Polishing continuous operation control method and control system

Abstract

The invention discloses a continuous polishing operation control method and system, and belongs to the technical field of polishing. The continuous polishing operation control method comprises the following steps that 1, a polishing system is divided into a plurality of device units according to the influence on polishing operation continuity; step 2, constructing a Markov model based on series and parallel connection relations among the device units; 3, a polishing plan of the polishing system in future preset time is obtained, and production tasks of all device units are generated; 4, based on a Markov model, the normal state probability of each device unit completing the production task is calculated, and the normal state probability of the whole polishing system completing the production task is calculated; and 5, the production speed of each polishing plan is calculated, and the optimal polishing plan is screened out based on the normal state probability and the production speed of completing the production task by the polishing system. According to the technical scheme, under the continuous operation scene, redundancy configuration of the polishing system can be accurately adjusted, maximum utilization of resources is achieved with low system construction cost, and therefore the overall production speed is remarkably increased.

Description

Technical Field

[0001] The present application relates to the technical field of polishing, and in particular, to a polishing continuous operation control method and a control system. Background Art

[0002] The polishing operation is a key step in the surface processing of workpieces, aiming to improve the surface finish and gloss of the workpieces. The current polishing process usually includes two main steps: rough polishing and fine polishing. The main purpose of rough polishing is to remove the rough parts on the surface of the workpiece and level the substrate, laying a foundation for the subsequent fine polishing; fine polishing further refines the surface to improve the smoothness and gloss. After the fine polishing is completed, the surface quality of the workpiece needs to be strictly inspected. If it meets the specified quality standards, it is considered qualified; if not, the fine polishing needs to be carried out again.

[0003] In the actual production environment, the number of rough polishing devices is usually less than that of fine polishing devices. The main reason for this configuration is that the processing speed of the rough polishing device is greater than that of the fine polishing device. Therefore, the workpieces after rough polishing will be transported to multiple fine polishing devices for subsequent processing. This layout aims to ensure that both the rough polishing and fine polishing processes can operate at full capacity, thereby improving the overall production speed.

[0004] However, this configuration has certain risks in actual operation. Since the number of rough polishing devices is small and it is in a key position in the entire production process, once the rough polishing device fails during continuous operation, the entire production process will be interrupted. This will not only affect the production speed but also may cause delays in the polishing plan and increase the production cost. In addition, the failure of the rough polishing device may also cause the fine polishing devices to be idle, further reducing the equipment utilization rate. Summary of the Invention

[0005] The content part of the present application is used to briefly introduce the concepts, which will be described in detail in the following detailed implementation part. The content part of the present application is not intended to identify the key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0006] As the first aspect of the present application, in order to solve the technical problems mentioned in the above background art part, some embodiments of the present application provide a polishing continuous operation control method, including the following steps:

[0007] Step 1: Divide the polishing system into several device units according to the impact on the continuity of the polishing operation;

[0008] Step 2: Construct a Markov model based on the series and parallel relationships between the device units;

[0009] Step 3: Obtain the polishing plan of the polishing system within a predetermined future time, and generate the production tasks for each device unit;

[0010] Step 4: Based on the Markov model, calculate the normal state probability of each device unit to complete the production task, and calculate the normal state probability of the entire polishing system to complete the production task;

[0011] Step 5: Calculate the production speed of each polishing plan, and select the best polishing plan based on the normal state probability and production speed of the polishing system to complete the production task.

[0012] In this technical solution, the polishing system is divided into several device units according to whether it affects the continuous operation characteristics. Based on the relationship between these device units, a Markov model is constructed. Through the state transition probability, the Markov model can accurately describe the dynamic evolution process of the system from one state to another. This modeling method has high accuracy in analyzing the transient behavior of the system (such as the process of the system gradually degrading from the normal state until failure), and can effectively evaluate the normal state probability of the system at different time points, rather than just focusing on the reliability performance under the steady state. Through the Markov model, effective processing of the system repair process and redundant design can be achieved. For example, after a device unit in the system fails, either repair measures can be taken to restore it to the normal operating state, or redundant components can be used to take over its function. In the continuous operation scenario, this modeling method based on the Markov model can accurately adjust the redundant configuration of the polishing system, maximize the utilization of resources at a lower system construction cost, and thus significantly improve the overall production speed.

[0013] When establishing the Markov model, the more detailed the system division, the more accurate the prediction of the stability of the entire system. However, overly detailed part division will increase the calculation amount, affect the calculation efficiency, and waste calculation resources. For this reason, this application provides the following technical solution:

[0014] Step 1 includes the following steps:

[0015] Step 11: Obtain all the part lists related to the polishing process, and remove the basic parts that will affect each step of the entire polishing process to obtain a preliminary list; the basic parts at least include various parts in the power system;

[0016] Step 12: Take the parts that need to be stopped synchronously during maintenance in the preliminary list as a device unit, and use the device unit as the smallest component unit of the polishing system.

[0017] In the technical solution of this application, when constructing the Markov model, those basic parts are excluded first. Once these basic parts fail, the entire polishing system will not be able to operate normally. Since it has a decisive impact on continuous operation, it is impossible to improve its impact on the overall system by adjusting the redundant states of other components within the system. Therefore, removing these basic parts can effectively reduce the complexity of the Markov model. At the same time, the parts that need to be stopped synchronously during the maintenance process are classified as an independent device unit. In this way, while ensuring compliance with the actual production requirements, the complexity of the model is minimized. Specifically, when any part in a device unit fails, the other parts within the device unit cannot operate normally either, so it can be determined as a damaged device unit as a whole. Through the above method, while ensuring that the prediction accuracy of the Markov model can meet the actual production requirements, the effective control of the model complexity is achieved.

[0018] Currently, for the calculation of system stability, the normal state probabilities of each device unit are used as the evaluation basis. However, the mutual correlations between device units are not considered during this process, resulting in a negative impact on the accuracy of system reliability evaluation.

[0019] Step 21: Obtain the correlations between each device unit, where the correlations include series relationship, parallel relationship, and standby relationship;

[0020] Step 22: Connect each device unit through signal flow according to the correlations to generate a Markov model.

[0021] In the technical solution provided by this application, when constructing the Markov model, the relationships between each device unit are determined according to the series relationship, parallel relationship, and standby relationship. Therefore, the finally established Markov model can accurately describe the internal connections between each device unit in the system. In this way, after establishing the Markov model, the reliability of the polishing system can be accurately evaluated, increasing the accuracy of the management of the polishing system.

[0022] When calculating the probability of whether the entire system can operate normally in the prior art, it all depends on the analysis and application of historical statistical data. However, in the actual situation of factory management, such data is often scarce and not representative. Therefore, evaluating based on limited historical data cannot accurately reflect the stability of the polishing system. This limitation makes the redundant design of the polishing system lack sufficient basis, affecting its reliability and practicality.

[0023] Furthermore, in the parallel structure of the Markov model, the probability of its normal state is 1 - P E , and the probability of its abnormal state is P E ;

[0024]

[0025] j represents the index of the device unit in the parallel structure, M represents the total number of all device units in the parallel structure, and p j represents the probability of the normal state of the j-th device unit in the parallel structure.

[0026] Furthermore, in the parallel structure of the Markov model, the probability of its normal state is 1 - P E , and the probability of the abnormal state is P E ;

[0027]

[0028] j represents the index of the device unit in the parallel structure, M represents the total number of all device units in the parallel structure, and p j represents the probability of the normal state of the j-th device unit in the parallel structure.

[0029] The technical solution provided by this application divides the entire polishing system into multiple independent device units. In this way, even when the data samples of the overall system are relatively scarce, each individual device unit still has sufficient data samples. Through this method, the occurrence probability of the normal state of a single device unit can be accurately evaluated. On this basis, combined with the normal state probability calculation formula under the series-parallel structure, the reliability of the entire system can be accurately evaluated. Finally, according to the actual requirements and system performance, the redundant design of the polishing system can be reasonably adjusted to improve the production speed.

[0030] In the prior art, a polishing plan with the highest production speed and as full utilization of production equipment as possible is usually preferred. However, this solution often ignores the importance of redundant settings, resulting in a relatively high risk of system shutdown. To address the above problems, this application provides the following technical solution:

[0031] Furthermore, step 3 includes the following specific steps:

[0032] Step 31: Obtain the production tasks within a future predetermined time, and the production tasks include all polishing tasks to be completed;

[0033] Step 32: Generate all possible polishing plans according to the redundancy of the polishing system;

[0034] Step 33: Adjust the connection information of each device unit in the Markov model for each generated polishing plan, and record the working parameters of the corresponding device unit.

[0035] The technical solution provided by this application pre-generates all possible polishing plans and incorporates each plan into the scope of comprehensive consideration. In this way, when selecting a specific polishing plan, it is possible to give priority to the option with the highest production speed and ensure that the selected plan has a high degree of redundancy, thereby reducing the risk of failure. This method can effectively improve the stability and reliability of the system while increasing the production speed.

[0036] When generating the polishing plan, if it is designed only to meet the production requirements, many polishing plans will result in some device units being idle, thus reducing the utilization rate of the device units in the entire polishing system.

[0037] Furthermore, step 32 includes the following specific steps:

[0038] Step 321: Divide the Markov model into several series units;

[0039] Step 322: Obtain the lower limit of the lowest production speed and the upper limit of the highest production speed that each series unit can achieve;

[0040] Step 323: Starting from the lowest production speed, gradually increase to the highest production speed in steps of a preset production speed interval to generate all possible polishing plans, where the production speed interval is dynamically adjusted according to the accuracy requirements of the production task.

[0041] The technical solution provided by this application divides the Markov model into several series units and finds the "lower limit of the lowest production speed and the upper limit of the highest production speed that each series unit can achieve". When setting the polishing plan, it can not only meet the load capacity of all device units but also keep all device units in working state, thereby reducing the idle rate of device units. Moreover, in practice, the production speed of some device units can be reduced as much as possible to reduce the aging time of these device units and extend the service life of the polishing system.

[0042] When setting the polishing plan, if excessive pursuit of production speed may lead to an increase in the system failure rate. In this case, the production process may be interrupted, which not only affects the continuity of production but also reduces the overall efficiency and increases the production volume of defective products. To solve this problem, this application provides the following technical solution:

[0043] In step 322, when calculating the lowest production speed of each series unit, only consider the efficiency of one device in the parallel device units within the series unit.

[0044] In each series unit, all device units can be connected in parallel to improve production speed and can also serve as backup equipment to ensure production continuity. By this method, the generated polishing plan not only takes into account the considerations of efficiency and continuity, but also optimizes overall production and product quality.

[0045] In step 33, the working parameters are the key parameters affecting the normal state of each device unit.

[0046] By calculating the normal state probability of each device unit using these parameters, the operating conditions of the entire system can be comprehensively understood.

[0047] Furthermore, step 4 includes the following steps:

[0048] Step 41: Obtain the working parameters of each device unit within a future predetermined time;

[0049] Step 42: Calculate the normal state probability of each device unit according to the working parameters;

[0050] Step 43: Calculate the normal state probability of the entire polishing system according to the normal state probability of each device unit.

[0051] The technical solution of this application estimates the working time and intensity of each device unit within a future predetermined time, thereby evaluating the continuity of the polishing operation in the next operating cycle. This method helps to comprehensively grasp the reliability of the system and ensure stable production.

[0052] Furthermore, step 5 includes the following steps:

[0053] Step 51: Calculate the time required to complete each polishing plan;

[0054] Step 52: Screen out the best polishing plan according to the time required to complete each polishing plan and the normal state probability of each polishing plan.

[0055] In the technical solution provided by this application, the finally determined polishing plan not only considers production speed but also takes into account production continuity. In this way, it can be ensured that the polishing operation is both efficient and continuous.

[0056] As the second aspect of this application: This application provides a polishing operation control system that controls the polishing plan of the polishing operation based on the polishing continuous operation control method described above.

[0057] In this technical solution, according to the impact on the continuity of the polishing operation, the polishing system is divided into several device units. Based on the relationships between these device units, a Markov model is constructed. Through the state transition probability, the Markov model can accurately describe the dynamic evolution process of the system from one state to another. This modeling method has high accuracy in analyzing the transient behavior of the system (such as the process of the system gradually degrading from the normal state until failure), and can effectively evaluate the normal state probability of the system at different time points, rather than only focusing on the reliability performance under the steady state. Through the Markov model, effective processing of the system repair process and redundant design can be achieved. For example, after a device unit in the system fails, either repair measures can be taken to restore it to the normal operating state, or redundant components can be used to take over its function. In the continuous operation scenario, this modeling method based on the Markov model can accurately adjust the redundant configuration of the polishing system, maximize the utilization of resources at a lower system construction cost, and thus significantly improve the overall production speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The accompanying drawings, which form a part of this application, are used to provide a further understanding of this application, making other features, objectives, and advantages of this application more obvious. The schematic embodiments and descriptions of the accompanying drawings of this application are used to explain this application and do not constitute an improper limitation of this application.

[0059] In addition, throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the elements and components are not necessarily drawn to scale.

[0060] In the drawings:

[0061] Figure 1 is a flowchart of the continuous operation control method for polishing.

[0062] Figure 2 is a schematic diagram of a series structure.

[0063] Figure 3 is a schematic diagram of a parallel structure.

[0064] Figure 4 is a signal flow diagram of the polishing system. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0065] Embodiments of this application will be described in more detail below with reference to the accompanying drawings. Although some embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand this application. It should be understood that the drawings and embodiments of this application are only for exemplary purposes and are not used to limit the protection scope of this application.

[0066] In addition, it should be noted that for the convenience of description, only the parts related to the relevant invention are shown in the drawings. Without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0067] The present application will be described in detail below with reference to the drawings and in combination with embodiments.

[0068] Embodiment 1:

[0069] Refer to Figure 1 , a polishing continuous operation control method, including the following steps:

[0070] Step 1: Divide the polishing system into several device units according to the impact on the continuity of the polishing operation.

[0071] Step 1 further includes the following steps:

[0072] Step 11: Obtain all the part lists related to the polishing process, and remove the basic parts that will affect each step in the entire polishing process to obtain a preliminary list.

[0073] The basic parts include various parts in the power system (such as power supplies, circuit boards, etc.). Once these parts fail, the entire system will not be able to operate. Assume that the polishing system includes the following parts: rough polishing equipment A, fine polishing equipment B, fine polishing equipment C, and the power supply module X in the power supply system. Among them, the power supply module X belongs to the basic parts. Once it fails, the entire polishing system will not be able to operate. Therefore, the stability of the power supply module X will affect the entire system. Thus, when considering the impact of the power supply module X on the polishing system, it is necessary to separate the power supply module X from it.

[0074] Step 12: Take the parts that need to be stopped synchronously during maintenance in the preliminary list as a device unit, and use the device unit as the smallest component unit of the polishing system.

[0075] For example, the polishing system includes rough polishing equipment A, fine polishing equipment B, and fine polishing equipment C. These equipments do not affect each other, so they can be divided into independent device units respectively. Specifically, the parts that make up the rough polishing equipment A can be used as an independent device unit because its stop working will not affect the normal operation of the fine polishing equipment B and the fine polishing equipment C. Similarly, the parts that make up the fine polishing equipment B can also be used as an independent device unit. Therefore, in Step 1, each finally determined device unit can operate independently, and removing any one device unit from the system will not interfere with the work of other device units.

[0076] Step 2: Generate a Markov model based on the series and parallel relationships between the device units.

[0077] Step 2 includes the following steps:

[0078] Step 21: Obtain the correlations between device units, where the correlations include series connection, parallel connection, and standby relationships;

[0079] Step 22: Connect the device units according to the correlations through signal flow to generate a Markov model.

[0080] Among them, when the device units are in a series connection relationship, they are connected in sequence according to the polishing process. When two device units are in a parallel connection or standby relationship, they are connected head-to-tail. Connecting in sequence means that the output of the previous device unit serves as the input of the next device unit. Connecting head-to-tail means that two or more device units share the same input port and one output port.

[0081] Both connecting in sequence and connecting head-to-tail are signal flow connection relationships. Signal flow refers to the working sequence of each device unit during the polishing process, that is, the sequence of the start signals of each device unit. For example, when the product passes through the rough polishing equipment, transportation equipment, fine polishing equipment, and inspection equipment in sequence, the direction of the signal flow is "rough polishing equipment → transportation equipment → fine polishing equipment → inspection equipment".

[0082] When there is a series connection relationship between device unit A and device unit B, the output of device unit A will serve as the input of device unit B. When there is a parallel connection relationship between device unit B1 and device unit B2, that is, device unit B1 and device unit B2 are two devices of the same model or devices with the same working content, the materials generated upstream will flow into device unit B1 and device unit B2 respectively for processing, and then flow out from device unit B1 and device unit B2 to the next process. When device unit B2 serves as the standby device of device unit B1, the materials upstream will flow to device unit B1 and be processed by device unit B1. When device unit B1 cannot work, device unit B2 will replace device unit B1 to work and receive the materials upstream for processing.

[0083] The essential principle of constructing the Markov model is: Based on the correlations between device units, construct a dynamic probability model (i.e., Markov chain) to describe the transition probabilities of each device unit in different states. The states include normal state and abnormal state.

[0084] As Figure 2 shown, Figure 2 is a typical Markov model with a series structure. In the series structure of the Markov model, the probability of the normal state is P R , and the probability of the abnormal state is 1 - P R . Among them, i represents the index of the device unit in the series structure, N represents the total number of all device units in the series structure, and p i represents the probability of the normal state of the i-th device unit in the series structure.

[0085] As Figure 3 shown, Figure 3 is a Markov model of a typical parallel structure. In the parallel structure of the Markov model, the probability of its normal state is 1 - P E , and the probability of the abnormal state is P E . j represents the index of the device unit in the parallel structure, M represents the total number of all device units in the parallel structure, and p j represents the probability of the normal state of the j-th device unit in the parallel structure.

[0086] After understanding the series and parallel relationships among the device units in the polishing system clearly, the Markov model can be generated accurately. Then, using the calculation formulas for series and parallel in the Markov model, the probability of the normal state of the Markov model can be calculated.

[0087] In this way, in step 2, the polishing system is constructed into a basic Markov model. In this Markov model, each device unit is connected by a signal flow. Furthermore, only by giving the probability of the normal state of each device unit within a future predetermined time, the probability of the normal state of the entire polishing system within the future predetermined time can be calculated.

[0088] Step 3: Obtain the polishing plan of the polishing system within a future predetermined time and generate the production tasks of each device unit. The "polishing plan within a future predetermined time" in this solution refers to all polishing tasks that need to be completed within a maintenance cycle or without shutting down. For example, if 30,000 workpieces need to be polished continuously within 48 hours, then the polishing plan within the future predetermined time is to complete the polishing task of 30,000 workpieces within 48 hours. According to the total time and total number of tasks of the task, the workload that needs to be completed per unit time can be calculated. For example, the number of polishing pieces that need to be completed per hour can be calculated, and the production tasks of each device unit can be arranged accordingly.

[0089] Furthermore, step 3 includes the following specific steps:

[0090] Step 31: Obtain the production tasks within a future predetermined time, and the production tasks include all polishing tasks that need to be completed;

[0091] Step 32: Generate all possible polishing plans according to the redundancy situation of the polishing system;

[0092] Step 33: Adjust the connection information of each device unit in the Markov model for each generated polishing plan, and record the working parameters of the corresponding device unit.

[0093] The polishing system is not a simple series system, but a system composed of a combination of series and parallel connections. For example, the rough polishing unit and the fine polishing unit are in series, and there are multiple parallel fine polishing devices in the fine polishing unit. In this case, all the fine polishing devices can be used as the main working devices, or one or several of the fine polishing devices can be selected as standby devices.

[0094] The redundancy situation refers to whether there are some redundant devices in each device unit (redundant devices, that is, devices exceeding the minimum number required for the normal operation of the system), which can also be understood as the number of alternative devices in each device unit. For example, multiple fine polishing devices are set, and the fine polishing devices that are redundant and can be used as alternatives are redundant devices.

[0095] Furthermore, step 32 includes the following specific steps:

[0096] Step 321: Divide the Markov model into several series units.

[0097] The "series unit" here refers to the basic product unit when calculating the normal state probability of the Markov model. Since the polishing process is roughly a linear process, there is a main series line. In the fine polishing operation link, multiple parallel fine polishing devices are added, and all the fine polishing devices can be regarded as a whole series unit.

[0098] Regarding all the fine polishing devices as a series unit can set the fine polishing devices in a parallel relationship or a standby relationship as needed, so as to generate different production plans. For example, there are a total of 3 fine polishing devices, and these 3 fine polishing devices are regarded as a series unit. When these 3 fine polishing devices are in a parallel relationship, the series unit composed of the 3 fine polishing devices has the maximum production speed, but if one fine polishing device fails, it will cause the polishing system to stop. If 2 fine polishing devices are used as standby devices for another fine polishing device, that is, the 3 fine polishing devices are in a standby relationship, then the series unit composed of the 3 fine polishing devices has the maximum normal state probability. When one of the fine polishing devices fails, the other two fine polishing devices can be used as standby devices to take over the non-working fine polishing device and continue working, ensuring the continuity of polishing.

[0099] Step 322: Obtain the lower limit of the minimum production speed and the upper limit of the maximum production speed that each series unit can achieve.

[0100] In step 322, when calculating the minimum production speed of each series unit, only the efficiency of one device unit among the device units connected in parallel within the series unit is considered. For example, if precision polishing equipment A and precision polishing equipment B are connected in parallel to form a series unit, and the minimum production speed of precision polishing equipment A is lower than that of precision polishing equipment B, then the minimum production speed of this series unit is the lower limit of the minimum production speed of precision polishing equipment A; if the maximum production speed of precision polishing equipment A is lower than that of precision polishing equipment B, then the maximum production speed of this series unit is the upper limit of the maximum production speed of precision polishing equipment A. In this way, the selected lower limit of the minimum production speed and the upper limit of the maximum production speed are the ranges within which each series unit can undertake production tasks.

[0101] Step 323: Taking the preset production speed interval as the step size, gradually increase from the minimum production speed to the maximum production speed to generate all possible polishing plans, where the production speed interval is dynamically adjusted according to the precision requirements of the production task.

[0102] The production speed interval is the minimum unit for adjusting the production speed. To avoid duplicate and excessive polishing plans, when setting the production speed interval, it is necessary to ensure that each production speed generated at this production speed interval will cause a change in the Markov model corresponding to the polishing plan.

[0103] For example, for a device unit, if its minimum production speed is 50 pieces per hour, the maximum production speed is 100 pieces per hour, and the production speed interval is 10 pieces per hour, then the following production speeds can be generated: 50 pieces per hour, 60 pieces per hour, 70 pieces per hour, 80 pieces per hour, 90 pieces per hour, and 100 pieces per hour. At each production speed level, the series-parallel structure in the Markov model corresponding to the generated polishing plan needs to change. For example, when reducing the production speed, some precision polishing equipment needs to be withdrawn from the parallel equipment and converted into standby equipment.

[0104] Step 33: Adjust the connection information of each device unit in the Markov model for each generated polishing plan, and record the working parameters of the corresponding device unit.

[0105] Since when adjusting the polishing plan, the device units connected in parallel need to be adjusted from the parallel relationship to the standby relationship, the initial Markov model needs to be adjusted. In this way, each polishing plan will correspond to a Markov model. And under different polishing plans, the production speed of each device unit is different, and the corresponding working parameters are different. Thus, under different polishing plans, the normal state probability of each device unit will change, resulting in a change in the normal state probability of the entire polishing system, aiming to find the best polishing plan.

[0106] In step 33, the working parameters are the key parameters affecting the normal state of each device unit. Since the types of each device unit are inconsistent, the key parameters are also different. The specific parameter types will be described in detail in step 4.

[0107] Step 4: Based on the Markov model, calculate the normal state probability of each device unit to complete the production task, and calculate the normal state probability of the entire polishing system to complete the production task.

[0108] Step 4 includes the following steps:

[0109] Step 41: Obtain the working parameters of each device unit within a future predetermined time;

[0110] Step 42: Calculate the normal state probability of each device unit according to the working parameters;

[0111] Step 43: Calculate the normal state probability of the entire polishing system according to the normal state probability of each device unit.

[0112] The normal state probability here refers to reliability. A high normal state probability indicates a high probability that the device unit can work normally and high reliability. The calculation methods of the normal state probabilities of different types of device units are different. In the polishing system, the device units are divided into the following three types:

[0113] The first type of device unit: display system, recognition system, etc. These systems are mainly electronic components in the polishing system. For example, the counting sensor on the conveyor belt and the display screen for showing the working status. The calculation formula is: R(t) = e -λt ; where R(t) is the reliability of the device unit within t; λ is the failure rate, obtained from the product quality specification; t is the working time. This type of device unit belongs to electronic components, and the operation of each device unit is independent of each other. And the manufacturer has conducted quality tests at the time of leaving the factory. Therefore, the failure rate can be obtained from the quality specification, and then the normal state probability at different working durations can be obtained. Suppose the failure rate of an electronic component is 0.001 per hour -1 , that is, λ = 0.001 per hour -1 , calculate its reliability within 1000 hours: R(1000) = 0.368. Therefore, the working parameter of this type of device unit is the working time. When formulating the polishing plan, determine its normal state probability (reliability) according to the next working time required by the device unit.

[0114] The second type of device unit: conveyor belt conveying equipment, rough polishing equipment, etc. The life of this type of device unit is mainly affected by the mechanical part, and the wear of the easily damaged part of this type of device unit is uniform, so the life progression is uniform, and its life distribution is symmetric and concentrated near the mean value. R(t) represents the reliability of the device unit within time t. μ is the mean value, representing the average life of the device unit. σ is the standard deviation, representing the degree of dispersion of the life. Φ is the cumulative distribution function of the standard normal distribution, and t is the time. Among them, the mean value and the standard deviation need to be obtained according to the daily data of the factory. Since the conveyor belt in the factory has a long service time, there is sufficient data to calculate the mean value and the standard deviation. Therefore, the operating parameter of this type of device unit is the operating time.

[0115] The third type of device unit: fine polishing equipment. Since the fine polishing equipment needs to have sufficient polishing accuracy during polishing, in practice, whether the fine polishing equipment can operate normally is affected by many factors, and whether it operates normally cannot be regarded as the same as ordinary mechanical equipment.

[0116] The specific calculation method of the normal state probability of the fine polishing equipment includes the following steps:

[0117] S1: Establish the state space equation of the fine polishing equipment;

[0118]

[0119] Among them, x(t) is the state vector, which contains the rotational speed information of the grinding disc. w(t) is the wear disturbance input, u(t) is the control input, A is the system matrix equation, which is used to describe the internal dynamic relationship. B1 and B2 are the input matrices corresponding to w(t) and u(t) respectively; C1 and C2 are the output matrices, D2 is the direct transfer matrix; y(t): output vector, z(t): system output vector;

[0120]

[0121] z s (t): grinding disc displacement vector, representing the displacement of the grinding disc at time t;

[0122] z u (t): abrasive grain quantity, representing the number of abrasive grains in the polishing liquid sprayed at the displacement at time t.

[0123] z r (t): grinding disturbance, representing the longitudinal pressure of the grinding disc on the surface smoothness and hardness of the workpiece to be ground;

[0124] x(t) describes the current mechanical state of the system, including the rotation of the grinding sheet, the relationship between the abrasive grains and the grinding disc, and the relationship information between the workpiece to be ground and the grinding disc;

[0125] w(t) = [z rf (t), z rr (t)] T ;

[0126] z rf (t): The influence disturbance on the upper side, which reflects the displacement caused by the uneven surface of the workpiece on the upper part of the grinding disc;

[0127] z rr (t): The influence disturbance on the lower side, which reflects the displacement caused by the uneven surface of the workpiece on the lower part of the grinding disc;

[0128] u(t) = [u f (t), u r (t)] T

[0129] u f (t): The ideal control force generated by the speed actuator, which is used to adjust the speed of the grinding disc.

[0130] u r (t): The ideal control force generated by the pressure actuator, which is used to adjust the pressure of the grinding disc.

[0131] u(t) is the control input actively applied by the system;

[0132] y(t) represents the measured output, which is fed back to the control system to help adjust the control force input u(t) to achieve the stable control of the system;

[0133] z(t) comprehensively reflects the overall behavior of the system at time t and is used for further analysis and evaluation of the system performance.

[0134] S2: Construct the transition probability matrix P;

[0135]

[0136] Where: P NN : The probability from the normal state to the normal state, P NF : The probability from the normal state to the fault state, P FN : The probability from the fault state to the normal state, P FF : The probability from the fault state to the fault state;

[0137] Among them, when constructing the transition probability matrix P, the state threshold of the state component x(t) is defined. When x(t) is not greater than the state threshold, it belongs to the normal state, otherwise it belongs to the fault state.

[0138] S3: Calculate the probability from the normal state to the fault state:

[0139] p NF = α·||w(t)|| k ;

[0140] α is a proportional constant, integrating the state space equations A, B1 Stability and state threshold x i，max 。

[0141] ||w(t)|| represents the magnitude of the external disturbance, and k is a positive exponent that reflects the degree of influence of the disturbance on the failure probability;

[0142] p FN = β·||u(t)|| m ;

[0143] Where:

[0144] β is a proportionality constant that controls the suppression performance of matrix B 2 。

[0145] ||u(t)|| represents the magnitude of the control input.

[0146] m is a positive exponent that reflects the degree of influence of the control input on the recovery probability;

[0147] Complete transition probability matrix:

[0148]

[0149] Therefore, in this solution, when calculating the normal state probability of the first type and the second type of device units, the working parameters include the working time. When calculating the third type of device units, the working parameters include: working time, abrasive concentration, polishing disc rotation speed, and polishing disc pressure.

[0150] Step 43: Calculate the normal state probability of the entire polishing system based on the normal state probability of each device unit.

[0151] In this way, according to the technical solution provided in step 42, the normal state probability of each device unit under different polishing plans can be calculated. In step 33, the connection information of each device unit in the Markov model is adjusted for each generated polishing plan, and for each polishing plan, a corresponding Markov model is generated. According to the solution provided in step 22, the normal state probability of the entire polishing system can be calculated. After obtaining the normal state probability data of the entire polishing system, it is possible to determine whether the entire polishing system will stop when executing the next polishing plan. If the polishing system will stop, the polishing system is repaired; if it does not stop, the polishing plan is executed. Thus, the adverse effects on production continuity and polishing plans are avoided, the material loss caused by production line downtime is reduced, and the product defect rate increased due to the reduced stability of each device before production line downtime is lowered.

[0152] Step 5: Calculate the production speed of each polishing plan, and select the best polishing plan based on the normal state probability and production speed of the polishing system to complete the production task.

[0153] Step 5 includes the following steps:

[0154] Step 51: Calculate the time required for each polishing plan to be completed;

[0155] Step 52: Screen out the best polishing plan according to the time required for each polishing plan to be completed and the normal state probability of each polishing plan.

[0156] In this way, for each polishing plan, each device unit will have a new production task, and for this new production task, each device unit will have a corresponding normal state probability. In this way, the normal state probability of the entire polishing system under each polishing plan can be calculated. If the normal state probability is high, it means that the continuity of the polishing plan is strong and the probability of sudden shutdown is low.

[0157] In practice, the best polishing plan can be screened out by adjusting the time required for the polishing plan to be completed and the weights of the normal state probabilities of each polishing plan.

[0158] For example: Set the weight score Q, Q = 1 / T + P 0 , where T is the time required for the polishing plan to be completed, and P 0 is the normal state probability of the polishing plan. The smaller T is and the larger P 0 is, the larger the weight score is. The polishing plan with the largest weight score is taken as the best polishing plan.

[0159] In practice, a minimum threshold for the weight score Q can also be set. If the weight score Q is less than the preset threshold, it means that the entire polishing system is likely to malfunction. To avoid waste of raw materials and increase in defective products caused by equipment failures, equipment maintenance can be arranged first. After the equipment maintenance, the working parameters of each device unit will change accordingly, the normal state probability will increase, and thus the weight score Q will also increase.

[0160] Embodiment 2:

[0161] Embodiment 2 provides a flowchart of the polishing system in actual production. The polishing system includes an input module, a control module, a rough polishing module, a conveying module, 5 fine polishing modules, and a detection module, and its signal flowchart is as Figure 4 shown.

[0162] Composition and functions of the polishing system:

[0163] The input module is used to receive production task instructions. It belongs to electronic components and calculates the normal state probability according to the calculation method of the first type of device unit.

[0164] The control module is responsible for receiving the information forwarded by the input module and passing the instructions to other modules. The control module belongs to communication electronic products, and it calculates the normal state probability by using the calculation method of the first type of device unit.

[0165] The rough polishing module is used for the preliminary polishing of workpieces. It belongs to the mechanical components in the polishing system. Due to the low precision of the internal mechanical equipment, its life distribution is symmetric and concentrated near the mean value. The normal state probability is calculated according to the calculation method of the second type of device unit.

[0166] The conveying module is responsible for transporting the workpieces from the rough polishing module to the fine polishing module. The working wear of this module is uniform and the life progression is uniform. The normal state probability can be calculated according to the calculation method of the second type of device unit.

[0167] The fine polishing module is used for the finishing of workpieces. In actual production, the system adopts a redundant design: the 5 fine polishing modules are divided into 3 main devices and 2 standby devices. In step 32, different operation strategies can be selected according to the requirements of the production task.

[0168] The fine polishing module calculates the normal state probability according to the calculation method of the third type of device unit.

[0169] The detection module is used for the quality inspection of the polished workpieces. It belongs to electronic components and calculates the normal state probability according to the calculation method of the first type of device unit.

[0170] The signal flow chart of the polishing system (as Figure 4 shown):

[0171] Input module → Control module:

[0172] The production task instructions are transmitted from the input module to the control module.

[0173] Control module → Rough polishing module:

[0174] The control module passes the instructions to the rough polishing module to start the preliminary polishing process.

[0175] Rough polishing module → Conveying module:

[0176] After the workpieces are rough polished, they are transported to the fine polishing module through the conveying module.

[0177] Fine polishing module → Detection module:

[0178] After the workpieces are finely processed by 5 fine polishing modules, they enter the detection module for quality inspection.

[0179] Implementation steps:

[0180] Step 1: Divide the series-parallel units;

[0181] Divide the polishing system into multiple series-parallel units;

[0182] The rough polishing module is used as a series unit. The fine polishing module is used as a parallel unit (including 3 main devices and 2 standby devices (omitted in the figure)). The detection module is used as an independent series unit.

[0183] Step 2: Based on the series and parallel relationships between the device units, construct a Markov model..

[0184] Step 3: According to the strategy in Step 32, generate a polishing plan:

[0185] Specifically, it includes the following two polishing plans:

[0186] Polishing Plan 1: Enable all 5 fine polishing modules.

[0187] This has a high production speed, but the simultaneous operation of redundant devices may lead to an increase in the failure rate.

[0188] Polishing Plan 2: Only enable 3 main devices and put 2 standby devices on standby.

[0189] This has a slightly lower production speed, but the overall reliability of the system is higher.

[0190] For each strategy, the Markov model needs to be adjusted to obtain Markov models of two structures.

[0191] Step 4: Based on the Markov model, calculate the normal state probability of each device unit to complete the production task, and calculate the normal state probability of the entire polishing system to complete the production task.

[0192] Calculate the working parameters of each device unit (Step 41), calculate the normal state probability of each device unit according to the working parameters (Step 42), and calculate the normal state probability of the entire polishing system.

[0193] In this solution, calculate the normal state probabilities of the input module, control module, rough polishing module, conveying module, fine polishing module, and detection module respectively, and then multiply the normal state probabilities of the input module, control module, rough polishing module, conveying module, fine polishing module, and detection module to obtain the normal state probability of the polishing system.

[0194] Step 5: According to the time required to complete each polishing plan and the normal state probability, assign corresponding weights, and select the polishing plan with the optimal comprehensive performance. For example, set the weight score Q, Q = 1 / T + P 0 , where T is the time required to complete the polishing plan, P 0is the normal state probability of the polishing plan. The larger Q is, the better the comprehensive performance is. Moreover, two weight coefficients α and β can be respectively set for (1 / T) and (P 0 ) to control the screening ability of the two factors for the polishing plan.

[0195] Embodiment 3:

[0196] A polishing operation control system controls the polishing plan of the polishing operation based on the polishing continuous operation control method described above.

[0197] The above description is only some preferred embodiments of the present application and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present application is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the technical solutions formed by mutually replacing the above features with the technical features (but not limited to) having similar functions disclosed in the embodiments of the present application.

Claims

1. A method for controlling a continuous polishing operation, characterized in that: The steps include: Step 1: Divide the polishing system into several device units according to the impact on the continuity of the polishing operation; Step 2: Construct a Markov model based on the series and parallel relationship between each device unit; Step 3: Obtain the polishing plan of the polishing system within a predetermined time in the future and generate the production tasks of each device unit; Step 4: Based on the Markov model, the normal state probability of each device unit completing the production task is calculated, and the normal state probability of the entire polishing system completing the production task is calculated; Step 5: Calculate the production speed of each polishing plan, and select the best polishing plan based on the normal state probability and production speed of the polishing system completing the production task.

2. The polishing continuous operation control method according to claim 1, characterized in that: Step 1 includes the following steps: Step 11: Obtain a list of all parts related to the polishing process and remove the basic parts that will affect each step in the entire polishing process to obtain a preliminary list; The basic parts at least include various parts in the power system; Step 12: Treat the parts in the preliminary list that need to be stopped synchronously during maintenance as a device unit, and treat the device unit as the smallest component unit of the polishing system.

3. The polishing continuous operation control method according to claim 1, characterized in that: Step 2 includes the following steps: Step 21: Obtain the correlation between each device unit, the correlation including series relationship, parallel relationship and standby relationship; Step 22: Based on the correlation, each device unit is connected through the signal flow to generate a Markov model.

4. The polishing continuous operation control method according to claim 3, characterized in that: In the serial structure of the Markov model, the probability of its normal state is P R , the probability of abnormal state is 1-P R ; Where i represents the index of the device in the series structure, N represents the total number of all devices in the series structure, and p i Represents the probability of the normal state of the i-th device in the series structure.

5. The polishing continuous operation control method according to claim 3, characterized in that: In the parallel structure of the Markov model, the probability of its normal state is 1-P E , the probability of abnormal state is P E ; j represents the index of the device unit in the parallel structure, M represents the total number of all device units in the parallel structure, and p j Represents the probability of the normal state of the jth device unit in the parallel structure.

6. The polishing continuous operation control method according to claim 1, characterized in that: Step 3 includes the following specific steps: Step 31: Obtain production tasks within a predetermined time in the future, wherein the production tasks include all polishing tasks that need to be completed; Step 32: Generate all possible polishing plans according to the redundancy of the polishing system; Step 33: Adjust the connection information of each device unit in the Markov model for each generated polishing plan, and record the working parameters of the corresponding device unit.

7. The polishing continuous operation control method according to claim 5, characterized in that: Step 32 includes the following specific steps: Step 321: Divide the Markov model into a number of series units; Step 322: Obtain the minimum production speed lower limit and the maximum production speed upper limit that can be achieved by each series unit; Step 323: Using the preset production speed interval as the step length, gradually increase from the lowest production speed to the highest production speed, and generate all possible polishing plans, wherein the production speed interval is dynamically adjusted according to the accuracy requirements of the production task.

8. The polishing continuous operation control method according to claim 7, characterized in that: Step 4 includes the following steps: Step 41: Obtaining the operating parameters of each device unit within a predetermined time in the future; Step 42: Calculate the normal state probability of each device unit according to the operating parameters; Step 43: Calculate the normal state probability of the entire polishing system according to the normal state probability of each device unit.

9. The polishing continuous operation control method according to claim 8, characterized in that: Step 5 includes the following steps: Step 51: Calculate the time required to complete each polishing plan; Step 52: Filter out the best polishing plan according to the time required for each polishing plan to complete and the normal state probability of each polishing plan.

10. A polishing operation control system, characterized in that: A polishing plan for a polishing operation is controlled based on the polishing continuous operation control method according to any one of claims 1 to 9.

Images