1-0.015) / 1.2, which is approximately 81.44 minutes. The heat treatment process originally had a concentrated duration of 150 minutes and a defect rate of 1%. After correction, the duration was approximately 150 × 1 / (1-0.01) / 1.2, which is approximately 151.81 minutes. Because the dynamic balancing process has an extremely low defect rate (typically <0.1%), its concentrated duration can be considered to include quality redundancy and is therefore not corrected.

[0035] Finally, the system executes impeller processing production line resource scheduling based on the corrected five-axis machining correction time (122.88 minutes), blade polishing correction time (81.44 minutes), heat treatment processing correction time (151.81 minutes) and the uncorrected dynamic balancing debugging concentrated time (such as 90 minutes).

[0036] This process quantifies the impact of defective rates on effective output, transforming the traditional scheduling model based on theoretical duration into a dynamic scheduling model that takes actual yield rates into account. This shortens the delivery cycle prediction error and significantly improves customer satisfaction and production line economic benefits.

[0037] In a preferred embodiment, based on the target production line's five-axis machining equipment model and first service time, blade polishing equipment model and second service time, heat treatment equipment model and third service time, dynamic balancing and debugging equipment model and fourth service time, a weight distribution of service time differences is performed on several impeller machining logs to obtain a machining log weight distribution, including:

[0038] A first impeller processing log is extracted from the plurality of impeller processing logs, wherein the first impeller processing log includes recording the model of the five-axis processing equipment and the first recorded service time, recording the model of the blade polishing processing equipment and the second recorded service time, recording the model of the heat treatment processing equipment and the third recorded service time, and recording the model of the dynamic balancing and debugging equipment and the fourth recorded service time.

[0039] According to the record of the five-axis machining equipment model and the first recorded service time, the record of the blade polishing equipment model and the second recorded service time, the record of the heat treatment equipment model and the third recorded service time, the record of the dynamic balancing and debugging equipment model and the fourth recorded service time, and the service time difference weight distribution of the five-axis machining equipment model and the first service time, the blade polishing equipment model and the second service time, the heat treatment equipment model and the third service time, the dynamic balancing and debugging equipment model and the fourth service time, the weight distribution of the first impeller machining log is obtained.

[0040] The first impeller process log weight distribution is added to the process log weight distribution.

[0041] Specifically, a single log is extracted from a massive collection of impeller processing logs for analysis. For example, the first impeller processing log is extracted. The log records in detail the models of the equipment used in each key process and their corresponding recorded service time, including the five-axis processing equipment model and the first recorded service time, the blade polishing processing equipment model and the second recorded service time, the heat treatment processing equipment model and the third recorded service time, and the dynamic balancing and debugging equipment model and the fourth recorded service time.

[0042] Subsequently, the equipment model and service time information recorded in the first impeller processing log are compared one by one with the actual model and current service time (i.e., first service time, second service time, third service time, and fourth service time) of the corresponding equipment in the target production line. Based on a predefined service time difference weight distribution rule, where the greater the service time difference of the equipment, the smaller the weight assigned to the log, the credibility degradation of each log due to equipment aging differences is quantified. Similar calculations are used to obtain the weight distribution of the first impeller processing log across each process, namely, the first impeller processing log weight distribution. Finally, the system integrates the first impeller processing log weight distribution into the overall processing log weight distribution set, providing data support for subsequent resource scheduling based on the dynamic status of the equipment. This ensures that scheduling decisions fully consider the impact of equipment aging on processing time and quality, thereby improving production line resource utilization efficiency and product delivery quality.

[0043] In a preferred embodiment, the weight distribution of the first impeller processing log is obtained by recording the five-axis processing equipment model and the first recorded service time, recording the blade polishing processing equipment model and the second recorded service time, recording the heat treatment processing equipment model and the third recorded service time, recording the dynamic balancing and debugging equipment model and the fourth recorded service time, and performing service time difference weight distribution with the five-axis processing equipment model and the first service time, the blade polishing processing equipment model and the second service time, the heat treatment processing equipment model and the third service time, the dynamic balancing and debugging equipment model and the fourth service time, including:

[0044] When the recorded five-axis machining equipment model is the same as the five-axis machining equipment model.

[0045] Calculate twice the product of the first recorded service time and the first service time, and then calculate the sum of twice the product of the service time and a preset small constant, and set it as a first evaluation parameter.

[0046] Calculate the sum of the squares of the first recorded service time and the first service time, and then calculate the sum of the squares of the service time and the preset small constant, and set it as the second evaluation parameter.

[0047] The ratio of the first evaluation parameter to the second evaluation parameter is calculated, set as the first log five-axis machining weight distribution, and added to the first impeller machining log weight distribution.

[0048] When the recorded five-axis machining equipment model is different from the five-axis machining equipment model, the first log five-axis machining weight distribution is set to 0.

[0049] Furthermore, when constructing the weight distribution for the first impeller machining log, the five-axis machining equipment model and first recorded service time recorded in the log are compared with the five-axis machining equipment model and first recorded service time in the target production line. Specifically, if the recorded five-axis machining equipment model matches the five-axis machining equipment model in the target production line, the system initiates the weight calculation process.

[0050] First, calculate the product of the first recorded service time and the first service time, multiplied by two. This step aims to quantify the relative difference between the two service times. This product is then added to a small, preset constant (e.g., 0.001, to avoid division by zero errors and fine-tune the weight distribution) to obtain the first evaluation parameter. For example, if the first recorded service time is 30 months and the first service time is 36 months, then the product doubled is 1800. After adding the preset small constant 0.001, the first evaluation parameter is approximately 1800.001.

[0051] Next, calculate the sum of the squares of the first recorded service time and the first service time, and then add this sum of squares to a small preset constant to obtain the second evaluation parameter. Continuing with the previous example, the sum of squares is 30² + 36² = 900 + 1296 = 2196. After adding the small preset constant, the second evaluation parameter is approximately 2196.001.

[0052] Next, the ratio of the first evaluation parameter to the second evaluation parameter is calculated as the five-axis machining weight distribution for the first log. This ratio reflects the log's confidence in the five-axis machining process, taking into account differences in equipment model matching and service life. For example, in the previous example, a ratio of approximately 0.8197 indicates that the log's confidence in the five-axis machining process is 0.8197. Finally, the five-axis machining weight distribution for the first log is added to the weight distribution for the first impeller machining log.

[0053] On the contrary, if the recorded five-axis machining equipment model is inconsistent with the five-axis machining equipment model in the target production line, the weight of the log in the five-axis machining process is directly determined to be 0, that is, the five-axis machining weight distribution of the first log is set to 0, so as to exclude invalid data that may be introduced due to equipment model mismatch, and ensure that subsequent resource scheduling decisions are based on accurate and reliable data.

[0054] In a preferred embodiment, based on the weight distribution of the processing logs, several impeller processing logs are sorted by process duration, and concentrated value evaluation is performed on the selected process durations to obtain the concentrated durations of five-axis processing, blade polishing processing, heat treatment processing, and dynamic balancing debugging, and perform impeller processing production line resource scheduling, including:

[0055] From the machining log weight distribution, the first log five-axis machining weight distribution to the Qth log five-axis machining weight distribution are extracted.

[0056] The first log five-axis machining weight distribution to the Qth log five-axis machining weight distribution are clustered according to a weight deviation threshold configured by a user end to obtain a multi-cluster five-axis machining weight distribution.

[0057] The first log five-axis machining duration to the Qth log five-axis machining duration are grouped based on the multi-cluster five-axis machining weight distribution to obtain the multi-cluster five-axis machining duration.

[0058] Step 1: Calculate the variance of the five-axis machining times of multiple centroids of the multi-cluster five-axis machining times.

[0059] Step 2: When the variance value is greater than or equal to the variance threshold, the centroid five-axis machining time corresponding to the cluster with the minimum centroid five-axis machining weight in the multi-cluster five-axis machining weight distribution is deleted from the multiple centroid five-axis machining times, and then the process returns to step 1 to execute the loop.

[0060] Step 3: When the variance value is less than the variance threshold, the multiple centroid five-axis machining times are added to the selected five-axis machining time.

[0061] The selected five-axis machining duration is added to the selected process duration.

[0062] Optionally, based on the constructed processing log weight distribution, all relevant weight data from the first five-axis processing log weight distribution to the Qth five-axis processing log weight distribution is extracted from this distribution. Subsequently, a cluster analysis is performed on these five-axis processing weight distributions using a weight deviation threshold pre-configured by the user. This aims to group logs with similar weights together, thereby obtaining multiple clusters of five-axis processing weight distributions. For example, if the weight deviation threshold is set to 0.1, the system will cluster logs with weight differences of no more than 0.1 into one cluster.

[0063] Next, based on the obtained multi-cluster five-axis machining weight distribution, the corresponding first log five-axis machining time to the Qth log five-axis machining time are grouped to ensure that the machining time in each group corresponds to the same cluster weight distribution, thereby obtaining the multi-cluster five-axis machining time.

[0064] After that, the iterative processing phase begins: Step 1: Calculate the variance of multiple centroid five-axis machining times of multiple clusters of five-axis machining times. This variance is used to measure the degree of dispersion between the machining times of each cluster. Step 2: If the variance value is greater than or equal to the preset variance threshold, it indicates that the machining time difference between the current clusters is large, and the centroid five-axis machining time corresponding to the cluster with the minimum centroid five-axis machining weight needs to be deleted from the multiple centroid five-axis machining times, and then return to Step 1 to recalculate the variance. This step aims to eliminate clusters that may introduce noise data due to too low weights, so as to improve the accuracy of subsequent analysis. Step 3: When the variance value is less than the variance threshold, it indicates that the machining time of the remaining clusters has become stable and the difference is small. At this time, the multiple centroid five-axis machining times are added to the selected five-axis machining time set.

[0065] Ultimately, the selected five-axis machining durations are incorporated into the set of selected process durations, providing data support for subsequent resource scheduling based on process durations. For example, if, after the above processing, the centroids of the remaining three clusters of five-axis machining durations are 120 minutes, 122 minutes, and 121 minutes, respectively, and their variances are less than the threshold, then these three centroid durations will be added to the selected five-axis machining durations and thus participate in resource scheduling decisions for the impeller machining production line.

[0066] In a preferred embodiment, according to the five-axis machining defect rate, the blade polishing defect rate and the heat treatment defect rate, the five-axis machining concentrated time, the blade polishing concentrated time and the heat treatment concentrated time are invalidated, and the five-axis machining corrected time, the blade polishing corrected time and the heat treatment corrected time are obtained, including:

[0067] The correction coefficient encoder bound to the target production line is used to process the five-axis machining defect rate, the blade polishing defect rate and the heat treatment defect rate, as well as the five-axis machining concentration time, the blade polishing concentration time and the heat treatment concentration time to generate the five-axis machining correction coefficient, the blade polishing correction coefficient and the heat treatment correction coefficient.

[0068] The five-axis machining correction time is obtained by multiplying the five-axis machining correction coefficient by the five-axis machining concentrated time.

[0069] The blade polishing correction time is obtained by multiplying the blade polishing correction coefficient by the blade polishing concentrated time.

[0070] The heat treatment correction time is obtained by multiplying the heat treatment correction coefficient by the heat treatment concentrated time.

[0071] For example, in order to accurately consider the impact of processing quality on process duration, the system uses a correction coefficient encoder bound to the target production line to comprehensively process key quality indicators such as five-axis processing defective rate, blade polishing defective rate, and heat treatment defective rate, as well as process duration data such as five-axis processing concentrated time, blade polishing processing concentrated time, and heat treatment processing concentrated time.

[0072] Based on a preset algorithm model, the correction factor encoder converts the nonlinear relationship between defect rate and concentrated duration into quantifiable correction factors, specifically generating correction factors for five-axis machining, blade polishing, and heat treatment. For example, if the five-axis machining defect rate is 3%, and historical data for this process shows that every 1% increase in defect rate leads to a decrease in effective output of approximately 2.5%, the correction factor encoder might generate a five-axis machining correction factor slightly less than 1 (e.g., 0.925) to reflect the negative impact of the defect rate on process duration.

[0073] Subsequently, the five-axis machining correction factor is multiplied by the five-axis machining concentrated time. For example, if the five-axis machining concentrated time is 120 minutes and the correction factor is 0.925, the five-axis machining correction time is 120 × 0.925 = 111 minutes. This correction time is closer to the effective machining time that needs to be adjusted due to the impact of defective rates in actual production. Similarly, the blade polishing correction time is calculated by multiplying the blade polishing correction factor by the blade polishing concentrated time; the heat treatment correction factor is multiplied by the heat treatment concentrated time to obtain the heat treatment correction time.

[0074] Through this series of calculations, the correction time for five-axis machining, blade polishing, and heat treatment machining was obtained, providing a more accurate data basis for subsequent resource scheduling based on the correction time, ensuring that scheduling decisions can fully consider the impact of machining quality on process duration, thereby improving the overall efficiency of the production line and product delivery quality.

[0075] In a preferred embodiment, the correction coefficient encoder includes a discrimination decision layer and a correction coefficient evaluation layer, the correction coefficient evaluation layer includes a five-axis machining correction coefficient calibration branch, a blade polishing machining correction coefficient calibration branch, and a heat treatment machining correction coefficient calibration branch, and processes the five-axis machining defect rate, the blade polishing defect rate, and the heat treatment defect rate, as well as the five-axis machining concentration time, the blade polishing concentration time, and the heat treatment concentration time through a correction coefficient encoder bound to the target production line to generate a five-axis machining correction coefficient, a blade polishing correction coefficient, and a heat treatment machining correction coefficient, including:

[0076] The execution logic of the decision-making layer includes:

[0077] When the five-axis machining defect rate is less than or equal to the five-axis machining defect rate threshold, the five-axis machining correction coefficient is equal to 1; otherwise, the five-axis machining defect rate and the five-axis machining concentrated time are input into the five-axis machining correction coefficient calibration branch to obtain the five-axis machining correction coefficient.

[0078] When the blade polishing defect rate is less than or equal to the blade polishing defect rate threshold, the blade polishing correction coefficient is equal to 1; otherwise, the blade polishing defect rate and the blade polishing concentrated time are input into the blade polishing correction coefficient calibration branch to obtain the blade polishing correction coefficient.

[0079] When the heat treatment defect rate is less than or equal to the heat treatment defect rate threshold, the heat treatment processing correction coefficient is equal to 1; otherwise, the heat treatment defect rate and the heat treatment processing concentrated time are input into the heat treatment processing correction coefficient calibration branch to obtain the heat treatment processing correction coefficient.

[0080] In detail, the correction coefficient encoder, as a key component, is responsible for converting the processing defect rate and the concentrated time of the process into the correction coefficient. Its structure includes a judgment decision layer and a correction coefficient evaluation layer. The correction coefficient evaluation layer is further subdivided into five-axis processing correction coefficient calibration branch, blade polishing processing correction coefficient calibration branch and heat treatment processing correction coefficient calibration branch.

[0081] The logical execution process of the judgment decision layer is as follows: for the five-axis machining process, when the five-axis machining defect rate is less than or equal to the preset five-axis machining defect rate threshold (the threshold is set according to the acceptable defect rate of the quality inspection standard, for example, 2%), it indicates that the current machining quality is within an acceptable range and no additional correction is required to the concentration time, so the five-axis machining correction coefficient is directly set to 1; if the five-axis machining defect rate exceeds this threshold, the system will use the five-axis machining defect rate and the five-axis machining concentration time as inputs, and pass them to the five-axis machining correction coefficient calibration branch for processing. The branch calculates and outputs the five-axis machining correction coefficient based on a preset algorithm model (such as a correction formula derived from historical data regression analysis). For example, when the defect rate is 3% and the concentration time is 120 minutes, the correction coefficient may be 0.95 to reflect the negative impact of the defect rate on the process time.

[0082] Similarly, for the blade polishing process, when the blade polishing defect rate is less than or equal to the blade polishing defect rate threshold (e.g., 1.5%), the blade polishing correction factor is set to 1. If it exceeds the threshold, the defect rate and the concentration duration are input into the blade polishing correction factor calibration branch to calculate the correction factor. The same logic applies to the heat treatment process: when the heat treatment defect rate is less than or equal to the heat treatment defect rate threshold (e.g., 1%), the heat treatment correction factor is set to 1. If it exceeds the threshold, the correction factor is calculated through the heat treatment correction factor calibration branch.

[0083] Through this series of judgment and calculation processes, the five-axis machining correction coefficient, blade polishing correction coefficient and heat treatment machining correction coefficient can be accurately generated, providing reliable data support for subsequent resource scheduling based on correction time, ensuring that scheduling decisions can fully consider the impact of machining quality on process time, thereby improving the overall operation efficiency of the production line and product delivery quality.

[0084] In a preferred embodiment, the five-axis machining correction coefficient calibration branch is constructed by the following steps:

[0085] Collect five-axis machining defect rate record data, five-axis machining time record data and rework time record data.

[0086] The five-axis machining time record data and the rework time record data are added together to obtain a theoretical correction time.

[0087] The ratio of the theoretical correction time to the five-axis machining time record data is calculated and set as a label identifying the five-axis machining correction coefficient.

[0088] Based on the five-axis machining defect rate record data, the five-axis machining time record data and the label identifying the five-axis machining correction coefficient, the five-axis machining correction coefficient calibration branch is obtained through machine learning.

[0089] Specifically, in the process of building the calibration branch for the five-axis machining correction coefficient, data collection is first carried out, specifically covering five-axis machining defect rate records, five-axis machining time records, and rework time records. This data serves as the basis for building the calibration branch and provides a rich source of information for subsequent analysis. For example, from the historical production records of a certain impeller machining production line, multiple sets of five-axis machining defect rate data (such as 2%, 3%, etc.), five-axis machining time records (such as 120 minutes, 125 minutes, etc.), and corresponding rework time records (such as 5 minutes, 8 minutes, etc.) were collected.

[0090] The five-axis machining time and the repair time are then added together to obtain the theoretical correction time. For example, if the five-axis machining time is 120 minutes and the repair time is 5 minutes, the theoretical correction time is 120 + 5 = 125 minutes.

[0091] Next, calculate the ratio of the theoretical correction time to the recorded five-axis machining time and set this ratio as the label identifying the five-axis machining correction factor. Continuing with the previous example, the ratio of a theoretical correction time of 125 minutes to a five-axis machining time of 120 minutes is approximately 1.042, which serves as the five-axis machining correction factor label for the corresponding data set.

[0092] Finally, based on the collected data on five-axis machining defect rate and duration, as well as labels identifying the five-axis machining correction coefficients, machine learning algorithms (such as decision trees and support vector machines) were used to train the model. By continuously adjusting the model parameters, the model accurately learned the inherent correlation between the five-axis machining defect rate, five-axis machining duration, and the correction coefficient, ultimately obtaining a five-axis machining correction coefficient calibration branch. This calibration branch can quickly and accurately output the corresponding five-axis machining correction coefficient based on the input five-axis machining defect rate and five-axis machining duration, providing strong support for correcting process durations in subsequent impeller machining production line resource scheduling.

[0093] The embodiment of the present invention provides a method for dynamically scheduling resources for an impeller processing production line, which has at least the following technical effects:

[0094] 1. By considering the models and service life of the five-axis machining, blade polishing, heat treatment, and dynamic balancing equipment on the target production line, the impeller processing logs are weighted according to their service life differences, thereby obtaining a processing log weight distribution. The log weights can be dynamically adjusted according to the actual usage of the equipment, making the subsequent process duration sorting and resource scheduling more accurate. The impact of factors such as equipment aging and wear on processing quality and efficiency is fully considered, and the scientificity and rationality of resource scheduling are improved.

[0095] 2. Based on the weight distribution of the processing logs, the impeller processing logs are sorted by process duration, and the concentrated value evaluation is used to obtain the concentrated duration of each process. The centroid variance of multiple clusters of five-axis processing durations is further calculated iteratively, and the centroid five-axis processing duration is screened and optimized based on the variance threshold. Finally, the selected five-axis processing duration is obtained. This can effectively eliminate abnormal data points, improve the representativeness and accuracy of the concentrated duration, provide a more reliable data basis for subsequent resource scheduling, and help optimize production line resource allocation and improve production efficiency.

[0096] 3. By calculating the defect rate of each process and using a correction coefficient encoder to correct the invalid duration of the concentrated process duration, the corrected process duration is obtained. The correction coefficient encoder includes a judgment decision layer and a correction coefficient evaluation layer. It can automatically generate a correction coefficient based on the defect rate and the concentrated process duration, dynamically adjusting the process duration. It can fully consider the impact of processing quality on process duration. By correcting the invalid duration, resource scheduling is more closely aligned with actual production conditions, helping to improve product quality and overall production line efficiency. At the same time, the application of the correction coefficient encoder also realizes the automation and intelligence of the correction process, reducing errors caused by human intervention.

[0097] Example 2:

[0098] like Figure 2 As shown, based on the same inventive concept as the method for dynamic scheduling of impeller processing production line resources provided in the first embodiment, an embodiment of the present invention further provides a dynamic scheduling system for impeller processing production line resources, the system comprising:

[0099] The weight distribution module 11 is used to perform service time difference weight distribution on a number of impeller processing logs based on the five-axis processing equipment model and the first service time, the blade polishing processing equipment model and the second service time, the heat treatment processing equipment model and the third service time, and the dynamic balancing and debugging equipment model and the fourth service time of the target production line, to obtain a processing log weight distribution, wherein the service time difference weight distribution rule is that the greater the service time difference, the smaller the processing log weight.

[0100] The resource scheduling module 12 is used to sort the process durations of several impeller processing logs based on the weight distribution of the processing logs, and perform centralized value evaluation on the selected process durations to obtain the centralized time of five-axis processing, the centralized time of blade polishing processing, the centralized time of heat treatment processing and the centralized time of dynamic balancing debugging, and perform resource scheduling of the impeller processing production line.

[0101] Furthermore, the resource scheduling module 12 is further configured to perform the following steps:

[0102] Based on the position number of the five-axis machining equipment, the position number of the blade polishing machining equipment and the position number of the heat treatment machining equipment of the target production line, the five-axis machining defect rate, the blade polishing defect rate and the heat treatment defect rate are counted; according to the five-axis machining defect rate, the blade polishing defect rate and the heat treatment defect rate, the five-axis machining concentrated time, the blade polishing concentrated time and the heat treatment concentrated time are invalidated to obtain the five-axis machining corrected time, the blade polishing corrected time and the heat treatment corrected time; based on the five-axis machining corrected time, the blade polishing corrected time, the heat treatment corrected time and the dynamic balancing debugging concentrated time, the impeller machining production line resource scheduling is performed.

[0103] Furthermore, the weight distribution module 11 is further configured to perform the following steps:

[0104] From the several impeller processing logs, a first impeller processing log is extracted, wherein the first impeller processing log includes recording the five-axis processing equipment model and the first recorded service time, recording the blade polishing processing equipment model and the second recorded service time, recording the heat treatment processing equipment model and the third recorded service time, recording the dynamic balancing and debugging equipment model and the fourth recorded service time; according to the record of the five-axis processing equipment model and the first recorded service time, recording the blade polishing processing equipment model and the second recorded service time, recording the heat treatment processing equipment model and the third recorded service time, recording the dynamic balancing and debugging equipment model and the fourth recorded service time, and the five-axis processing equipment model and the first service time, the blade polishing processing equipment model and the second service time, the heat treatment processing equipment model and the third service time, the dynamic balancing and debugging equipment model and the fourth service time, the service time difference weight distribution is performed to obtain the first impeller processing log weight distribution; the first impeller processing log weight distribution is added to the processing log weight distribution.

[0105] Furthermore, the weight distribution module 11 is further configured to perform the following steps:

[0106] When the recorded five-axis machining equipment model is the same as the five-axis machining equipment model: calculate twice the product of the first recorded service time and the first service time, and then calculate the sum of twice the service time product and a preset small constant, and set it as a first evaluation parameter; calculate the sum of the squares of the first recorded service time and the first service time, and then calculate the sum of the squares of the service time and the preset small constant, and set it as a second evaluation parameter; calculate the ratio of the first evaluation parameter to the second evaluation parameter, set it as the first log five-axis machining weight distribution, and add it to the first impeller machining log weight distribution; when the recorded five-axis machining equipment model is different from the five-axis machining equipment model, set the first log five-axis machining weight distribution to 0.

[0107] Furthermore, the resource scheduling module 12 is further configured to perform the following steps:

[0108] From the processing log weight distribution, extract the first log five-axis processing weight distribution to the Qth log five-axis processing weight distribution; cluster the first log five-axis processing weight distribution to the Qth log five-axis processing weight distribution through the weight deviation threshold configured on the user side to obtain a multi-cluster five-axis processing weight distribution; group the first log five-axis processing time to the Qth log five-axis processing time based on the multi-cluster five-axis processing weight distribution to obtain a multi-cluster five-axis processing time; step one: calculate the variance of the multiple centroid five-axis processing times of the multi-cluster five-axis processing time; step two: when the variance value is greater than or equal to the variance threshold, delete the centroid five-axis processing time corresponding to the cluster with the minimum centroid five-axis processing weight in the multi-cluster five-axis processing weight distribution from the multiple centroid five-axis processing times, and return to step one to execute the loop; step three: when the variance value is less than the variance threshold, add the multiple centroid five-axis processing times to the selected five-axis processing time; add the selected five-axis processing time to the selected process time.

[0109] Furthermore, the resource scheduling module 12 is further configured to perform the following steps:

[0110] The five-axis machining defect rate, the blade polishing defect rate and the heat treatment defect rate, as well as the five-axis machining concentration time, the blade polishing machining concentration time and the heat treatment machining concentration time are processed through the correction coefficient encoder bound to the target production line to generate a five-axis machining correction coefficient, a blade polishing correction coefficient and a heat treatment machining correction coefficient; the five-axis machining correction coefficient is multiplied by the five-axis machining concentration time to obtain the five-axis machining correction time; the blade polishing correction coefficient is multiplied by the blade polishing concentration time to obtain the blade polishing correction time; the heat treatment machining correction coefficient is multiplied by the heat treatment machining concentration time to obtain the heat treatment machining correction time.

[0111] Furthermore, the resource scheduling module 12 is further configured to perform the following steps:

[0112] The execution logic of the judgment decision layer includes: when the five-axis machining defect rate is less than or equal to the five-axis machining defect rate threshold, the five-axis machining correction coefficient is equal to 1, otherwise, the five-axis machining defect rate and the five-axis machining concentrated time are input into the five-axis machining correction coefficient calibration branch to obtain the five-axis machining correction coefficient; when the blade polishing machining defect rate is less than or equal to the blade polishing machining defect rate threshold, the blade polishing machining correction coefficient is equal to 1, otherwise, the blade polishing machining defect rate and the blade polishing machining concentrated time are input into the blade polishing machining correction coefficient calibration branch to obtain the blade polishing correction coefficient; when the heat treatment defect rate is less than or equal to the heat treatment defect rate threshold, the heat treatment machining correction coefficient is equal to 1, otherwise, the heat treatment defect rate and the heat treatment machining concentrated time are input into the heat treatment machining correction coefficient calibration branch to obtain the heat treatment machining correction coefficient.

[0113] Furthermore, the resource scheduling module 12 is further configured to perform the following steps:

[0114] Collect five-axis machining defect rate record data, five-axis machining time record data and rework time record data; add the five-axis machining time record data and the rework time record data to obtain a theoretical correction time; calculate the ratio of the theoretical correction time to the five-axis machining time record data, and set it as a label identifying the five-axis machining correction coefficient; based on the five-axis machining defect rate record data, the five-axis machining time record data and the label identifying the five-axis machining correction coefficient, obtain the five-axis machining correction coefficient calibration branch through machine learning.

[0115] Through the above-mentioned detailed description of a method for dynamic scheduling of resources for an impeller processing production line in this specification, those skilled in the art can clearly understand a dynamic scheduling system for resources for an impeller processing production line in this embodiment. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0116] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A dynamic resource scheduling method for an impeller processing production line, characterized in that: include: Based on the target production line's five-axis machining equipment model and first service time, blade polishing equipment model and second service time, heat treatment equipment model and third service time, and dynamic balancing and debugging equipment model and fourth service time, a service time difference weight distribution is performed on several impeller machining logs to obtain a machining log weight distribution. The service time difference weight distribution rule is that the larger the service time difference, the smaller the machining log weight. Based on the weight distribution of the processing logs, several impeller processing logs are sorted by process duration, and concentrated value evaluation is performed on the selected process durations to obtain the concentrated durations of five-axis processing, blade polishing processing, heat treatment processing, and dynamic balancing debugging, and perform resource scheduling for the impeller processing production line; Based on the weight distribution of the processing logs, several impeller processing logs are sorted by process duration, and concentrated value evaluation is performed on the selected process durations to obtain the concentrated durations of five-axis processing, blade polishing processing, heat treatment processing, and dynamic balancing debugging. Resource scheduling of the impeller processing production line is then performed, including: Extracting the first log five-axis machining weight distribution to the Qth log five-axis machining weight distribution from the machining log weight distribution; Clustering the first log five-axis machining weight distribution to the Qth log five-axis machining weight distribution according to a weight deviation threshold configured by a user end to obtain a multi-cluster five-axis machining weight distribution; Grouping the first log five-axis machining duration to the Qth log five-axis machining duration based on the multi-cluster five-axis machining weight distribution to obtain the multi-cluster five-axis machining duration; Step 1: Calculate the variance of the five-axis machining time of multiple centroids of the multi-cluster five-axis machining time; Step 2: When the variance value is greater than or equal to the variance threshold, the centroid five-axis machining time corresponding to the cluster with the minimum centroid five-axis machining weight in the multi-cluster five-axis machining weight distribution is deleted from the multiple centroid five-axis machining time, and then the process returns to step 1 to execute the loop; Step 3: When the variance value is less than the variance threshold, the plurality of centroid five-axis machining times are added to the selected five-axis machining time; The selected five-axis machining duration is added to the selected process duration.

2. The method for dynamic resource scheduling of an impeller processing production line according to claim 1, characterized in that: Based on the weight distribution of the processing logs, several impeller processing logs are sorted by process duration, and concentrated value evaluation is performed on the selected process durations to obtain the concentrated durations of five-axis processing, blade polishing processing, heat treatment processing, and dynamic balancing debugging. Resource scheduling of the impeller processing production line is then performed, which also includes: Based on the position numbers of the five-axis machining equipment, blade polishing equipment, and heat treatment equipment of the target production line, the defective rates of five-axis machining, blade polishing, and heat treatment are calculated; According to the five-axis machining defect rate, the blade polishing defect rate and the heat treatment defect rate, invalid time correction is performed on the five-axis machining concentrated time, the blade polishing concentrated time and the heat treatment concentrated time to obtain the five-axis machining corrected time, the blade polishing corrected time and the heat treatment corrected time; Based on the five-axis machining correction time, the blade polishing correction time, the heat treatment machining correction time and the dynamic balancing debugging concentrated time, impeller machining production line resource scheduling is performed.

3. The method for dynamic resource scheduling of an impeller processing production line according to claim 1, wherein: Based on the target production line's five-axis machining equipment model and first service time, blade polishing equipment model and second service time, heat treatment equipment model and third service time, dynamic balancing and debugging equipment model and fourth service time, a weight distribution of service time differences is performed on several impeller machining logs to obtain the machining log weight distribution, including: Extracting a first impeller processing log from the plurality of impeller processing logs, wherein the first impeller processing log includes recording a five-axis processing equipment model and a first record of service time, recording a blade polishing processing equipment model and a second record of service time, recording a heat treatment processing equipment model and a third record of service time, and recording a dynamic balancing debugging equipment model and a fourth record of service time; According to the record of the five-axis processing equipment model and the first recorded service time, the record of the blade polishing processing equipment model and the second recorded service time, the record of the heat treatment processing equipment model and the third recorded service time, the record of the dynamic balancing and debugging equipment model and the fourth recorded service time, and the five-axis processing equipment model and the first service time, the blade polishing processing equipment model and the second service time, the heat treatment processing equipment model and the third service time, the dynamic balancing and debugging equipment model and the fourth service time, the service time difference weight distribution is performed to obtain the weight distribution of the first impeller processing log; The first impeller process log weight distribution is added to the process log weight distribution.

4. The method for dynamic resource scheduling of an impeller processing production line according to claim 3, wherein: According to the record of the five-axis processing equipment model and the first recorded service time, the record of the blade polishing processing equipment model and the second recorded service time, the record of the heat treatment processing equipment model and the third recorded service time, the record of the dynamic balancing and debugging equipment model and the fourth recorded service time, and the service time difference weight distribution of the five-axis processing equipment model and the first service time, the blade polishing processing equipment model and the second service time, the heat treatment processing equipment model and the third service time, the dynamic balancing and debugging equipment model and the fourth service time, the weight distribution of the first impeller processing log is obtained, including: When the recorded five-axis machining equipment model is the same as the five-axis machining equipment model: Calculate twice the product of the first recorded service time and the first service time, and then calculate the sum of twice the product of the service time and a preset small constant, and set the sum as a first evaluation parameter; Calculating the sum of the squares of the first recorded service time and the first service time, and then calculating the sum of the squares of the service time and the preset small constant, and setting the sum as a second evaluation parameter; Calculating a ratio of the first evaluation parameter to the second evaluation parameter, setting the ratio as a first log five-axis machining weight distribution, and adding the ratio to the first impeller machining log weight distribution; When the recorded five-axis machining equipment model is different from the five-axis machining equipment model, the first log five-axis machining weight distribution is set to 0.

5. The method for dynamic resource scheduling of an impeller processing production line according to claim 2, wherein: According to the five-axis machining defect rate, the blade polishing defect rate, and the heat treatment defect rate, invalid time correction is performed on the five-axis machining concentrated time, the blade polishing concentrated time, and the heat treatment concentrated time to obtain the five-axis machining corrected time, the blade polishing corrected time, and the heat treatment corrected time, including: The correction coefficient encoder bound to the target production line is used to process the five-axis machining defect rate, the blade polishing defect rate, and the heat treatment defect rate, as well as the five-axis machining concentrated time, the blade polishing concentrated time, and the heat treatment concentrated time, to generate a five-axis machining correction coefficient, a blade polishing correction coefficient, and a heat treatment correction coefficient; The five-axis machining correction time is obtained by multiplying the five-axis machining correction coefficient by the five-axis machining concentrated time; The blade polishing correction time is obtained by multiplying the blade polishing correction coefficient by the blade polishing concentrated time; The heat treatment correction time is obtained by multiplying the heat treatment correction coefficient by the heat treatment concentrated time.

6. The method for dynamic resource scheduling of an impeller processing production line according to claim 5, characterized in that: The correction coefficient encoder includes a discrimination decision layer and a correction coefficient evaluation layer. The correction coefficient evaluation layer includes a five-axis machining correction coefficient calibration branch, a blade polishing machining correction coefficient calibration branch, and a heat treatment machining correction coefficient calibration branch. The correction coefficient encoder bound to the target production line processes the five-axis machining defect rate, the blade polishing defect rate, and the heat treatment defect rate, as well as the five-axis machining concentrated time, the blade polishing concentrated time, and the heat treatment concentrated time to generate the five-axis machining correction coefficient, the blade polishing correction coefficient, and the heat treatment correction coefficient, including: The execution logic of the decision-making layer includes: When the five-axis machining defect rate is less than or equal to the five-axis machining defect rate threshold, the five-axis machining correction coefficient is equal to 1; otherwise, the five-axis machining defect rate and the five-axis machining concentrated time are input into the five-axis machining correction coefficient calibration branch to obtain the five-axis machining correction coefficient; When the blade polishing defect rate is less than or equal to the blade polishing defect rate threshold, the blade polishing correction coefficient is equal to 1; otherwise, the blade polishing defect rate and the blade polishing processing concentrated time are input into the blade polishing correction coefficient calibration branch to obtain the blade polishing correction coefficient; When the heat treatment defect rate is less than or equal to the heat treatment defect rate threshold, the heat treatment processing correction coefficient is equal to 1; otherwise, the heat treatment defect rate and the heat treatment processing concentrated time are input into the heat treatment processing correction coefficient calibration branch to obtain the heat treatment processing correction coefficient.

7. The method for dynamic resource scheduling of an impeller processing production line according to claim 6, wherein: The five-axis machining correction coefficient calibration branch is constructed through the following steps: Collect five-axis machining defect rate record data, five-axis machining time record data and rework time record data; Adding the five-axis machining time record data and the rework time record data to obtain a theoretical correction time; Calculating a ratio of the theoretical correction time to the five-axis machining time record data, and setting the ratio as a label identifying the five-axis machining correction coefficient; Based on the five-axis machining defect rate record data, the five-axis machining time record data and the label identifying the five-axis machining correction coefficient, the five-axis machining correction coefficient calibration branch is obtained through machine learning.

8. A dynamic resource scheduling system for an impeller processing production line, characterized in that: The system for implementing the method for dynamic resource scheduling of an impeller processing production line according to any one of claims 1 to 7 comprises: The weight distribution module is used to perform service time difference weight distribution on several impeller processing logs based on the target production line's five-axis processing equipment model and first service time, blade polishing processing equipment model and second service time, heat treatment processing equipment model and third service time, and dynamic balancing and debugging equipment model and fourth service time, to obtain a processing log weight distribution. The service time difference weight distribution rule is that the larger the service time difference, the smaller the processing log weight. The resource scheduling module is used to sort the process durations of several impeller processing logs based on the weight distribution of the processing logs, and perform centralized value evaluation on the selected process durations to obtain the centralized time of five-axis processing, the centralized time of blade polishing processing, the centralized time of heat treatment processing, and the centralized time of dynamic balancing debugging, and perform resource scheduling of the impeller processing production line.

Images