Manufacturing system health analysis method and device

Abstract

The present disclosure relates to the field of electronic technology, and in particular, provides a manufacturing system health analysis method and device, the method comprising: determining at least one process parameter set related to the manufacturing system, the process parameter set including a process parameter combination corresponding to at least one machine of the manufacturing system; predicting the processing results generated when the manufacturing system works according to the configuration of each process parameter set, the processing results including the quality characteristic values ​​generated when each machine works according to the respective process parameter combination in the process parameter set; selecting a target process parameter combination corresponding to at least one machine according to the processing results of each process parameter set; if the estimated processing results of the target process parameter combination corresponding to at least one machine meet the preset target processing results, then controlling the manufacturing system to work according to the target process parameter combination of each machine. The present disclosure obtains a target process parameter combination with extremely high operating stability, which improves the operating efficiency and stability of the manufacturing system.

Description

Technical Field

[0001] The present disclosure relates to the field of electronic technology, and in particular to a manufacturing system health analysis method and device. Background Art

[0002] With the rapid development of science and technology, artificial intelligence is increasingly being used in the manufacturing industry. For a more common manufacturing system, it is generally composed of multiple machines and can be used to manufacture industrial products. In order to ensure that the manufacturing system can produce qualified products, it is necessary to ensure that the system operates stably.

[0003] In actual production, fluctuations in influencing factors such as operators, machines, materials, processing methods, measurements and environment can lead to unstable product quality. Therefore, before using the manufacturing system, it is necessary to know the stability of the manufacturing system during operation. Therefore, how to predict the operation of the manufacturing system to improve the operating efficiency and qualification rate of the manufacturing system is a technical problem that needs to be solved urgently. Summary of the invention

[0004] In view of the above problems, the technical solution of the present disclosure is proposed: a manufacturing system health analysis method and device.

[0005] In a first aspect, the present disclosure provides a manufacturing system health analysis method, comprising:

[0006] Determine at least one process parameter set related to the manufacturing system, the process parameter set includes a process parameter combination corresponding to at least one machine of the manufacturing system, the process parameter combination includes at least one process parameter, and the process parameter refers to an input parameter when the machine is working;

[0007] Predict the processing results produced when the manufacturing system works according to the configuration of each process parameter set. The processing results include the quality characteristic values ​​produced when each machine works according to its own process parameter combination in the process parameter set. The quality characteristic values ​​are used to represent the values ​​of key quality characteristics after the machine processes the workpiece;

[0008] According to the processing results of each process parameter set, select at least one target process parameter combination corresponding to each machine;

[0009] If the estimated processing result of the target process parameter combination corresponding to at least one machine meets the preset target processing result, the manufacturing system is controlled to work according to the target process parameter combination of each machine.

[0010] According to a second aspect of the present disclosure, there is provided a manufacturing system health analysis device, comprising:

[0011] A determination unit, configured to determine at least one process parameter set related to the manufacturing system, wherein the process parameter set includes a process parameter combination corresponding to at least one machine of the manufacturing system, wherein the process parameter combination includes at least one process parameter, and the process parameter refers to an input parameter when the machine is working;

[0012] A prediction unit is used to predict the processing results generated when the manufacturing system works according to the configuration of each process parameter set. The processing results include the quality characteristic values ​​generated when each machine works according to the respective process parameter combination in the process parameter set. The quality characteristic values ​​are used to represent the values ​​of key quality characteristics after the machine processes the workpiece;

[0013] A selection unit, for selecting a target process parameter combination corresponding to at least one machine according to the processing results of each process parameter set;

[0014] The control unit is used to control the manufacturing system to work according to the target process parameter combination of each machine if the estimated processing result of the target process parameter combination corresponding to at least one machine meets the preset target processing result.

[0015] As will be described in detail below, according to the manufacturing system health analysis method of the embodiment of the present disclosure, at least one process parameter set related to the manufacturing system can be determined, and the process parameter set can include the process parameter combination of each machine in the manufacturing system. In addition, the process parameter set of the manufacturing system is used to simulate the production of the manufacturing system to obtain the processing results corresponding to each process parameter set. The processing results may include the quality characteristic value of each machine in the corresponding process parameter combination, and the quality characteristic value can directly characterize the processing effect of each process parameter set, obtain the target process parameter combination with better processing effect of each machine, and realize the preliminary screening of the target process parameter combination of each machine. Afterwards, when it is confirmed that the target process parameter combination of each machine meets the preset functional health prediction condition, the target process parameter combination is used for actual processing. Through the steps of the composition of the parameter combination, the production simulation of the manufacturing system, the screening of the target process parameter combination of each machine, the determination of the functional health prediction condition, etc., the dimension of predicting the manufacturing health status of the manufacturing system is established, and a process parameter combination with a healthier operation is selected for the manufacturing system, and a more stable and more accurate process parameter combination is provided for the operation health of the manufacturing system, thereby ensuring the healthy operation of the manufacturing system.

[0016] It is to be understood that both the foregoing general description and the following detailed description are exemplary, and are intended to provide further explanation of the technology as claimed. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and other purposes, features and advantages of the present disclosure will become more apparent by describing the embodiments of the present disclosure in more detail in conjunction with the accompanying drawings. The accompanying drawings are used to provide a further understanding of the embodiments of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and do not constitute a limitation of the present disclosure. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0018] Figure 1 is a structural schematic diagram illustrating a manufacturing system according to an embodiment of the present disclosure;

[0019] Figure 2 is a flow chart illustrating a manufacturing system health analysis method according to an embodiment of the present disclosure;

[0020] Figure 3 is a diagram further illustrating an example of application of the HHO-BP neural network model in the manufacturing system health analysis method according to an embodiment of the present disclosure;

[0021] Figure 4 is another flow chart illustrating a manufacturing system health analysis method according to an embodiment of the present disclosure;

[0022] Figure 5 is another flow chart illustrating a manufacturing system health analysis method according to an embodiment of the present disclosure;

[0023] Figure 6 is another flow chart illustrating a manufacturing system health analysis method according to an embodiment of the present disclosure;

[0024] Figure 7 is a block diagram illustrating a manufacturing system health analysis apparatus according to an embodiment of the present disclosure;

[0025] Figure 8 is a hardware block diagram illustrating an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the present disclosure more obvious, the exemplary embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments of the present disclosure, and it should be understood that the present disclosure is not limited to the exemplary embodiments described here.

[0027] The technical solution disclosed in the present invention can be applied in intelligent manufacturing scenarios, by configuring a process parameter set for one or more machines in a manufacturing system. The process parameter set may include a combination of process parameters corresponding to one or more machines, and performing a stability test on the manufacturing system through the process parameter set to select a process parameter set with higher stability, so as to make the operation stability of the manufacturing system higher.

[0028] In the related art, a manufacturing system can generally be composed of one or more machines, and the machines in the manufacturing system can run in parallel or in series. In order to improve the operational stability of the manufacturing system, it is necessary to first perform a stability analysis on the manufacturing system. At present, prediction methods such as models, data or knowledge are generally used to predict the operational stability of the manufacturing system. For example, interference factors of the manufacturing system are used, and the interference factors may include internal interference factors and external interference factors. Internal interference usually includes processing machine failures, defects in work-in-progress, and changes in production tasks. External interference includes changes in environmental factors, changes in customer demand and other factors. The operational stability of the manufacturing system is predicted by interference factors. However, the prediction results currently obtained by using the interference factor analysis method are not accurate enough, resulting in the manufacturing system not being stable enough after being predicted in this way.

[0029] In order to solve the above problems, in the technical solution disclosed in the present invention, the process parameter combinations of each machine in the manufacturing system are combined to predict the processing quality of each process parameter set, and then the target process parameter combination is selected for each machine. Then, the manufacturing system is simulated using the target process parameter combination to obtain the system estimated processing result of the manufacturing system, and it is determined whether the system estimated processing result meets the preset target processing result. If it meets the target process parameter combination, the manufacturing system is controlled to work according to the target process parameter combination.

[0030] Figure 1 This is an example diagram of a manufacturing system provided by an embodiment of the present disclosure. Figure 1 As shown, the manufacturing system may include a machine 10, a machine 20, a machine 30 and a machine 40, wherein the machine 10 and the machine 30 are in parallel, the machine 10 and the machine 20 are in series, the machine 30 and the machine 40 are in series, and in addition, the machine 10 may also be in series with the machine 40.

[0031] For example, the raw material can be fed into the machine 10 and the machine 30, and after the processing is completed by the machine 10, the workpiece is fed into the machine 20 or the machine 40. After the processing is completed by the machine 30, the workpiece is fed into the machine 40. Figure 1 The manufacturing system shown is merely exemplary and does not limit the parallel or serial connection method of the machines in the manufacturing system or the number of machines.

[0032] The manufacturing system may further include an electronic device 50, in which the manufacturing system health analysis method provided by the present disclosure may be configured. The electronic device 50 may serve as a host computer of the machine 10-40 to control the machine 10-40 through parameters or instructions, thereby achieving smooth and healthy operation of the manufacturing system.

[0033] The manufacturing system health analysis method provided in the embodiment of the present disclosure requires the following assumptions before use:

[0034] Assumption 1: The degradation of all machines in the manufacturing system follows a Markov process.

[0035] Assumption 2: The performance status of each processing machine refers to the amount of work that the processing machine can process under normal circumstances.

[0036] Assumption 3: The critical quality characteristics / critical quality characteristics processed by each machine are independent. The critical quality characteristics / critical quality characteristics can refer to the geometric quantities of the workpiece processed by each machine, such as diameter, length, weld length or roughness.

[0037] Assumption 4: There is an absolutely reliable checkpoint after each processing is completed to check the quality status of the output product.

[0038] Assumption 5: The processing capability of a machine has several discrete states, and the corresponding probabilities are proportional.

[0039] Hypothesis 6: The stability of process parameters will affect the probability of the processing machine being in various performance states.

[0040] Assumption 7: After the process parameters are selected, the impact of human labor on the processing is not considered during the manufacturing process of the manufacturing system.

[0041] Based on the above assumptions, a manufacturing system health analysis method according to an embodiment of the present disclosure is proposed. The manufacturing system health analysis method will be described in detail below in conjunction with the accompanying drawings.

[0042] like Figure 2 FIG. 1 is a schematic flow chart of a manufacturing system health analysis method provided by an embodiment of the present disclosure. The method may include the following steps:

[0043] S201. Determine at least one process parameter set related to a manufacturing system, where the process parameter set includes a process parameter combination corresponding to at least one machine in the manufacturing system, and the process parameter combination includes at least one process parameter, which refers to an input parameter when the machine is working.

[0044] Optionally, S201 further includes: determining at least one process parameter set related to the manufacturing system in response to a trigger operation performed by a user. The trigger operation may refer to a click operation performed by the user on a device control control. The electronic device may display a system operation page, which may include a device control control.

[0045] Optionally, the manufacturing system may include at least one machine, and the at least one machine may be connected in series and / or in parallel.

[0046] Each machine may have a corresponding machine type, such as a centerless grinder, a surface grinder, a welder, or a grinder.

[0047] Each machine may also have a process corresponding to the machine type, key quality characteristics set during machining, workpiece specifications, and process parameter combinations.

[0048] Table 1 below shows examples of several machine types, processes, key quality characteristics, specifications, and process parameter combinations.

[0049]

[0050] Table 1

[0051] As shown in Table 1, the key quality characteristic of the centerless grinder is the diameter, the key quality characteristic of the surface grinder is the length, the key quality characteristic of the electric welder is the weld length, and the key quality characteristic of the grinder is the roughness.

[0052] Of course, the above Table 1 is only an example of machine types and various machine parameters, and does not constitute a detailed limitation. In actual applications, the number of machines and machine types in the manufacturing system can be set according to processing needs, and this embodiment does not impose too many limitations on this.

[0053] Optionally, the process parameter set refers to a parameter set formed by a combination of process parameters respectively set for at least one machine in the manufacturing system. The process parameter combination may include at least one process parameter, and the process parameter refers to an input parameter when the machine is working, and specifically may refer to a parameter set by the machine to complete the processing target of the corresponding process. The process parameter combination of each machine may include one or more process parameters.

[0054] For example, the process parameter combination of a centerless grinder may include parameters such as grinding wheel speed, workpiece speed, and feed rate. The process parameter combination of a surface grinder may include parameters such as grinding wheel speed, workpiece speed, and feed rate. The process parameter combination of an electric welder may include parameters such as welding speed, arc voltage, and welding current. The process parameter combination of a grinder may include parameters such as grinding wheel speed, workpiece speed, and backcut depth.

[0055] That is, a process parameter set can be, for example, {[machine 1 (centerless grinder): grinding wheel speed, workpiece speed, feed rate], [machine 2 (surface grinder): grinding wheel speed, workpiece speed, feed rate], [machine 3 (electric welder): welding speed, arc voltage, welding current], [machine 4 (grinder): grinding wheel speed, workpiece speed, backcut depth]}

[0056] In order to determine one or more process parameter sets related to the manufacturing system, at least one machine in the manufacturing system may be determined first, that is, the machine composition of the manufacturing system may be clarified. After obtaining at least one machine in the manufacturing system, a process parameter combination of each machine is determined. That is, the process parameter combination of each machine is determined based on the parameters that can affect the processing of each machine. The process parameter combinations of each machine are combined into a set of process parameter sets.

[0057] S202. Predict the processing results generated when the manufacturing system works according to the configuration of each process parameter set. The processing results include the quality characteristic values ​​generated when each machine works according to its own process parameter combination in the process parameter set. The quality characteristic values ​​represent the values ​​of key quality characteristics after the machine processes the workpiece.

[0058] Optionally, a key quality characteristic refers to a characteristic used to indicate the processing quality of a workpiece. After processing the workpiece using a combination of process parameters and according to the processing target or processing result of the workpiece, the value of the key quality characteristic of the workpiece is obtained, which is the quality characteristic value. The quality characteristic value is used to indicate the value of the key quality characteristic of the workpiece after the machine processes the workpiece.

[0059] For example, a centerless grinder can grind a workpiece into a cylindrical or spherical shape. The key quality characteristic corresponding to the processing function of the centerless grinder is the diameter, that is, the quality characteristic value is the value of the diameter of the workpiece after grinding. The centerless grinder can use a combination of process parameters (such as grinding wheel speed, workpiece speed, and feed rate) to process the workpiece, and the processing target is a workpiece diameter of 15 cm. After the workpiece is processed by the centerless grinder, the diameter of the workpiece is 15.1 cm. The 15.1 cm is the quality characteristic value.

[0060] Optionally, when the process parameter set is known, the processing result generated by the manufacturing system when the process parameter set is operated can be predicted. The processing result can include the quality characteristic value of each machine.

[0061] Furthermore, a production model of the manufacturing system may be obtained, and each process parameter set may be input into the production model of the manufacturing system to obtain a processing result corresponding to each process parameter set.

[0062] The production model may refer to a machine learning model that simulates the production process of at least one machine in a manufacturing system. The production model may be obtained through training.

[0063] For example, a production model with unknown parameters is established for a manufacturing system, and the historical process parameter set of the manufacturing system's historical processing and the actual processing results corresponding to the historical process parameter set are obtained. The historical process parameter set and the corresponding processing results are used as training data to train the production model until the model converges. The production model parameters corresponding to the target parameters finally obtained are then applied to the processing simulation process of the manufacturing system.

[0064] Optionally, the manufacturing system works according to the configuration of each process parameter set, which may specifically refer to configuring at least one machine in the manufacturing system according to respective process parameter combinations, and performing production simulation through the manufacturing system when the configuration is completed.

[0065] Optionally, each machine operating according to its own process parameter combination in the process parameter set may mean that a processing simulation is performed according to the process parameter combination of each machine to obtain a quality characteristic value of each machine.

[0066] Specifically, third-party simulation software can be used to simulate the working environment of the manufacturing system. The processing actions of each machine in the manufacturing system can be simulated through the third-party simulation software, and after inputting the process parameter set, the processing result of each process parameter set can be obtained.

[0067] S203: Select at least one target process parameter combination corresponding to each machine according to the processing results of each process parameter set.

[0068] Optionally, S203 may include: selecting, for each machine, a target process parameter combination with the highest processing effect according to the processing results of each process parameter set.

[0069] The processing result may include the quality characteristic value generated when each machine works according to the process parameter combination. Furthermore, for any machine, the quality characteristic value generated by the process parameter combination corresponding to at least one process parameter set of the machine may be used. Based on the quality characteristic value generated by the machine in at least one process parameter combination, the processing effect of the machine corresponding to at least one process parameter combination may be determined, and the process parameter combination with the highest processing effect may be selected as the target process parameter combination.

[0070] The processing effect of the machine can be represented by the estimated processing result of the manufacturing system. The higher the estimated processing result of the machine in the manufacturing system, the higher the processing effect of the machine. Conversely, the lower the system estimated processing result of the machine in the manufacturing system, the lower the processing effect of the machine.

[0071] Selecting machines based on processing effects can ensure that the selected target process parameter combination can achieve a higher processing effect, thereby achieving effective screening of process parameter combinations.

[0072] S204: If the estimated processing result of the target process parameter combination corresponding to at least one machine satisfies the preset target processing result, the manufacturing system is controlled to operate according to the target process parameter combination of each machine.

[0073] Optionally, S204 may include: determining, based on the target process parameter combinations corresponding to at least one machine, an estimated processing result when the manufacturing system operates according to the target process parameter combinations of each machine. If the estimated processing result reaches the preset target processing result, controlling the manufacturing system to operate according to the target process parameter combinations of each machine.

[0074] Optionally, the estimated processing result may refer to a value used to indicate the healthy operation status of the system when the manufacturing system operates according to the target process parameter combination corresponding to at least one machine. The higher the value of the estimated processing result, the healthier the manufacturing system operates, and the more stable and accurate the manufactured products are. The lower the value of the estimated processing result, the more abnormal the manufacturing system operates, and the worse the accuracy of the manufactured products is.

[0075] In the disclosed embodiment, at least one process parameter set related to the manufacturing system can be determined, and the process parameter set can include a process parameter combination of each machine in the manufacturing system. In addition, the process parameter set of the manufacturing system is used to perform production simulation on the manufacturing system to obtain processing results corresponding to each process parameter set. The processing results can include the quality characteristic value of each machine in the corresponding process parameter combination, and the quality characteristic value can directly characterize the processing effect of each process parameter set, obtain the target process parameter combination with better processing effect of each machine, and realize the preliminary screening of the target process parameter combination of each machine. Afterwards, when it is confirmed that the estimated processing result of the target process parameter combination corresponding to at least one machine of the target process parameter combination of each machine meets the preset target processing result, the target process parameter combination is used for actual processing. Through the steps of the composition of the parameter combination, the production simulation of the manufacturing system, the screening of the target process parameter combination of each machine, the determination of functional health, etc., the dimension of predicting the manufacturing health status of the manufacturing system is established, and a process parameter combination with healthier operation is selected for the manufacturing system, and a process parameter combination with more stability and higher processing accuracy is provided for the operation health of the manufacturing system, thereby ensuring the healthy operation of the manufacturing system.

[0076] It is understandable that different combinations of process parameters are used as inputs to the corresponding machines in the manufacturing system, and the values ​​of the key quality characteristics of the workpieces in process output by each machine should meet the preset requirements, that is, the processing results are the values ​​of the key quality characteristics of the workpieces taken by the machine. In screening the target process parameter combinations of each machine, on the one hand, the output of the process parameter combination of each machine needs to meet the quality characteristic requirements, and on the other hand, the signal-to-noise ratio must be large enough, that is, stable.

[0077] As mentioned above, the production model can be used to simulate the production process of the manufacturing system to obtain the processing results of each process parameter set. In one possible design, the production model can be, for example, a HHO-BP (Harris Hawk Optimization-Back Propagation) neural network model.

[0078] Therefore, as an embodiment, S202 may include the following steps:

[0079] S202a, obtaining a trained HHO-BP neural network model, where the HHO-BP neural network model is used to predict processing results generated when the manufacturing system works according to input data.

[0080] S202b, using at least one process parameter set as input data and inputting it into the HHO-BP neural network model to obtain processing results of each process parameter set.

[0081] Optionally, the HHO-BP neural network model can be obtained by pre-training or real-time training. In this embodiment, there are no excessive restrictions on the training steps of the HHO-BP neural network model.

[0082] Figure 3 The application example diagram of the HHO-BP neural network model in the embodiment of the present disclosure is shown. The input of the HHO-BP neural network model is each process parameter set, and the output is the processing result of each process parameter set.

[0083] Optionally, S202a may include acquiring a pre-constructed HHO-BP neural network model, loading model parameters of the HHO-BP neural network model into the constructed HHO-BP neural network model, and obtaining a trained HHO-BP neural network model.

[0084] Optionally, the model input of the HHO-BP neural network model is each process parameter set, and the model output of the HHO-BP neural network model is the processing result corresponding to each process parameter set.

[0085] In the disclosed embodiment, the HHO-BP neural network model is involved in the processing simulation of the manufacturing system. The model has a high global search capability and can better fit the inherent laws of the data, rather than simply memorizing the training data. In this way, when faced with each set of process parameters, more accurate predictions can be made, which effectively enhances the accuracy of the network's processing simulation of the manufacturing system, allowing it to play a more reliable role in the processing process of the simulated manufacturing system.

[0086] As an embodiment, the training steps of the HHO-BP neural network model may include:

[0087] Obtain at least one training sample, the training sample includes training data and true values, the training data includes a process parameter set of the manufacturing system when it is actually working, and the true value is a real quality characteristic value generated by each machine when the manufacturing system works according to the actual process parameter set;

[0088] Taking the prediction result being the same as the true value corresponding to each training sample as the training goal, the HHO-BP neural network model is trained using at least one training sample to obtain the HHO-BP neural network model at the end of the training.

[0089] Among them, the true quality characteristic value is the true processing result of the process parameter set when the manufacturing system actually works, such as diameter, length, weld length, roughness and other data.

[0090] Optionally, before obtaining at least one training sample, the method further includes: establishing a HHO-BP neural network model. Specifically, the BP neural network model may be optimized according to the Harris Hawk algorithm.

[0091] Optionally, after acquiring at least one training sample, the HHO-BP neural network model may be trained using the at least one training sample to obtain the HHO-BP neural network model at the end of the training.

[0092] Optionally, taking the prediction result being the same as the true value corresponding to each training sample as a training goal, using at least one training sample to train the HHO-BP neural network model, and obtaining the HHO-BP neural network model at the end of training, can include: determining a first model parameter of the HHO-BP neural network model, inputting the training data in at least one training sample into the HHO-BP neural network model corresponding to the first model parameter, and obtaining the prediction result of each training sample through the HHO-BP neural network model. Calculating the error between the prediction result and the true value of each training sample to obtain the error corresponding to at least one training sample. Adding and summing the errors corresponding to at least one training sample to obtain the model prediction error.

[0093] If the model prediction error meets the preset convergence condition, the HHO-BP neural network model corresponding to the first model parameter is determined to be the HHO-BP neural network model obtained after the training. If the model prediction error does not meet the preset convergence condition, the first model parameter is updated, and the training data in at least one training sample is returned to the HHO-BP neural network model corresponding to the first model parameter to continue execution.

[0094] Optionally, in order to obtain training data, key process parameter combinations, key quality characteristics, and product quality data of products can be collected during the actual operation of the manufacturing system, as well as different process parameter sets in actual production. In addition, data such as the pass rate at the end of production can also be collected.

[0095] In the disclosed embodiment, the HHO-BP neural network model optimizes the network parameters with the help of HHO-BP, and the algorithm has a faster convergence speed. Compared with relying solely on the training mechanism of the BP neural network itself, the HHO-BP model can find more suitable network parameters in a shorter time, thereby speeding up the network training process and saving training time.

[0096] like Figure 4 As shown, it is a schematic flow chart of another manufacturing system health analysis method provided by the embodiment of the present disclosure. The difference from the above embodiment is that S203 may include the following steps:

[0097] S401: According to the processing results of each process parameter set, the quality characteristic value corresponding to each machine in at least one process parameter combination is obtained.

[0098] Optionally, the processing result of the process parameter set may include: the quality characteristic value corresponding to the process parameter combination of each machine in the process parameter set.

[0099] The quality characteristic value can be the value of the key quality characteristic obtained by processing the workpiece according to the process parameter combination of the machine. Taking the centerless grinder in Table 1 as an example, its key quality characteristic is the diameter, and the process parameter combination includes: grinding wheel speed, workpiece speed and feed rate. The centerless grinder can grind the diameter of the workpiece, and the quality characteristic value can be the value of the diameter of the workpiece. The quality characteristic value corresponding to the process parameter combination of the centerless grinder in each process parameter set can be obtained.

[0100] For example, assuming that the manufacturing system includes four machines, namely machines 1-4, the structure of the process parameter set of the manufacturing system may be as follows:

[0101] [Machine 1: process parameter combination A, Machine 2: process parameter combination B, Machine 3: process parameter combination C, Machine 4: process parameter combination D]. Taking Machine 1 as a centerless grinder as an example, process parameter combination A can include sub-parameters such as grinding wheel speed, workpiece speed, and feed amount. Taking Machine 2 as a surface grinder as an example, process parameter combination B can include sub-parameters such as grinding wheel speed, workpiece speed, and feed amount. Taking Machine 3 as an electric welder as an example, process parameter combination C can include sub-parameters such as welding speed, arc voltage, and welding current. Taking Machine 4 as a grinder as an example, process parameter combination D can include sub-parameters such as grinding wheel speed, workpiece speed, and backcut depth. Each process parameter combination can include multiple sub-parameters.

[0102] For the processing result of each process parameter set, each machine may have a corresponding quality characteristic value in at least one process parameter combination. Taking the process parameter combination A as an example, the quality characteristic value associated with the process parameter combination A may refer to the value of the key quality characteristic (i.e., diameter) obtained by machine 1 processing the workpiece according to the process parameter combination A (e.g., the diameter value is 14.514).

[0103] Table 2 below provides examples of process parameter sets, process parameter combinations, and quality characteristic values.

[0104]

[0105] Table 2

[0106] Referring to Table 2, process parameter set 1, the process parameter combination in machine 1 is A1, the process parameter combination in machine 2 is B1, the process parameter combination in machine 3 is C1, and the process parameter combination in machine 4 is D1.

[0107] The processing results corresponding to process parameter set 1 may include: quality characteristic value a1 corresponding to process parameter combination A1 of machine 1, quality characteristic value b1 corresponding to process parameter combination B1 of machine 2, quality characteristic value c1 corresponding to process parameter combination C1 of machine 3, and quality characteristic value d1 corresponding to process parameter combination d1 of machine 4.

[0108] Optionally, S401 may include: according to the processing results of each process parameter set, that is, the quality characteristic values ​​of each process parameter set corresponding to at least one machine, reading the quality characteristic values ​​corresponding to each process parameter set in each machine from the quality characteristic values ​​of each process parameter set corresponding to at least one machine, that is, when the process parameter set includes N sets, each machine contains N quality characteristic values.

[0109] For example, the quality characteristic values ​​corresponding to at least one process parameter combination of machine 1 can be read, that is, for process parameter combinations A1, A2, A3 and A4, the quality characteristic value a1 corresponding to process parameter combination A1, the quality characteristic value a2 corresponding to process parameter combination A2, the quality characteristic value a3 corresponding to process parameter combination A3, and the quality characteristic value a4 corresponding to process parameter combination A4 are read. Optionally, the quality characteristic value can be a quality characteristic value set according to historical experience or usage requirements. If there is a design requirement for the value range of the key quality characteristic of the workpiece (i.e., the processing object or product), for example, 50mm±1mm, the preset target processing result is [49mm, 51mm]. If the estimated processing result meets this range, it is determined that the corresponding parameter combination can be used. The preset quality characteristic value can determine whether the estimated processing result of the target process parameter combination of each machine is within the design requirement range. If it is within the design requirement range, the target process parameter combination of each machine can be used to participate in the processing of the manufacturing system. If it is not within the design requirement range, the target process parameter combination of each machine can be deleted and return to steps S201 or S203.

[0110] As an optional implementation, S401 may include: obtaining preset quality characteristic values ​​of workpieces required to be processed by each machine; and obtaining at least one quality characteristic value that meets the preset quality characteristic value of the machine from the quality characteristic values ​​of each machine based on the processing results of each process parameter set.

[0111] Optionally, the preset quality characteristic value of each machine refers to the value range that each machine needs to meet in the key quality characteristic. Specifically, the quality characteristic value of each machine in all processing results can be obtained; and according to all quality characteristic values ​​and preset quality characteristic values ​​of each machine, the quality characteristic value of each machine is screened to obtain at least one quality characteristic value that meets the preset quality characteristic value of each machine. In the screening process, the quality characteristic values ​​of each machine that do not meet the preset quality characteristic value can be directly deleted.

[0112] Furthermore, based on all the quality characteristic values ​​and preset quality characteristic values ​​of each machine, the quality characteristic values ​​of each machine are screened to obtain at least one quality characteristic value that meets the preset quality characteristic value of each machine, which may include: for any machine, determining, from at least one quality characteristic value of the machine, a valid quality characteristic value that belongs to the value range of the preset quality characteristic value corresponding to the machine.

[0113] For example, assuming that the preset value range of the machine is [49mm, 50mm], the mass characteristic value greater than or equal to 49mm and less than or equal to 50mm can be determined as the effective mass characteristic value.

[0114] By presetting quality characteristic values, preliminary screening of the parameters of each machine that participate in subsequent quality characteristic values ​​can obtain parameter combinations corresponding to quality characteristic values ​​with relatively stable processing quality, thereby further improving the screening accuracy of parameter combinations.

[0115] S402: Calculate the system health index corresponding to at least one quality characteristic value of each machine. Optionally, for any machine, the system health index corresponding to each quality characteristic value of the machine can be determined according to each quality characteristic value of the machine.

[0116] Taking the process parameter set in Table 2 as an example, the system health indicators of each quality characteristic value are illustrated. The quality characteristic value a1 corresponds to the system health indicator s1, the quality characteristic value a2 corresponds to the system health indicator s2, the quality characteristic value a3 corresponds to the system health indicator s3, and the quality characteristic value a4 corresponds to the system health indicator s4.

[0117] Optionally, S402 may include: calculating the system health index corresponding to the quality characteristic value of each machine according to the health index calculation model.

[0118] S403: Selecting a target process parameter combination for each machine from the corresponding at least one process parameter combination according to the system health indicator corresponding to each machine in the at least one process parameter combination.

[0119] Optionally, each process parameter combination may include one or more system health indicators. The system health indicator may refer to an indicator used to indicate the operating health status of the machine under the effect of the process parameter combination.

[0120] For any machine, the machine may be associated with a process parameter combination in at least one process parameter set, and each process parameter combination may correspond to a system health indicator, so that a corresponding target process parameter component may be selected from the at least one process parameter combination according to the system health indicator corresponding to the at least one process parameter combination of the machine.

[0121] Specifically, a target process parameter combination with a more stable system health indicator can be selected from at least one process parameter combination corresponding to each machine. For example, a system health indicator with the smallest gap between the system health indicator and the processing target can be selected from at least one process parameter combination corresponding to each machine, and the process parameter combination corresponding to the system health indicator with the smallest gap can be determined as the target process parameter combination.

[0122] Taking the deviation ratio and signal-to-noise ratio as system health indicators, for at least one process parameter combination of any machine, a process parameter combination with a relatively small deviation and a relatively high signal-to-noise ratio can be selected as the target process parameter combination. Specifically, the selection can be made from the two dimensions of the deviation ratio and signal-to-noise ratio of each process parameter combination of the machine, so that the selected target process parameter combination can ensure a relatively balanced deviation ratio and signal-to-noise ratio, and the selected target process parameter combination can have a relatively stable and efficient processing effect.

[0123] In the disclosed embodiment, when selecting the target process parameter combination for each machine from each process parameter set, it is clearly pointed out that the quality characteristic values ​​corresponding to each machine under different process parameter combinations are obtained based on the processing results of each process parameter set, which lays the foundation for the subsequent analysis and operation. The system health index is further calculated based on the quality characteristic values ​​obtained previously, and the data association between the quality characteristic values ​​of each machine in at least one corresponding process parameter combination and the system health index of each process parameter combination is clearly demonstrated. Then, the selection of the target process parameter combination is completed based on the calculated system health index. The entire process is logically coherent, which provides strong support for the reasonable selection of the target process parameter combination and improves the scientificity and accuracy of the decision-making process of the target process parameter combination of each machine.

[0124] As an optional implementation, S402: calculating the system health index corresponding to each machine in at least one process parameter combination according to the quality characteristic value corresponding to each machine in at least one process parameter combination may include:

[0125] S402a. According to at least one process parameter combination of each machine and in combination with a preset response model, the expectations of each machine corresponding to at least one process parameter combination are calculated. The response model refers to a formula for calculating the expectations of each machine in each process parameter combination.

[0126] Optionally, S402a may include: inputting each process parameter combination of each machine into a response model to obtain an expectation of each process parameter combination.

[0127] Alternatively, the response model of a manufacturing system with multiple machines in series can be expressed as follows:

[0128]

[0129] in, is the regression coefficient of the variable of the ith machine, is the regression coefficient of the controllable factor, is the regression coefficient of the qth noise factor, is the quadratic regression coefficient of the jth controllable variable, is the regression coefficient of the interaction term between the jth controllable variable and the qth noise variable, is a residual variable that follows a normal distribution. is a matrix formed by at least one process parameter combination of each machine, A matrix formed by the expectations corresponding to at least one process parameter combination of each machine. It is a matrix formed by the expectations corresponding to at least one process parameter group of each machine during the last calculation. is a constant, which can be calculated in advance. For details, please refer to the content of the relevant technology and will not be repeated here. m is the total number of machines in the manufacturing system. A matrix formed by combining at least one noise factor for each processing machine. Refers to the sequence of workstations or the location of machines, such as workstation It refers to Workstation, or located at a location Machine, or machine It refers to A machine.

[0130] Among them, at the workstation (or machine) The controllable variables (such as process parameter combinations) on , at the workstation The noise variable on (which can be obtained by definition or data collection) can be expressed as , at the workstation The response variable (e.g., the output of a response model) on can be expressed as ,in The workstations The total number of controllable variables, noise variables, and response variables, where .

[0131] The following is an example of the calculation matrix of the response model:

[0132]

[0133] Among them, the expectation of each process parameter can be obtained through the above matrix calculation. and Take the calculation as an example.

[0134]

[0135]

[0136] S402b. Calculate the deviations of each machine in at least one process parameter combination based on the quality characteristic values ​​of each machine in at least one process parameter combination in combination with a preset deviation model. The deviation model refers to a formula for calculating the deviations of each machine in different process parameter combinations.

[0137] Optionally, S402b may include inputting the quality characteristic value of each process parameter combination of each machine into a preset deviation model to calculate the deviation of each process parameter combination of each machine.

[0138] Alternatively, the deviation model of a manufacturing system with multiple machines in series can be expressed as follows:

[0139]

[0140] in, It refers to A machine or workstation, a machine can be located at a workstation, The value range of is 1-m, where m is the total number of machines in the manufacturing system. ]、[ ]、[ ] is the Jacobian matrix, which can be represented by F( , , ), is the model error in the variance equation. is the process parameter combination, is the expected value of the process parameter combination, For the The noise parameters of a processing machine. F ( , , ) is the response model The variation of , specifically used to represent the mapping relationship between an upstream workstation (a workstation is, for example, a machine in a manufacturing system) and a downstream workstation.

[0141] is the deviation of the kth machine; For the Deviation of the machine; For the The deviation between the process parameters actually executed by a machine and the input process parameters; For the The deviation between the actual noise interference parameters of a machine and the ideal noise interference parameters.

[0142] After formula conversion, the response model and deviation model can be calculated by the following formula:

[0143]

[0144]

[0145] in, is the response model; Q is ( )’s state space; is the process parameter combination; is the expectation of process parameter combination; is the noise factor, represents the expectation calculated by the response model corresponding to the i-th machine, i is the i-th machine, the value range is 1-m, m is the number of machines in the manufacturing system, The deviation obtained by calculating the deviation mode corresponding to the i-th machine.

[0146] S402c, according to the expectations and deviations respectively associated with each machine in at least one process parameter combination, in combination with a preset system health indicator quantification model, calculate the system health indicators respectively corresponding to each machine in at least one process parameter combination.

[0147] Taking the expected value and deviation as examples, the system health indicators for different process parameter combinations are illustrated.

[0148] Table 3 below shows the key quality characteristics, quality characteristic values, expected values ​​and deviations.

[0149]

[0150] Table 3

[0151] In the disclosed embodiment, expectations are calculated based on at least one process parameter combination of each machine in combination with a preset response model, which lays the analytical foundation for subsequent analysis. The quality characteristic values ​​of each machine under the process parameter combination are combined with a preset deviation model to calculate the deviations corresponding to at least one processing parameter of each machine. The expectations and deviations obtained previously are then combined with the preset system health indicator quantification model to calculate the system health index, so that the entire process can achieve the acquisition of system health indicators for each machine from multiple dimensions. The comprehensive consideration of multiple factors makes the final system health indicator more convincing and valuable for reference, and can better guide the subsequent process parameter combination screening operations.

[0152] As an optional implementation manner, the difference from other embodiments is that, S402c, according to the expectation and deviation respectively associated with each machine in at least one process parameter combination, in combination with a preset system health indicator quantification model, calculating the system health indicator corresponding to each machine in at least one process parameter combination, may include the following steps:

[0153] S1. Determine the design characteristic formula and actual characteristic formula in the system health indicator quantification model.

[0154] Optionally, the system health indicator quantification model can be shown as the following formula:

[0155]

[0156] in, is the maximum value of the ith process parameter combination, is the minimum value of the i-th process parameter combination. is the deviation of the ith process parameter combination. is the variance-covariance matrix determined by the deviation of each machine. C refers to the cost matrix reflecting economic efficiency. It reflects the stability of the manufacturing system, and the stability is inversely proportional to the variance. is the expected mass loss, is the expectation corresponding to the i-th process parameter combination of the machine, is the processing target corresponding to the i-th process parameter combination of the machine, that is, the actual value of the machine at the i-th process parameter combination, is the weight of the i-th machine.

[0157] From the above formula, we can see that is the design characteristic formula,

[0158] is the actual characteristic formula. When calculating the system design characteristic and critical mass characteristic value of each process parameter combination respectively, the value of n can be 1 and the value of r can be 1.

[0159] S2. Input the processing target of each machine and the expectation of associating each machine with at least one process parameter combination into the design characteristic formula to obtain the system design characteristics of each machine corresponding to each process parameter combination.

[0160] S3. Input the deviation associated with each machine at each process parameter combination into the actual characteristic formula to obtain the critical mass characteristic corresponding to each machine at each process parameter combination.

[0161] S4. Calculate the quotient of the system design characteristic and the critical mass characteristic of each machine at each process parameter combination to obtain the deviation ratio of each machine at each process parameter combination.

[0162] S5. Calculate the signal-to-noise ratio corresponding to the expectation and deviation of each process parameter combination according to the signal-to-noise ratio calculation formula.

[0163] This study takes the signal-to-noise ratio of the eye-viewing characteristic as an example. The signal-to-noise ratio calculation formula is: ,in , , is the expectation of the ith machine, is the expected mean of each machine in the manufacturing system.

[0164] S6. Determine the deviation ratio and signal-to-noise ratio of each process parameter combination as the system health indicator of each process parameter combination.

[0165] The system health index for each process parameter combination can be composed of a deviation ratio and a signal-to-noise ratio.

[0166] Table 4 below shows examples of system health indicators for process parameter combinations.

[0167]

[0168] Table 4

[0169] In the disclosed embodiment, the key formula in the quantitative model of the system health index is determined, and then different data (processing goals, expectations and deviation ratios) are input into the corresponding formulas to obtain two intermediate results: the system design characteristics and the critical mass characteristics. Finally, the system health index is obtained by calculation based on the two intermediate results obtained above. This step-by-step process setting provides a reliable and quantifiable means for accurately measuring the system health status of each machine under different process parameter combinations, and improves the scientificity and accuracy of the evaluation of the system health index of each process parameter combination.

[0170] like Figure 5 As shown, it is a schematic flow chart of another manufacturing system health analysis method provided by the embodiment of the present disclosure. The difference from the above embodiment is that S204 may include the following steps:

[0171] S501. When the manufacturing system works according to the target process parameter combination of each machine, the overall stability index and the overall processing index of the manufacturing system are determined. The overall stability index refers to an index that can represent the stability of the manufacturing system, and the overall processing index refers to an index that can represent the health status of the processing results of the manufacturing system.

[0172] Referring to Table 4 above, the target process parameter combination of machine 1 is selected as A2, the target process parameter combination of machine 2 is selected as B5, the target process parameter combination of machine 3 is selected as C1, and the target process parameter combination of machine 4 is selected as D7.

[0173] Furthermore, the overall stability index and overall processing index of the manufacturing system composed of machine 1, machine 2, machine 3 and machine 4 can be determined according to the target process parameter combination A2 of machine 1, the target process parameter combination B5 of machine 2, the target process parameter combination C1 of machine 3 and the target process parameter combination D7 of machine 4.

[0174] Optionally, S501 may include: calculating an overall stability index and an overall processing index of the manufacturing system according to a target process parameter combination of each machine.

[0175] S502. Calculate an estimated processing result of the manufacturing system according to the overall stability index and the overall processing index of the manufacturing system.

[0176] Optionally, the estimated processing result of the manufacturing system may be calculated according to a calculation formula for the estimated processing result using the overall stability index and the overall processing index.

[0177] S503: If the estimated processing result reaches the preset target processing result, the manufacturing system is controlled to work according to the target process parameter combination of each machine.

[0178] Optionally, the preset target processing result may be a threshold value or a value range.

[0179] Optionally, the estimated processing result reaching a preset target processing result may mean that the estimated processing result is greater than or equal to a threshold of a preset target processing result or falls within a value range of a preset target processing result.

[0180] In the disclosed embodiment, the overall stability index and overall processing index of the manufacturing system are first determined when the manufacturing system works according to the target process parameter combination of each machine. These two indicators quantify the operational stability of the manufacturing system from different angles. The overall stability index focuses on the stability of the manufacturing system, while the overall processing index focuses on the processing results of the manufacturing system. By considering these two important aspects at the same time, the actual situation of the manufacturing system when working according to the target process parameter combination of each machine can be grasped more comprehensively and objectively, avoiding the one-sidedness caused by evaluating the stability of the system from only a single dimension. Therefore, according to the overall stability index and overall processing index of the manufacturing system, the estimated processing results calculated can more comprehensively characterize the health of the system. It provides a reliable and comparable basis for the subsequent accurate judgment of whether the estimated processing results meet the preset target processing results, which helps to make parameter decisions more scientifically.

[0181] Further, based on any of the above embodiments, S501, when the manufacturing system works according to the target process parameter combination of each machine, determining the overall stability index and the overall processing index of the manufacturing system may include the following steps:

[0182] S501a. When the manufacturing system operates according to the target process parameter combination of each machine, an overall stability index of the manufacturing system is obtained.

[0183] S501b. Based on the stability-oriented quality random flow network model, the task execution status, machine performance status and product quality status of the manufacturing system are modeled to obtain the overall processing indicators of the manufacturing system when it works according to the target process parameter combination of each machine.

[0184] Optionally, the robustness-oriented quality stochastic flow network (RQSFN) model can refer to a model that models the task execution status, machine performance status, and work-in-process quality status based on existing data and process solutions. Specifically, it refers to quantitatively analyzing the task execution status based on the actual production characteristics of the manufacturing machine, where the calculation of the output qualification rate is the key to analyzing the machine task execution status.

[0185] In the method provided by the embodiment of the present disclosure, the overall stability index is obtained under the actual operating condition that the manufacturing system works according to the target process parameter combination of each machine. The obtained overall stability index can truly reflect the stability of the manufacturing system under the setting of a specific process parameter combination. Based on the stability-oriented quality random flow network model, the three key and interrelated aspects of the task execution status, machine performance status and product quality status of the manufacturing system are modeled to obtain the overall processing index. This multi-dimensional modeling method fully considers the different levels of the manufacturing system operation process, not only focusing on the performance of the machine itself, but also involving the smoothness of task execution and the quality of the final product. It can more comprehensively and accurately characterize the overall processing effect of the manufacturing system when working according to the target process parameter combination, so that the obtained overall processing index has stronger comprehensiveness and representativeness.

[0186] As an optional implementation, S501a may include:

[0187] Obtain the machining targets, expectations and deviation ratios associated with the target process parameter combinations for each machine.

[0188] According to the expectations and processing goals associated with the target process parameter combination of each machine, combined with the quality loss function of the manufacturing system, the overall quality loss of the manufacturing system is calculated.

[0189] According to the overall quality loss and the expected loss function of the manufacturing system, the overall processing index of the manufacturing system is calculated.

[0190] According to the overall quality loss, the overall expected loss and the deviation ratio of the target process parameter combination of each machine, the overall stability index of the manufacturing system is calculated in combination with the system health index quantification model. The deviation ratio of the target process parameter combination is obtained by calculating the expectation and deviation of each machine in the corresponding target process parameter combination.

[0191] It should be noted that the technical solution of this embodiment is an analysis of the machine after selecting the target process parameters, and the process of selecting the target process parameter combination involves the calculation of data such as the processing objectives, expectations, deviations, deviation ratios, and signal-to-noise ratios of each process parameter combination. Therefore, after determining the target process parameter combination of each machine, the expectations, processing objectives, and deviation ratios calculated for the target process parameter combination of each machine can be used to participate in the operation analysis process of the target process parameter combination of each machine. Therefore, for the calculation of the processing objectives, expectations, deviations, signal-to-noise ratios, deviation ratios, and other parameters related to the target process parameter combination, reference can be made to the relevant contents of the above embodiments, which will not be repeated here.

[0192] Alternatively, the quality loss function can be as follows:

[0193]

[0194] in, is the expectation of the i-th machine, i ranges from 1 to r, and r is the number of machines. is the weight of the i-th machine, is the processing target of the i-th machine.

[0195] Alternatively, the expected loss function can be expressed as follows:

[0196]

[0197] Taking the manufacturing system including the above four devices as an example, the weight of each machine can be set ,like , and the processing objectives of each machine are as follows: . Input the above values ​​into the quality loss function and we get:

[0198]

[0199] Combined with the expectations of each machine , the overall mass loss L=0.544 is calculated.

[0200] Combined with the overall mass loss, the overall processing index is calculated:

[0201]

[0202] The overall processing index E=0.076 is obtained.

[0203] Optionally, the system health indicator quantification model can be as follows:

[0204]

[0205] Among them, L is the overall mass loss, E is the overall processing index, is the deviation ratio of the target process parameter combination of the i-th machine, and r is the number of machines.

[0206] Also combined with the above example, the manufacturing system uses the target process parameters of each machine to combine the corresponding overall stability index The calculation formula for can be, for example:

[0207]

[0208] In the method provided by the embodiment of the present disclosure, the key elements of the target process parameter combination association of each machine, namely, the processing target, expectation and deviation ratio, are first clearly obtained, which lays the foundation for the subsequent calculations. Then, based on the expectations of the target process parameter combination association of each machine obtained, and combined with the quality loss function of the manufacturing system, the overall quality loss of the manufacturing system is calculated, and the progressive calculation method of the overall quality loss of the manufacturing system is demonstrated. Subsequently, based on the calculated overall quality loss, the overall processing index is further calculated in combination with the expected loss function, so that the calculation process continues to deepen. Finally, the overall quality loss, the overall expected loss and the target quality characteristic values ​​of each machine are combined, and the overall stability index is calculated through the system health index quantification model. The whole process steps are coherent, allowing people to clearly understand the complete context from basic data acquisition to the final calculation of each important index. This provides a reliable and quantifiable means for accurately measuring the overall stability index of the manufacturing system, and improves the scientificity and accuracy of the evaluation.

[0209] As an optional implementation, S501b may specifically include several steps a to h:

[0210] a. Divide the processing capacity of each machine to obtain multiple discrete states corresponding to the processing capacity of each machine.

[0211] Optionally, in step a, the processing capacity of each machine may be divided to obtain multiple discrete states of each machine.

[0212] Furthermore, the number of divisions can be obtained in advance, and the processing capacity of each machine can be divided into m discrete states according to the preset number of divisions m, such as using represents the m discrete states of machine i.

[0213] Furthermore, the processing capability of each machine may be scored, and the processing capability of each machine may be divided into a plurality of discrete states according to the score of each machine.

[0214] The following still takes 4 machines divided into 11 discrete states as an example, showing an example of discrete division of machines 1 to 4:

[0215]

[0216]

[0217]

[0218]

[0219] b. Based on the multiple discrete states corresponding to the processing capacity of each machine and the first probability parameter of each machine, construct probability formulas corresponding to the multiple discrete states of each machine, where the first probability parameter is unknown.

[0220] Optionally, after obtaining multiple discrete states of each machine, probability formulas corresponding to the multiple discrete states of the machine can be constructed using the first probability parameter of each machine. The probability formulas corresponding to the multiple discrete states of each machine can be expressed as: ,in, represents the proportional relationship between machine i in different discrete states, represents the first probability parameter of the i-th machine. And the sum of the probability formulas corresponding to the multiple discrete states of each machine is equal to 1, which is expressed as: .

[0221] The following still takes machines 1 to 4 as an example to show the probability formula of each machine:

[0222]

[0223]

[0224]

[0225]

[0226] in, is the first probability parameter of machine 1, The first probability parameter of machine 2, The first probability parameter of machine 3, is the first parameter probability of machine 4.

[0227] Optionally, constructing the probability formulas corresponding to the multiple discrete states of each machine may include determining the proportion data for the multiple discrete states of each machine. , the first constant and corresponding proportional data of each discrete state Multiply them together to get the probability formula for each discrete state. For example, the probability formula for the discrete state 0 of machine 1 is , the probability formula for the discrete state 50 is 2 , and so on.

[0228] c. Obtain the expected downtime and failure rate of each machine.

[0229] Optionally, the estimated downtime and failure rate of each machine may be determined based on the actual operation data of each machine. The actual operation data of each machine may refer to the data of the actual operation process of each machine. The estimated downtime may refer to the time that each machine may be downtime during operation. The failure rate may refer to the ratio of the abnormal operation time of the machine in one day to the time of one day.

[0230] The expected downtime and failure rates for each machine are as follows:

[0231]

[0232]

[0233] d. Input the expected downtime and failure rate of each machine into the preset processing state calculation formula to calculate the parameter value of the first probability parameter of each machine.

[0234] Optionally, the processing state calculation formula is as follows:

[0235]

[0236] in, is the maximum value of the discrete state of the i-th machine, is the value of the discrete state x of the i-th machine, is the downtime of the i-th machine, is the failure rate of the i-th machine, is the number of failures / malfunctions of machine i in time t, t is the number of days the manufacturing system needs to process, and m is the number of discrete states.

[0237] Taking T=300 as an example, the parameter value of the first probability parameter of each machine is calculated:

[0238]

[0239]

[0240]

[0241]

[0242] e. Input the parameter value of the first probability parameter of each machine into the probability formula of each machine for multiple discrete states, and obtain the probability of each machine corresponding to the corresponding multiple discrete states.

[0243] By substituting the parameter value of the first probability parameter of each machine into the probability formula corresponding to the multiple discrete states, the probabilities corresponding to the multiple discrete states of each machine can be obtained.

[0244] Taking the first machine as an example, Substitute into the probability formula of machine 1 ,get:

[0245] f. Obtain the minimum input flow of each machine.

[0246] Optionally, obtaining the minimum input flow of each machine may include: determining the minimum input flow of each machine according to a product qualification rate of each machine.

[0247] Optionally, the product qualification rate of each machine can be determined based on the historical processing results of each machine. The historical processing results may include the number of qualified products and the total amount of products. The quotient of the number of qualified products and the total amount of products is calculated to obtain the product qualification rate of the machine.

[0248] For example, the product qualification rate of each machine is:

[0249]

[0250] Optionally, the minimum input flow rate of each machine may refer to the minimum flow rate required for each machine to operate normally. For example, the minimum input flow rate of each machine is:

[0251]

[0252] Specifically, the minimum input flow of each machine can be determined based on the product qualification rate of each machine and the normal working flow required for each machine under normal working conditions.

[0253] g. Based on the minimum input flow of each machine and the probabilities corresponding to the corresponding multiple discrete states, calculate the probability that the machine processing state of each machine is not less than its minimum input flow.

[0254] Optionally, the probability that the machine processing state of each machine is not less than the minimum input flow rate may refer to the probability that the machine processing state of each machine is greater than or equal to the minimum input flow rate.

[0255] For example, the minimum flow rate of machine 1 is 260.23, and among the multiple discrete states of machine 1, the discrete states that are greater than the minimum flow rate are 270, 290, and 315. Accordingly, the probability that the machine processing state of machine 1 is not less than its minimum input flow rate is:

[0256]

[0257] Similarly, the probability that the machine processing state of machine 2-4 is not less than its minimum input flow is:

[0258]

[0259]

[0260]

[0261] h. Multiply the probability that the machine processing state of each machine is not less than its minimum input flow rate, and calculate the overall processing index of the manufacturing system.

[0262] Optionally, the machine processing state of each machine is not less than its minimum input flow rate can be used The probability that the machine processing state of each machine is not less than its minimum input flow can be multiplied by the formula: Indicates that the product obtained by calculation That is, the overall processing index of the entire manufacturing system. For example, the probability multiplication formula of the machine processing state of each machine being not less than its minimum input flow can be expressed as:

[0263] .

[0264] In the method provided by the embodiment of the present disclosure, the processing capacity of each machine is first divided to obtain discrete states, which provides a basic classification basis for subsequent steps. Then, a probability formula is constructed based on the previous discrete states and the unknown first probability parameter, and the relationship between each discrete state and the probability is clarified. Then, two key data, the expected downtime and the failure rate, are obtained for subsequent calculation of the first probability parameter. The value of the first probability parameter is calculated using the acquired data, and it is substituted into the probability formula to obtain the probability of each machine in different discrete states. After that, new data such as product qualification rate and minimum input flow are obtained, and the specific probability of each machine is calculated based on the various data obtained previously, and finally the relevant probabilities of each machine are multiplied to obtain the overall processing index of the manufacturing system. By incorporating these different factors into the calculation system of the overall processing index, the actual processing status of the manufacturing system can be more comprehensively and objectively reflected, avoiding the one-sidedness brought by the evaluation of only a single factor, so that the overall processing index finally obtained is more convincing and has reference value.

[0265] Further, based on any of the above embodiments, S502, calculating the estimated processing result of the manufacturing system according to the overall stability index and the overall processing index of the manufacturing system, may include the following steps:

[0266] A health status quantitative model of the manufacturing system is obtained; the overall processing index and overall stability index of the manufacturing system are input into the health status quantitative model to obtain an estimated processing result of the manufacturing system.

[0267] Optionally, the functional health of a manufacturing system is defined as:

[0268]

[0269] in, represents the stability index of the manufacturing system, Indicates the machine performance status of the manufacturing system, Indicates the quality status of the products manufactured by the manufacturing system, Represents the production task execution status of the manufacturing system. The estimated processing results of the manufacturing system obtained by calculation. ,in, For the The task execution status of each machine, For the The minimum input flow of a machine, It refers to Machines meet Therefore, according to the system health indicator quantification model, the health status quantification model can be converted to:

[0270]

[0271] Still taking the above as an example, the overall processing indicators =0.5325, overall stability index =0.1766, inputting the overall processing index and the overall stability index into the health status quantitative model, the estimated processing result of the manufacturing system as a whole can be obtained H (t) = 0.5325 (1-0.1766)=0.4385.

[0272] In the method provided by the embodiment of the present disclosure, the health status quantification model of the manufacturing system can quantitatively express the health status of the complex manufacturing system through mathematical formulas, algorithms, etc., so that the health status that was originally vague and difficult to measure accurately has a specific and operational evaluation basis. It can comprehensively consider many factors and convert these factors into quantifiable estimated processing results in a systematic way, thereby improving the scientificity and accuracy of the health evaluation of the manufacturing system and being able to evaluate the health status of the manufacturing system relatively quickly.

[0273] For ease of understanding, Figure 6An exemplary flow chart of a manufacturing system health analysis method provided by an embodiment of the present disclosure is shown. The manufacturing system health analysis method may include the following steps:

[0274] S601. Determine at least one process parameter set related to a manufacturing system, where the process parameter set includes a process parameter combination corresponding to at least one machine in the manufacturing system, and the process parameter combination includes at least one process parameter, which refers to an input parameter when the machine is working.

[0275] S602. Predict the processing results generated when the manufacturing system works according to the configuration of each process parameter set. The processing results include the quality characteristic values ​​generated when each machine works according to its own process parameter combination in the process parameter set. The quality characteristic values ​​are used to represent the values ​​of key quality characteristics after the machine processes the workpiece.

[0276] S603. According to the processing results of each process parameter set, the quality characteristic value corresponding to each machine in at least one process parameter combination is obtained.

[0277] S604. Calculate the expectation of each machine in at least one process parameter combination according to the process parameter combination of each machine in each process parameter set in combination with a preset response model. The response model refers to a formula for calculating the expectation of each machine in each process parameter combination.

[0278] S605. Calculate the deviations of each machine in at least one process parameter combination based on the quality characteristic values ​​of each machine in at least one process parameter combination and in combination with a preset deviation model. The deviation model refers to a formula for calculating the deviations of each machine in different process parameter combinations.

[0279] S606. Calculate the system health indicators corresponding to at least one process parameter combination of each machine according to the expectations and deviations respectively associated with at least one process parameter combination of each machine in combination with a preset system health indicator quantification model. The system health indicator quantification model refers to a calculation formula used to quantify the stability of each machine.

[0280] S607 . Select a target process parameter combination for each machine from the corresponding at least one process parameter combination according to the system health indicator corresponding to each machine in at least one process parameter combination.

[0281] S608. Obtain the processing target, expectation and deviation ratio corresponding to each machine when it operates with its target process parameter combination.

[0282] S609: Calculate the overall quality loss of the manufacturing system according to the expectations and processing targets associated with the target process parameter combinations of each machine and the quality loss function of the manufacturing system.

[0283] S610. Calculate the overall expected loss of the manufacturing system according to the overall quality loss and the expected loss function of the manufacturing system.

[0284] S611. According to the overall quality loss, the overall expected loss and the deviation ratio corresponding to the target process parameter combination of each machine, combined with the system health index quantification model, the overall stability index of the manufacturing system is calculated. The deviation ratio corresponding to the target process parameter combination is obtained by calculating the expectation and deviation of each machine in the corresponding target process parameter combination.

[0285] S612. Based on the stability-oriented quality random flow network model, the task execution status, machine performance status and product quality status of the manufacturing system are modeled to obtain the overall processing indicators of the manufacturing system when it works according to the target process parameter combination of each machine.

[0286] S613. Obtain a quantitative model of the health status of the manufacturing system.

[0287] S614: Input the overall processing index and overall stability index of the manufacturing system into the health status quantification model to obtain the estimated processing result of the manufacturing system.

[0288] S615: If the estimated processing result of the target process parameter combination corresponding to at least one machine satisfies the preset target processing result, control the manufacturing system to operate according to the target process parameter combination of each machine.

[0289] In the disclosed embodiment, starting from determining the process parameter set of the manufacturing system, going through the prediction of the processing results under different process parameter combination configurations, the acquisition and calculation of various indicators such as relevant quality characteristic values, and then selecting the target process parameter combination for the machine, and then paying attention to the situation of the machine under the target process parameter combination and the acquisition of the overall indicators of the manufacturing system, and finally obtaining the estimated processing results of the manufacturing system through the health status quantification model. The entire process covers all aspects from process parameter combination setting, machine individual indicator analysis to the overall evaluation of the manufacturing system, forming an organic whole, comprehensively and systematically handling many matters related to the operation and health evaluation of the manufacturing system, and ultimately improving the stability and accuracy of the acquisition of the estimated processing results of the entire manufacturing system.

[0290] like Figure 7 FIG. 1 is a schematic diagram of a structure of a manufacturing system health analysis device provided by an embodiment of the present disclosure. The manufacturing system health analysis device 700 may include the following units:

[0291] The determination unit 701 is used to determine at least one process parameter set related to the manufacturing system, the process parameter set includes a process parameter combination corresponding to at least one machine in the manufacturing system, the process parameter combination includes at least one process parameter, and the process parameter refers to an input parameter when the machine is working.

[0292] The prediction unit 702 is used to predict the processing results generated when the manufacturing system works according to the configuration of each process parameter set. The processing results include the quality characteristic values ​​generated when each machine works according to its own process parameter combination in the process parameter set. The quality characteristic values ​​are used to represent the values ​​of key quality characteristics after the machine processes the workpiece.

[0293] The selection unit 703 is used to select a target process parameter combination corresponding to at least one machine according to the processing results of each process parameter set.

[0294] The control unit 704 is used to control the manufacturing system to operate according to the target process parameter combination of each machine if the estimated processing result of the target process parameter combination corresponding to at least one machine meets the preset target processing result.

[0295] As an embodiment, the control unit may include:

[0296] The first determination module is used to determine the overall stability index and the overall processing index of the manufacturing system when the manufacturing system works according to the target process parameter combination of each machine. The overall stability index refers to an index that can represent the stability of the manufacturing system, and the overall processing index refers to an index that can represent the health status of the processing results of the manufacturing system.

[0297] The first calculation module is used to calculate the estimated processing result of the manufacturing system according to the overall stability index and the overall processing index of the manufacturing system.

[0298] The parameter control module is used to control the manufacturing system to work according to the target process parameter combination of each machine if the estimated processing result reaches the preset target processing result.

[0299] As yet another embodiment, the first calculation module may include:

[0300] The first submodule is used to obtain a quantitative model of the health status of the manufacturing system.

[0301] The second submodule is used to input the overall processing index and the overall stability index of the manufacturing system into the health status quantification model to obtain the estimated processing result of the manufacturing system.

[0302] As yet another embodiment, the first determining module may include:

[0303] The third submodule is used to obtain the overall stability index of the manufacturing system when the manufacturing system works according to the target process parameter combination of each machine.

[0304] The fourth submodule is used to model the task execution status, machine performance status and product quality status of the manufacturing system based on the stability-oriented quality random flow network model, and obtain the overall processing indicators when the manufacturing system works according to the target process parameter combination of each machine.

[0305] As yet another embodiment, the third submodule is specifically configured to:

[0306] Obtain the processing target, expectation and deviation ratio corresponding to each machine when it works in its target process parameter combination; calculate the overall quality loss of the manufacturing system based on the expectation and processing target associated with each machine in the target process parameter combination, combined with the quality loss function of the manufacturing system; calculate the overall expected loss of the manufacturing system based on the overall quality loss, combined with the expected loss function of the manufacturing system; calculate the overall stability index of the manufacturing system based on the overall quality loss, the overall expected loss and the deviation ratio corresponding to the target process parameter combination of each machine, combined with the system health index quantification model, and the deviation ratio corresponding to the target process parameter combination is obtained by calculating the expectation and deviation of each machine in the corresponding target process parameter combination.

[0307] As another embodiment, the fourth submodule may be specifically used for:

[0308] The processing capacity of each machine is divided to obtain multiple discrete states corresponding to the processing capacity of each machine; according to the multiple discrete states corresponding to the processing capacity of each machine and the first probability parameter of each machine, a probability formula corresponding to the multiple discrete states of each machine is constructed, and the first probability parameter is unknown; the expected downtime and failure rate of each machine are obtained;

[0309] Input the expected downtime and failure rate of each machine into the preset processing state calculation formula to calculate the parameter value of the first probability parameter of each machine; input the parameter value of the first probability parameter of each machine into the probability formula of each machine's multiple discrete states to obtain the probability of each machine in the corresponding multiple discrete states; obtain the minimum input flow of each machine; based on the minimum input flow of each machine and the probabilities corresponding to the corresponding multiple discrete states, calculate the probability that the machine processing state of each machine is not less than its minimum input flow; multiply the probabilities of the machine processing state of each machine being not less than its minimum input flow, and calculate the overall processing index of the manufacturing system.

[0310] As an embodiment, the prediction unit may include:

[0311] The model acquisition module is used to acquire the trained HHO-BP neural network model, and the HHO-BP neural network model is used to predict the processing results generated when the manufacturing system works according to the input data.

[0312] The result prediction module is used to use each process parameter set as input data, input into the HHO-BP neural network model, and obtain the processing results of each process parameter set.

[0313] As yet another embodiment, the device further comprises:

[0314] A sample acquisition unit, used to acquire at least one training sample, the training sample includes training data and true values, the training data includes a process parameter set of the manufacturing system when it is actually working, and the true value is a real quality characteristic value generated by each machine when the manufacturing system works according to the actual process parameter set;

[0315] The model training unit is used to train the HHO-BP neural network model with the prediction result being the same as the true value corresponding to each sample as the training goal, using at least one training sample to obtain the HHO-BP neural network model at the end of the training.

[0316] As yet another embodiment, the selection unit may include:

[0317] The quality acquisition module is used to acquire at least one quality characteristic value corresponding to each machine according to the processing results of each process parameter set.

[0318] The indicator calculation module is used to calculate the system health indicator corresponding to at least one quality characteristic value of each machine.

[0319] a target selection module, configured to select a target quality characteristic value from at least one quality characteristic value for each machine according to the system health indicator corresponding to the at least one quality characteristic value of each machine;

[0320] The combination determination module is used to determine the process parameter combination associated with the target quality characteristic value of each machine as the target process parameter combination, and obtain the target process parameter combination corresponding to at least one machine.

[0321] As yet another embodiment, the indicator calculation module may include:

[0322] A response calculation submodule, for calculating the expectation of each machine in at least one process parameter combination according to the process parameter combination of each machine in each process parameter set and in combination with a preset response model, wherein the response model refers to a formula for calculating the expectation of each machine in each process parameter combination;

[0323] A deviation calculation submodule, used to calculate the deviation of each machine in at least one process parameter combination according to the quality characteristic value of each machine in at least one process parameter combination, in combination with a preset deviation model, wherein the deviation model refers to a formula for calculating the deviation of each machine in different process parameter combinations;

[0324] The indicator quantification submodule is used to calculate the system health indicators corresponding to each machine in at least one process parameter combination according to the expectations and deviations associated with each machine in at least one process parameter combination, combined with a preset system health indicator quantification model. The system health indicator quantification model refers to a calculation formula used to quantify the stability of each machine.

[0325] As another embodiment, the indicator quantification submodule may be specifically used for:

[0326] Determine the design characteristic formula and the actual characteristic formula in the system health index quantification model; input the processing target of each machine and the expectation associated with each machine in at least one process parameter combination into the design characteristic formula to obtain the system design characteristic corresponding to each machine in each process parameter combination; input the deviation associated with each machine in each process parameter combination into the actual characteristic formula to obtain the critical mass characteristic corresponding to each machine in each process parameter combination; calculate the quotient of the system design characteristic and the critical mass characteristic of each machine in each process parameter combination to obtain the deviation ratio of each machine in each process parameter combination; calculate the signal-to-noise ratio corresponding to the expectation and deviation of each process parameter combination according to the signal-to-noise ratio calculation formula; determine the deviation ratio and signal-to-noise ratio of each process parameter combination as the system health indicator of each process parameter combination.

[0327] As another embodiment, the quality acquisition module can be specifically used to: obtain the preset quality characteristic values ​​of the workpieces required to be processed by each machine; and obtain at least one quality characteristic value that meets the preset quality characteristic value of the machine from the quality characteristic values ​​of each machine based on the processing results of each process parameter set.

[0328] In the above, with reference to the accompanying drawings, the manufacturing system health analysis device according to the embodiment of the present disclosure is described, which can determine at least one process parameter set related to the manufacturing system, and the process parameter set can include the process parameter combination of each machine in the manufacturing system. In addition, the process parameter set of the manufacturing system is used to simulate the production of the manufacturing system to obtain the processing results corresponding to each process parameter set. The processing results can include the quality characteristic value of each machine in the corresponding process parameter combination, and the quality characteristic value can directly characterize the processing effect of each process parameter set, obtain the target process parameter combination with better processing effect of each machine, and realize the preliminary screening of the target process parameter combination of each machine. Afterwards, when it is confirmed that the target process parameter combination of each machine meets the preset functional health prediction condition, the target process parameter combination is used for actual processing. Through the steps of the composition of the parameter combination, the production simulation of the manufacturing system, the screening of the target process parameter combination of each machine, the determination of the functional health prediction condition, etc., the dimension of predicting the manufacturing health status of the manufacturing system is established, and a process parameter combination with healthier operation is selected for the manufacturing system, and a more stable and higher processing accuracy process parameter combination is provided for the operation health of the manufacturing system, thereby ensuring the healthy operation of the manufacturing system.

[0329] also, Figure 8 This is a hardware block diagram of an electronic device provided in an embodiment of the present disclosure. The electronic device may include: a memory 801, a processor 802, and a computer program stored in the memory 801. The processor executes the computer program to implement the steps of the method provided in any of the above embodiments.

[0330] For example, the processor 802 may be configured to execute:

[0331] Determine at least one process parameter set related to the manufacturing system, the process parameter set includes a process parameter combination corresponding to at least one machine of the manufacturing system, the process parameter combination includes at least one process parameter, and the process parameter refers to an input parameter when the machine is working; predict the processing results generated when the manufacturing system works according to the configuration of each process parameter set, the processing results include quality characteristic values ​​generated when each machine works according to the respective process parameter combination in the process parameter set, and the quality characteristic values ​​are used to represent the values ​​of key quality characteristics after the machine processes the workpiece; select at least one target process parameter combination corresponding to each machine according to the processing results of each process parameter set; if the estimated processing result of the target process parameter combination corresponding to at least one machine meets the preset target processing result, control the manufacturing system to work according to the target process parameter combination of each machine.

[0332] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, and are not limiting. The above details do not limit the present disclosure to the necessity of adopting the above specific details to be implemented.

[0333] The block diagrams of the devices, apparatuses, equipment, and systems involved in this disclosure are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagrams. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including," "comprising," "having," and the like are open words, referring to "including but not limited to," and can be used interchangeably therewith. The words "or" and "and" used herein refer to the words "and / or," and can be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to," and can be used interchangeably therewith.

[0334] Additionally, as used herein, "or" used in a list of items beginning with "at least one" indicates a separate list, so that, for example, a list of "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not mean that the example described is preferred or better than other examples.

[0335] It should also be noted that in the system and method of the present disclosure, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present disclosure.

[0336] Various changes, substitutions, and modifications of the technology herein may be made without departing from the technology of the teachings defined by the appended claims. Furthermore, the scope of the claims of the present disclosure is not limited to the specific aspects of the above processes, machines, manufactures, compositions of events, means, methods, and actions. Currently existing or later to be developed processes, machines, manufactures, compositions of events, means, methods, or actions that perform substantially the same functions or achieve substantially the same results as the corresponding aspects herein may be utilized. Thus, the appended claims include such processes, machines, manufactures, compositions of events, means, methods, or actions within their scope.

[0337] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0338] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A manufacturing system health analysis method, characterized in that: include: Determine at least one process parameter set related to the manufacturing system, the process parameter set includes a process parameter combination corresponding to at least one machine of the manufacturing system, the process parameter combination includes at least one process parameter, and the process parameter refers to an input parameter when the machine is working; Predicting the processing results generated when the manufacturing system works according to the configuration of each process parameter set, the processing results include the quality characteristic values ​​generated when each machine works according to the respective process parameter combination in the process parameter set, and the quality characteristic values ​​are used to represent the values ​​of key quality characteristics of the workpiece after the machine processes the workpiece; According to the processing results of each process parameter set, select a target process parameter combination corresponding to the at least one machine; If the estimated processing result of the target process parameter combination corresponding to the at least one machine satisfies the preset target processing result, the manufacturing system is controlled to operate according to the target process parameter combination of each machine; wherein if the estimated processing result of the target process parameter combination corresponding to the at least one machine satisfies the preset target processing result, the manufacturing system is controlled to operate according to the target process parameter combination of each machine, including: When the manufacturing system operates according to the target process parameter combination of each machine, an overall stability index and an overall processing index of the manufacturing system are determined, wherein the overall stability index refers to an index that can represent the stability of the manufacturing system, and the overall processing index refers to an index that can represent the health status of the processing result of the manufacturing system; Calculating an estimated processing result of the manufacturing system according to an overall stability index and an overall processing index of the manufacturing system; If the estimated processing result meets the preset target processing result, controlling the manufacturing system to operate according to the target process parameter combination of each machine; The overall stability index is calculated based on the overall quality loss of the manufacturing system, the overall expected loss and the deviation ratio corresponding to the target process parameter combination of each machine, combined with the system health index quantitative model.

2. The method according to claim 1, characterized in that: The step of calculating the estimated processing result of the manufacturing system according to the overall stability index and the overall processing index of the manufacturing system comprises: Obtaining a quantitative model of the health status of the manufacturing system; The overall processing index and the overall stability index of the manufacturing system are input into the health status quantification model to obtain the estimated processing result of the manufacturing system.

3. The method according to claim 1, characterized in that When the manufacturing system operates according to the target process parameter combination of each machine, determining the overall stability index and the overall processing index of the manufacturing system includes: When the manufacturing system operates according to the target process parameter combination of each machine, obtaining an overall stability index of the manufacturing system; Based on the stability-oriented quality random flow network model, the task execution status, machine performance status and product quality status of the manufacturing system are modeled to obtain the overall processing index of the manufacturing system when it works according to the target process parameter combination of each machine.

4. The method according to claim 3, characterized in that When the manufacturing system operates according to the target process parameter combination of each machine, obtaining the overall stability index of the manufacturing system includes: Obtain the processing target, expectation and deviation ratio corresponding to each machine when it works with its target process parameter combination; Calculating the overall quality loss of the manufacturing system based on the expectations and processing goals associated with the target process parameter combinations of each machine and the quality loss function of the manufacturing system; Calculating the overall expected loss of the manufacturing system according to the overall quality loss and in combination with the expected loss function of the manufacturing system; According to the overall quality loss, the overall expected loss and the deviation ratio corresponding to the target process parameter combination of each machine, combined with the system health index quantification model, the overall stability index of the manufacturing system is calculated. The deviation ratio corresponding to the target process parameter combination is obtained by calculating the expectation and deviation of each machine in the corresponding target process parameter combination.

5. The method according to claim 3, characterized in that: The stability-oriented quality random flow network model models the task execution status, machine performance status, and product quality status of the manufacturing system, and obtains the overall processing index of the manufacturing system when working according to the target process parameter combination of each machine, including: The processing capacity of each machine is divided to obtain a plurality of discrete states corresponding to the processing capacity of each machine; Constructing probability formulas corresponding to the multiple discrete states of each machine according to the multiple discrete states corresponding to the processing capacity of each machine and the first probability parameter of each machine, wherein the first probability parameter is unknown; Get the expected downtime and failure rate for each machine; Input the expected downtime and failure rate of each machine into a preset processing state calculation formula to calculate the parameter value of the first probability parameter of each machine; Inputting the parameter value of the first probability parameter of each machine into the probability formula of each machine for multiple discrete states, respectively, to obtain the probability of each machine corresponding to the corresponding multiple discrete states; Get the minimum input flow of each machine; Based on the minimum input flow of each machine and the probabilities corresponding to the corresponding multiple discrete states, the probability that the machine processing state of each machine is not less than its minimum input flow is calculated; The overall processing index of the manufacturing system is calculated by multiplying the probability that the machine processing state of each machine is not less than its minimum input flow.

6. The method according to any one of claims 1 to 5, characterized in that: The predicting of the processing results generated when the manufacturing system operates according to the configuration of each process parameter set includes: Obtaining a trained Harris Hawk Optimization Algorithm-Back Propagation HHO-BP neural network model, wherein the HHO-BP neural network model is used to predict a processing result generated when the manufacturing system works according to input data; Each process parameter set is used as input data and input into the HHO-BP neural network model to obtain the processing results of each process parameter set.

7. The method according to claim 6, characterized in that The training steps of the HHO-BP neural network model include: Acquire at least one training sample, wherein the training sample includes training data and a true value, wherein the training data includes a process parameter set of the manufacturing system when it is actually working, and the true value is a true quality characteristic value generated by each machine when the manufacturing system is working according to the actual process parameter set; Taking the prediction result being the same as the true value corresponding to each sample as a training goal, the HHO-BP neural network model is trained using the at least one training sample to obtain the HHO-BP neural network model at the end of the training.

8. The method according to any one of claims 1 to 5, characterized in that: The step of selecting target process parameter combinations corresponding to the at least one machine according to the processing results of each process parameter set includes: Obtaining at least one quality characteristic value corresponding to each machine according to the processing results of each process parameter set; Calculate the system health index corresponding to at least one quality characteristic value of each machine; selecting a target quality characteristic value from the at least one quality characteristic value for each machine according to the system health indicator corresponding to the at least one quality characteristic value of each machine; The process parameter combination associated with the target quality characteristic value of each machine is determined as the target process parameter combination, and the target process parameter combination corresponding to at least one machine is obtained.

9. The method according to claim 8, characterized in that The calculating of the system health index corresponding to at least one quality characteristic value of each machine includes: According to the process parameter combinations of each machine in each process parameter set, combined with a preset response model, the expectations of each machine corresponding to at least one process parameter combination are calculated, wherein the response model refers to a formula for calculating the expectations of each machine in each process parameter combination; Calculate the deviation of each machine at at least one process parameter combination according to the quality characteristic value of each machine at at least one process parameter combination in combination with a preset deviation model, wherein the deviation model refers to a formula for calculating the deviation of each machine at different process parameter combinations; According to the expectations and deviations respectively associated with each machine in at least one process parameter combination, combined with a preset system health indicator quantification model, the system health indicators corresponding to each machine in at least one process parameter combination are calculated. The system health indicator quantification model refers to a calculation formula used to quantify the stability of each machine.

10. The method according to claim 9, characterized in that The method of calculating the system health index corresponding to at least one process parameter combination of each machine according to the expectations and deviations respectively associated with at least one process parameter combination of each machine in combination with a preset system health index quantification model includes: Determining a design characteristic formula and an actual characteristic formula in the system health indicator quantification model; Inputting the processing target of each machine and the expectation of each machine being respectively associated with at least one process parameter combination into the design characteristic formula to obtain the system design characteristic corresponding to each machine in each process parameter combination; Input the deviation associated with each process parameter combination of each machine into the actual characteristic formula to obtain the critical mass characteristic corresponding to each process parameter combination of each machine; Calculate the quotient of the system design characteristic and the critical mass characteristic of each machine at each process parameter combination to obtain the deviation ratio of each machine at each process parameter combination; According to the signal-to-noise ratio calculation formula, the signal-to-noise ratio corresponding to the expectation and deviation of each process parameter combination is calculated; The deviation ratio and signal-to-noise ratio of each process parameter combination are determined as the system health indicators for each process parameter combination.

11. The method according to claim 8, characterized in that The step of obtaining at least one quality characteristic value corresponding to each machine according to the processing results of each process parameter set includes: Obtain the preset quality characteristic values ​​of the workpieces required to be processed by each machine; At least one quality characteristic value satisfying a preset quality characteristic value of the machine is obtained from the quality characteristic values ​​of each machine according to the processing results of each process parameter set.

12. A manufacturing system health analysis device, characterized in that: include: a determination unit, configured to determine at least one process parameter set related to a manufacturing system, wherein the process parameter set includes a process parameter combination corresponding to at least one machine of the manufacturing system, wherein the process parameter combination includes at least one process parameter, and the process parameter refers to an input parameter when the machine is working; A prediction unit, used to predict the processing results generated when the manufacturing system works according to the configuration of each process parameter set, the processing results include the quality characteristic values ​​generated when each machine works according to the respective process parameter combination in the process parameter set, and the quality characteristic values ​​are used to represent the values ​​of key quality characteristics after the machine processes the workpiece; A selection unit, configured to select target process parameter combinations corresponding to the at least one machine according to processing results of each process parameter set; A control unit, configured to control the manufacturing system to operate according to the target process parameter combination of each machine if the estimated processing result of the target process parameter combination respectively corresponding to the at least one machine satisfies the preset target processing result; Wherein, the control unit comprises: A first determination module is used to determine an overall stability index and an overall processing index of the manufacturing system when the manufacturing system works according to the target process parameter combination of each machine, wherein the overall stability index refers to an index that can represent the stability of the manufacturing system, and the overall processing index refers to an index that can represent the health status of the processing result of the manufacturing system; A first calculation module, configured to calculate an estimated processing result of the manufacturing system according to the overall stability index and the overall processing index of the manufacturing system; A parameter control module, configured to control the manufacturing system to operate according to a target process parameter combination of each machine if the estimated processing result reaches a preset target processing result; The overall stability index is calculated based on the overall quality loss of the manufacturing system, the overall expected loss and the deviation ratio corresponding to the target process parameter combination of each machine, combined with the system health index quantitative model.

Images