Overall equipment efficiency improving method and device applied to generic semiconductor field

Abstract

The invention discloses an overall equipment efficiency improving method and device applied to the field of generic semiconductors, and belongs to the technical field of computers. The method comprises the following steps: acquiring an operation parameter of target equipment at the current moment and an overall equipment efficiency parameter; obtaining a parameter optimization scheme based on a K-nearest neighbor model, the operation parameters at the current moment and the overall equipment efficiency parameters, and predicting the remaining service life of the target equipment based on the operation parameters at the current moment and the overall equipment efficiency parameters to obtain maintenance suggestions; the K neighbor model is constructed based on the operation parameters of the target equipment at multiple historical moments in a stable operation state and the overall equipment efficiency parameters; and visually displaying the parameter optimization scheme and the maintenance suggestion. According to the overall equipment efficiency improvement method applied to the generic semiconductor field, the problem that the production and manufacturing efficiency of production equipment is not ideal due to a parameter optimization scheme for adjusting the overall equipment efficiency determined in a manual mode can be solved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of computer, and particularly relates to a whole device efficiency improving method and device applied to the field of general semiconductor. BACKGROUND

[0002] In the field of general semiconductor, such as integrated circuit, solar cell and semiconductor device material industry, the whole device efficiency of production equipment is generally adjusted to improve the production and manufacturing efficiency of the production equipment.

[0003] However, generally, the improvement of the whole device efficiency of the production equipment depends on simple statistics, formulas or rules, and the optimization scheme for adjusting the whole device efficiency of the production equipment is obtained based on the result calculated by the artificial statistics or formula. In the complex industrial production environment or the industrial production environment including new type of equipment, the optimization scheme determined by the artificial experience is difficult to guarantee the improvement effect of the whole device efficiency, resulting in the unsatisfactory production and manufacturing efficiency of the production equipment. SUMMARY

[0004] The present application aims to at least solve one of the technical problems in the prior art. To this end, the present application provides a whole device efficiency improving method and device applied to the field of general semiconductor to solve the problem that the optimization scheme for adjusting the whole device efficiency determined in an artificial manner results in the unsatisfactory production and manufacturing efficiency of the production equipment.

[0005] In a first aspect, the present application provides a whole device efficiency improving method applied to the field of general semiconductor, which comprises:

[0006] obtaining the running parameter and the whole device efficiency parameter of the target device at the current time;

[0007] obtaining the parameter optimization scheme based on the K-neighbor model, the running parameter at the current time and the whole device efficiency parameter, and predicting the remaining service life of the target device based on the running parameter at the current time and the whole device efficiency parameter to obtain the maintenance suggestion; the K-neighbor model is constructed based on the historical values of the running parameter and the whole device efficiency parameter at multiple historical times under the stable running state of the target device;

[0008] visually displaying the parameter optimization scheme and the maintenance suggestion.

[0009] According to the method for improving overall device efficiency applied to the field of general-purpose semiconductors, the parameter optimization scheme is obtained based on the running parameters and overall device efficiency parameters of the target device at the current time and the K-nearest neighbor model, and the remaining service life of the target device is predicted based on the running parameters and overall device efficiency parameters at the current time, so as to obtain the maintenance suggestion. In this way, the parameter optimization scheme and the maintenance suggestion obtained can ensure the production efficiency of the target device to be improved.

[0010] According to an embodiment of the present application, after the parameter optimization scheme and the maintenance suggestion are visually displayed, the method further comprises:

[0011] Based on the large language model, the knowledge graph in the field of general-purpose semiconductors is searched according to the running parameters at the current time, the parameter optimization scheme and the maintenance suggestion, and the target maintenance scheme is output.

[0012] According to an embodiment of the present application, the running parameters include controllable parameters and uncontrollable parameters; the parameter optimization scheme is obtained based on the K-nearest neighbor model, the running parameters at the current time and the overall device efficiency parameters, and comprises:

[0013] Based on the uncontrollable parameters at the current time, K first target points in the index cluster of the K-nearest neighbor model are determined; each point in the index cluster corresponds to the running parameters and the overall device efficiency parameters at a historical time; K is a positive integer greater than 2;

[0014] Based on the controllable parameters and the overall device efficiency parameters at the historical time corresponding to the K first target points, the parameter optimization scheme is obtained; the parameter optimization scheme includes the target value of the controllable parameters.

[0015] According to an embodiment of the present application, the parameter optimization scheme further includes the predicted value of the overall device efficiency and the expected improvement rate of the overall device efficiency; the parameter optimization scheme is obtained based on the controllable parameters and the overall device efficiency parameters at the historical time corresponding to the K first target points, and comprises:

[0016] Based on the controllable parameters and the overall device efficiency parameters at the historical time corresponding to the K first target points, the target value of the controllable parameters is obtained;

[0017] Based on the target value of the controllable parameters, the predicted value of the overall device efficiency is obtained;

[0018] Based on the overall device efficiency at the current time and the predicted value of the overall device efficiency, the expected improvement rate of the overall device efficiency is obtained.

[0019] According to an embodiment of the present application, the target value of the controllable parameters is obtained based on the controllable parameters and the overall device efficiency parameters at the historical time corresponding to the K first target points, and comprises:

[0020] obtaining a first weight of each of the first target points based on the quality performance parameters corresponding to the historical time points of the K first target points and the first distances, wherein the first distance is a distance between a point corresponding to the current time running parameter and the overall device efficiency parameter in the index cluster and each of the first target points, and wherein the quality performance parameter is obtained based on the overall device efficiency parameter;

[0021] obtaining a target value of the controllable parameter based on the first weights of the K first target points and the controllable parameters corresponding to the historical time points of the K first target points.

[0022] According to an embodiment of the present application, after obtaining the predicted value of the overall device efficiency based on the target value of the controllable parameter, the method further comprises:

[0023] obtaining an actual value of the overall device efficiency after the target device is controlled with the target value of the controllable parameter as a control target;

[0024] updating the K nearest neighbor model based on the first difference, wherein the first difference is a difference between the actual value of the overall device efficiency and the predicted value of the overall device efficiency.

[0025] According to an embodiment of the present application, updating the K nearest neighbor model based on the first difference comprises:

[0026] in a case where the first difference is less than or equal to a threshold value, updating the quality performance parameters corresponding to the K first target points.

[0027] According to an embodiment of the present application, updating the K nearest neighbor model based on the first difference further comprises:

[0028] in a case where the first difference is greater than the threshold value, deleting M second target points in the index cluster, wherein the second target point is a point with a distance less than or equal to a preset value between the point corresponding to the controllable parameter at the current time.

[0029] According to an embodiment of the present application, updating the quality performance parameter comprises:

[0030] for each of the K first target points, obtaining a second weight based on the second distance and the smoothing factor coefficient, wherein the second distance is a distance between each of the first target points and the remaining (K-1) first target points;

[0031] obtaining an updated quality performance parameter based on the second weights of the K first target points and the quality performance parameters corresponding to each of the first target points.

[0032] In a second aspect, the present application provides a device for improving overall device efficiency applied to the field of general semiconductors, which comprises:

[0033] The first obtaining module is configured to obtain the running parameter of the target device at the current time and the overall device efficiency parameter;

[0034] The second obtaining module is configured to obtain the parameter optimization scheme based on the K-neighbor model, the running parameter at the current time and the overall device efficiency parameter, and predict the remaining service life of the target device based on the running parameter at the current time and the overall device efficiency parameter, and obtain the maintenance suggestion; the K-neighbor model is constructed based on the running parameter at a plurality of historical times and the overall device efficiency parameter in the stable running state of the target device;

[0035] The display module is configured to visually display the parameter optimization scheme and the maintenance suggestion.

[0036] According to the overall device efficiency improvement device applied to the field of general semiconductors provided by the present application, the parameter optimization scheme is obtained based on the running parameter at the current time of the target device and the overall device efficiency parameter and the K-neighbor model, and the remaining service life of the target device is predicted based on the running parameter at the current time and the overall device efficiency parameter, so as to obtain the maintenance suggestion, thereby replacing the optimization scheme determined by artificial experience, so that the obtained parameter optimization scheme and maintenance suggestion can guarantee the production efficiency improvement effect of the target device.

[0037] Additional aspects and advantages of the present application will be in part apparent and in part pointed out hereinafter. BRIEF DESCRIPTION OF DRAWINGS

[0038] The above and / or additional aspects and advantages of the present application will become apparent and be readily appreciated from the following description, including the appended drawings, wherein:

[0039] Figure 1 is one of the flowcharts of the overall device efficiency improvement method applied to the field of general semiconductors provided by the embodiments of the present application;

[0040] Figure 2 is another flowchart of the overall device efficiency improvement method applied to the field of general semiconductors provided by the embodiments of the present application;

[0041] Figure 3 is a method schematic diagram for obtaining K first target points based on an index cluster provided by the embodiments of the present application;

[0042] Figure 4 is a visual interface schematic diagram provided by the embodiments of the present application;

[0043] Figure 5 is a structural schematic diagram of the overall device efficiency improvement device applied to the field of general semiconductors provided by the embodiments of the present application;

[0044] Figure 6 FIG. 1 is a structural schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0045] The technical solutions in the embodiments of the present application will be clearly described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all of them. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art are within the scope of protection of the present application.

[0046] The terms "first", "second", and the like in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. are generally of a kind and do not limit the number of objects, for example, the first object can be one or more. In addition, "and / or" in the specification and claims means at least one of the connected objects, and the character " / ", generally represents a "or" relationship between the front and rear associated objects.

[0047] The application of the overall device efficiency improvement method applied to the field of general semiconductors, the overall device efficiency improvement device applied to the field of general semiconductors, the electronic device and the readable storage medium provided by the embodiments of the present application will be described in detail below in combination with the drawings and through specific embodiments and their application scenarios.

[0048] The overall device efficiency improvement method applied to the field of general semiconductors can be applied to a terminal, and can be executed by hardware or software in the terminal.

[0049] The terminal includes, but is not limited to, a portable communication device such as a mobile phone or a tablet computer having a touch-sensitive surface (e.g., a touchscreen display and / or a touchpad). It should also be understood that in some embodiments, the terminal can not be a portable communication device, but a desktop computer having a touch-sensitive surface (e.g., a touchscreen display and / or a touchpad).

[0050] In each of the following embodiments, a terminal including a display and a touch-sensitive surface is described. However, it should be understood that the terminal can include one or more other physical user interface devices such as a physical keyboard, a mouse, and a joystick.

[0051] The embodiment of the present application provides a method for improving overall equipment efficiency applied to the pan-semiconductor field. The execution subject of the method for improving overall equipment efficiency applied to the pan-semiconductor field can be an electronic device or a functional module or functional entity in the electronic device that can implement the method for improving overall equipment efficiency applied to the pan-semiconductor field. The electronic devices mentioned in the embodiment of the present application include but are not limited to mobile phones, tablets, computers, cameras and wearable devices, etc. The overall equipment efficiency improvement method applied to the pan-semiconductor field provided by the embodiment of the present application is explained below using electronic devices as the execution subject as an example.

[0052] In related technologies, improving the overall equipment efficiency (OEE) of production equipment relies on simple statistics, formulas, or rules. Based on the results calculated using fixed statistical methods or formulas, an optimization plan for adjusting the OEE of production equipment is derived. However, in complex industrial production environments, or those involving new types of equipment, fixed empirical formulas cannot fully account for the combined impact of multiple factors such as equipment utilization, performance, and quality. Consequently, the resulting optimization plan struggles to guarantee an improvement in OEE, resulting in suboptimal production efficiency for production equipment.

[0053] like Figure 1 As shown, the overall device efficiency improvement method applied to the pan-semiconductor field includes: step 110, step 120 and step 130.

[0054] Step 110: Obtain the current operating parameters and overall equipment efficiency parameters of the target equipment.

[0055] In actual implementation, the operating parameters of the target device may include the target device's operating time, downtime, total number of produced products, number of defective products, etc. The operating parameters of the target device can be collected in real time through sensors and IoT devices.

[0056] In some embodiments, after obtaining the operating parameters of the target device, the operating parameters of the target device can be detected and corrected for abnormal values ​​through statistical methods, machine learning models, or any other theoretically feasible methods, thereby removing abnormal values ​​and noise data in the operating parameters of the target device in turn.

[0057] In some embodiments, after obtaining the operating parameters of the target device, the collected operating parameters of the target device can be stored in a target database for subsequent analysis and processing. The target database can be a relational database, a NoSQL database, or any theoretically feasible database, and this application does not specifically limit this. In addition, data synchronization technology can be used, and data synchronization tools, such as ETL tools or data middleware, can also be used to ensure data consistency during the process of improving the overall device efficiency of the target device.

[0058] In actual implementation, the overall equipment efficiency parameter of the target device can include at least one of an availability parameter, a quality rate parameter or a performance parameter of the target device. In some embodiments, the overall equipment efficiency of the target device at different time points can be obtained based on the overall equipment efficiency parameter of the target device at the different time points.

[0059] In step 120, a parameter optimization scheme is obtained based on the K-neighbor model, the running parameter at the current time point and the overall equipment efficiency parameter, and a remaining service life of the target device is predicted based on the running parameter at the current time point and the overall equipment efficiency parameter to obtain a maintenance suggestion. The K-neighbor model is constructed based on the running parameter at a plurality of historical time points and the overall equipment efficiency parameter of the target device in a stable running state.

[0060] In actual implementation, the K-neighbor model can be constructed based on the running parameter at the historical time points and the overall equipment efficiency parameter of the target device in the stable running state. The K-neighbor model can include an index cluster, and the index cluster can include a plurality of points, each of which can be a multi-dimensional coordinate point or a one-dimensional coordinate point. Each point in the index cluster can be mapped from the running parameter at a historical time point and the overall equipment efficiency parameter.

[0061] In some embodiments, a target point similar to the running parameter at the current time point and the overall equipment efficiency parameter can be determined from the index cluster of the K-neighbor model based on the running parameter at the current time point and the overall equipment efficiency parameter of the target device. The running parameter can include a controllable parameter, and the controllable parameter corresponding to the target point at a historical time point and the overall equipment efficiency parameter can be obtained based on the mapping relationship between each point in the index cluster and the controllable parameter at a historical time point and the overall equipment efficiency parameter, and the parameter optimization scheme can be obtained based on the controllable parameter at the historical time point and the overall equipment efficiency parameter.

[0062] In some embodiments, after the parameter optimization scheme is obtained and the target device is controlled based on the parameter optimization scheme, an actual value of the overall equipment efficiency of the target device can be obtained, a difference between an expected value of the overall equipment efficiency of the target device before the parameter optimization scheme is executed and an actual value of the overall equipment efficiency after the parameter optimization scheme is executed is calculated, and the K-neighbor model is updated based on the calculation result of the difference.

[0063] In actual execution, the remaining useful life of the target device can be predicted based on a pre-trained remaining useful life (RUL) model, and maintenance recommendations can be obtained to reduce downtime caused by post-maintenance and regular maintenance through predictive maintenance, thereby improving the availability of the target device and further improving the overall device efficiency of the target device. The RUL model can predict the remaining useful life of the target device based on the operating parameters and overall device efficiency parameters of the target device at the current time.

[0064] In some embodiments, the RUL prediction model based on the baseline Transformer model and unsupervised transfer learning method can accurately predict the remaining useful life of the target device, and automatically generate maintenance recommendations when the RUL prediction value is lower than the threshold.

[0065] Step 130, visualizing the parameter optimization scheme and maintenance recommendations.

[0066] In actual execution, the parameter optimization scheme and maintenance recommendations can be visualized and displayed through a visualization device. The visualization device can be a visualization device installed on the target device side, or a visualization device installed on the control device side that controls the target device.

[0067] In some embodiments, after visualizing the parameter optimization scheme and maintenance recommendations, a target maintenance scheme for the target device can be obtained based on a pre-trained large language model.

[0068] In some examples, the operating parameters and overall device efficiency parameters of the target device, the parameter optimization scheme, and the maintenance recommendations can be dynamically displayed through an html front-end technology integrated with chart.js, helping operation and maintenance personnel to more intuitively and efficiently manage and maintain the target device, and further improving the overall device efficiency of the target device.

[0069] The overall device efficiency improvement method applied to the field of general semiconductors according to the embodiments of the present application obtains a parameter optimization scheme based on the operating parameters and overall device efficiency parameters of the target device at the current time and a K-nearest neighbor model, and obtains maintenance recommendations by predicting the remaining useful life of the target device based on the operating parameters and overall device efficiency parameters at the current time. Instead of the way of determining the optimization scheme by artificial experience, the obtained parameter optimization scheme and maintenance recommendations can ensure the production efficiency improvement effect of the target device.

[0070] In some embodiments, a large language model can be used to search a knowledge graph in the field of general semiconductors based on the operating parameters, parameter optimization scheme, and maintenance recommendations at the current time, and output a target maintenance scheme.

[0071] In actual implementation, the knowledge graph is a complex information structure for storing and representing the relationships between entities, such as people, places, and things, and various semantic connections between them. The knowledge graph can be composed of SPO triples. SPO triples are subject, relationship, and object. Among them, the subject and the object are both affiliated to the entity objects in the knowledge graph. The SPO triple structure can support various applications, such as search engines, recommendation systems, and artificial intelligence applications, and organize information through graphical structures for efficient information retrieval and analysis.

[0072] In some embodiments, based on a large language model, the enhanced knowledge graph in the field of general semiconductors can be retrieved based on the running parameters, parameter optimization schemes and maintenance suggestions at the current time to obtain retrieval results; the running parameters, parameter optimization schemes and maintenance suggestions at the current time are used to obtain processing suggestions for faults in the field of general semiconductors; the enhanced knowledge graph in the field of general semiconductors is obtained by associating the target vectors obtained by vectorizing the names of target objects and the description texts of target objects in the knowledge graph in the field of general semiconductors based on the fine-tuned BERT model; the target objects include entities and relationships; and the description texts of the target objects are extracted from target online documents. Based on the retrieval results, a target maintenance scheme is output.

[0073] In actual implementation, the BERT (Bidirectional Encoder Representations from Transformers) model is a pre-trained deep learning model commonly used in natural language processing (NLP). The core technology of the BERT model is the Transformer. The Transformer is an attention mechanism that can process all parts of the input data simultaneously. The BERT model learns the context relationship on a large amount of text through pre-training, and can then be fine-tuned to perform various NLP tasks. The above NLP tasks can be text classification, question and answer systems, or language reasoning, etc. In the embodiments of the present application, the fine-tuned BERT can be used as an encoder to encode natural language texts (i.e., names of target objects and description texts of target objects) into corresponding embedding vectors (target vectors corresponding to target objects).

[0074] According to the method for improving the overall equipment efficiency in the field of general semiconductors, by retrieving the knowledge graph in the field of general semiconductors based on a large language model according to the running parameters, parameter optimization schemes and maintenance suggestions at the current time, and outputting a target maintenance scheme, the operation and maintenance personnel can maintain the target equipment based on the target maintenance scheme, thereby ensuring the production efficiency of the target equipment.

[0075] In some embodiments, the operation parameters include controllable parameters and uncontrollable parameters; the K first target points in the index cluster of the K-neighbor model can be determined based on the uncontrollable parameters at the current time; each point in the index cluster corresponds to operation parameters at a historical time and the overall device efficiency parameter; K is a positive integer greater than 2; the parameter optimization scheme is obtained based on the controllable parameters and the overall device efficiency parameter at the historical time corresponding to the K first target points; and the parameter optimization scheme includes target values of the controllable parameters.

[0076] In actual implementation, the K-neighbor model can be constructed based on the uncontrollable parameters and the overall device efficiency parameter at the historical time of the target device in each stable operation state. The K-neighbor model can include an index cluster, and the index cluster can include a plurality of points, each of which can be a multi-dimensional coordinate point. Each point in the index cluster can have a mapping relationship with the uncontrollable parameters and the overall device efficiency parameter at a historical time. In actual implementation, the uncontrollable parameters and the overall device efficiency parameter at each historical time of the target device in a stable operation state can be obtained, and the uncontrollable parameters and the overall device efficiency parameter at each historical time can be processed by extreme sample exclusion, normalization, etc. Then, based on quantization technology, each uncontrollable parameter at a historical time in the processed uncontrollable parameters at each historical time is expressed as a multi-dimensional coordinate point, and a corresponding mapping relationship with the overall device efficiency parameter at the historical time is established.

[0077] In some embodiments, a corresponding multi-dimensional coordinate point can be generated for each historical time uncontrollable parameter, and a quantized inverted index can be established, for example, the open source Faiss framework can be used to implement the quantized inverted index. In some embodiments, based on the uncontrollable parameters at a historical time, the coordinate value of a point in the index cluster can be obtained, and a corresponding relationship between the coordinate point and the overall device efficiency parameter at the historical time can be established.

[0078] In some embodiments, the quality performance parameter can be obtained based on the overall device efficiency parameter. After obtaining a plurality of multi-dimensional coordinate points quantized from the uncontrollable parameters at each historical time of the target device, the index cluster of the K-neighbor model can be obtained according to the quality performance parameters corresponding to each of the plurality of multi-dimensional coordinate points.

[0079] In some embodiments, after obtaining a plurality of multi-dimensional coordinate points quantized from the uncontrollable parameters at each historical time of the target device, the multi-dimensional coordinate points corresponding to the quality performance parameters greater than or equal to a preset value can be inserted into the index cluster to obtain the index cluster of the K-neighbor model.

[0080] In actual execution, the uncontrollable parameter of the target device at the current time can be quantified as a multi-dimensional coordinate point, the K first target points can be the K coordinate points closest to the coordinate point quantified by the uncontrollable parameter of the target device at the current time in distance in the index cluster, and K is a positive integer greater than 2.

[0081] In some embodiments, after obtaining the K first target points from the index cluster, the parameter optimization scheme can be obtained based on the controllable parameters and the overall device efficiency parameters of the historical time corresponding to the K first target points; the parameter optimization scheme can include target values of controllable parameters, or can include predicted values of overall device efficiency and expected improvement rates of overall device efficiency.

[0082] According to the overall device efficiency improvement method applied to the field of general semiconductors, the K first target points in the index cluster of the K nearest neighbor model are determined based on the uncontrollable parameter at the current time, the parameter optimization scheme is obtained based on the controllable parameters and the overall device efficiency parameters of the historical time corresponding to the K first target points, and the way of obtaining the optimization scheme determined by artificial experience is replaced, so that the parameter optimization scheme and the maintenance suggestion obtained can guarantee the production efficiency improvement effect of the target device.

[0083] In some embodiments, the parameter optimization scheme can include target values of controllable parameters, predicted values of overall device efficiency, and expected improvement rates of overall device efficiency; the target values of controllable parameters are obtained based on the controllable parameters and the overall device efficiency parameters of the historical time corresponding to the K first target points; the predicted values of overall device efficiency are obtained based on the target values of controllable parameters; and the expected improvement rates of overall device efficiency are obtained based on the overall device efficiency at the current time and the predicted values of overall device efficiency.

[0084] In actual execution, the target values of controllable parameters can be obtained based on the controllable parameters and the overall device efficiency parameters of the historical time corresponding to the K first target points, and the target values of controllable parameters can be data used for adjusting the operation of the target device by the maintenance personnel of the target device. After obtaining the target values of controllable parameters, the predicted values of overall device efficiency can be obtained based on the target values of controllable parameters. The predicted values of overall device efficiency can be the overall device efficiency of the target device that is theoretically expected to be achieved under the condition that the target device is adjusted according to the target values of controllable parameters.

[0085] In some embodiments, the overall equipment efficiency of the target device at the current time can be obtained in real time, and based on the overall equipment efficiency of the target device at the current time and the predicted value of the overall equipment efficiency, an expected improvement rate of the overall equipment efficiency can be obtained, for example, by calculating the difference between the overall equipment efficiency of the target device at the current time and the predicted value of the overall equipment efficiency, and then calculating the ratio between the difference result and the overall equipment efficiency of the target device at the current time, and taking the calculation result as the expected improvement rate of the overall equipment efficiency.

[0086] According to the overall equipment efficiency improvement method applied to the field of general semiconductors, the target value of the controllable parameter is obtained based on the controllable parameter and the overall equipment efficiency parameter at the historical time corresponding to the K first target points; the predicted value of the overall equipment efficiency is obtained based on the target value of the controllable parameter; and the expected improvement rate of the overall equipment efficiency is obtained based on the overall equipment efficiency at the current time and the predicted value of the overall equipment efficiency, so as to obtain the parameter optimization scheme including the target value of the controllable parameter, the predicted value of the overall equipment efficiency and the expected improvement rate of the overall equipment efficiency. Therefore, the maintenance personnel can intuitively understand the effect of adjusting the operation of the target device based on the parameter optimization scheme, and then adjust the target device based on the parameter optimization scheme to ensure the production efficiency improvement effect of the target device.

[0087] In some embodiments, the first weight of each first target point can be obtained based on the quality performance parameter at the historical time corresponding to the K first target points and the first distance, wherein the first distance is the distance between the point corresponding to the running parameter and the overall equipment efficiency parameter at the current time in the index cluster and each first target point; and the quality performance parameter is obtained based on the overall equipment efficiency parameter. The target value of the controllable parameter is obtained based on the first weight of the K first target points and the controllable parameter at the historical time corresponding to the K first target points.

[0088] In actual execution, the quality performance parameter can be obtained based on the performance parameter and the quality rate parameter in the overall equipment efficiency parameter of the target device, for example, the quality performance parameter is the product of the performance parameter and the quality rate parameter of the target device, wherein the performance parameter can be obtained based on the total number of production parts, the running time and the idle rate of the target device, and the quality rate parameter can be obtained based on the total number of production parts and the number of defective parts of the target device. The overall equipment efficiency parameter of the target device can further include the availability parameter of the target device, and the availability parameter can be obtained based on the running time, the downtime and the planned running time of the target device.

[0089] In some embodiments, the availability parameter, the quality rate parameter and the performance parameter in the overall equipment efficiency parameter of the target device can be calculated based on the following formula:

[0090]

[0091] In some embodiments, a distance (i.e., a first distance) between each of the K first target points and the coordinate point quantified according to the uncontrollable parameter at the current time can be obtained, the first distance can be calculated based on the two coordinate values of the first target point and the coordinate point quantified according to the uncontrollable parameter at the current time, and the quality performance parameter of the historical time corresponding to each of the K first target points and the first distance are used to obtain a first weight of each of the first target points.

[0092] In actual implementation, the first weight of each of the first target points can be obtained based on the following formula:

[0093]

[0094] wherein ω i represents the first weight of the i-th first target point, QP i represents the quality performance parameter of the i-th first target point, d(x, x i represents the first distance between the coordinate point x quantified according to the uncontrollable parameter at the current time and the i-th first target point, and i is a positive integer greater than 0 and less than K.

[0095] In some embodiments, the target value of the controllable parameter can be obtained based on the first weight corresponding to the K first target points and the controllable parameter at the historical time. The controllable parameter can be temperature, humidity, pressure, speed, or other types of parameters that can be theoretically controlled.

[0096] In actual implementation, the target value of the controllable parameter can be obtained based on the following formula:

[0097]

[0098] wherein P i represents the controllable parameter value of the i-th first target point, P opt represents the target value of the controllable parameter.

[0099] According to the overall device efficiency improvement method applied to the field of general semiconductors, the first weight of each of the K first target points is obtained based on the quality performance parameter at the historical time corresponding to the K first target points and the first distance, the target value of the controllable parameter is obtained based on the first weight of the K first target points and the controllable parameter at the historical time corresponding to the K first target points, the target value of the controllable parameter is obtained, and the target device is adjusted based on the target value of the controllable parameter to ensure the production efficiency of the target device.

[0100] In some embodiments, after obtaining the predicted value of the overall device efficiency based on the target value of the controllable parameter, the actual value of the overall device efficiency can be obtained after the target device is controlled with the target value of the controllable parameter as the control target; the K nearest neighbor model is updated based on the first difference value, wherein the first difference value is the difference between the actual value of the overall device efficiency and the predicted value of the overall device efficiency.

[0101] In actual implementation, the actual value of the overall device efficiency of the target device after the target device is controlled based on the target value of the controllable parameter can be obtained.

[0102] In some embodiments, the difference (i.e., the first difference value) between the predicted value of the overall device efficiency and the actual value of the overall device efficiency can be obtained based on the predicted value of the overall device efficiency of the target device in the parameter optimization scheme, and the K nearest neighbor model is updated based on the first difference value.

[0103] In some embodiments, the K nearest neighbor model can be updated by Laplace smoothing based on the first difference value.

[0104] According to the overall device efficiency improvement method applied to the field of general semiconductors, the actual value of the overall device efficiency is obtained, and the K nearest neighbor model is updated based on the difference between the actual value of the overall device efficiency and the predicted value of the overall device efficiency, so that the K nearest neighbor model is dynamically adjusted and optimized, and the close real-time combination between controllable parameter optimization and overall device efficiency calculation is ensured, thereby improving the overall operation efficiency of the target device.

[0105] In some embodiments, the difference (i.e., the first difference value) between the actual value of the overall device efficiency and the predicted value of the overall device efficiency can be calculated, and the quality performance parameters corresponding to the K first target points are updated in a case where the first difference value is less than or equal to a threshold value.

[0106] In some embodiments, in a case where the first difference value is less than or equal to a threshold value, the quality performance parameters corresponding to the K first target points can be updated to update the K nearest neighbor model.

[0107] According to the overall device efficiency improvement method applied to the field of general semiconductors, the difference (i.e., the first difference value) between the actual value of the overall device efficiency and the predicted value of the overall device efficiency is calculated, and the quality performance parameters are updated in a case where the first difference value is less than or equal to a threshold value, so that the K nearest neighbor model is dynamically adjusted and optimized, thereby improving the overall operation efficiency of the target device.

[0108] In some embodiments, a difference (i.e., a first difference) between an actual value of the overall device efficiency and an expected value of the overall device efficiency can be calculated, and in a case where the first difference is greater than a threshold value, M second target points in the index cluster are deleted; wherein the second target points are points whose distance from points corresponding to the controllable parameter at the current time is less than or equal to a preset value.

[0109] In actual implementation, in a case where the first difference is greater than the threshold value, M points in the index cluster whose distance from points corresponding to the controllable parameter at the current time is less than or equal to the preset value can be deleted, or M second target points closest to the points corresponding to the controllable parameter at the current time in the index cluster, and the operating parameter corresponding to each of the M second target points and the overall device efficiency parameter can be deleted.

[0110] According to the overall device efficiency improvement method applied to the field of general semiconductors, by calculating a difference (i.e., a first difference) between an actual value of the overall device efficiency and an expected value of the overall device efficiency, and in a case where the first difference is greater than a threshold value, M second target points in the index cluster are deleted, the K-neighbor model is dynamically adjusted and optimized, thereby improving the overall operating efficiency of the target device.

[0111] In some embodiments, for each of the K first target points, a second weight can be obtained based on a second distance and a smoothing factor coefficient; wherein the second distance is the distance between each first target point and the remaining K-1 first target points; and an updated quality performance parameter is obtained based on the second weight of the K first target points and the quality performance parameter corresponding to each first target point.

[0112] In actual implementation, for each of the K first target points, the distance between the first target point and the remaining K-1 first target points can be obtained, and the second weight of the first target point can be obtained based on the distance between the first target point and the remaining K-1 first target points; and the quality performance parameter of the first target point can be updated based on the second weight of the remaining K-1 first target points and the quality performance parameter of the remaining K-1 first target points.

[0113] In actual implementation, the second weight of each of the K first target points can be obtained based on the following formula:

[0114]

[0115] wherein ω ij represents the second weight of the i th first target point, d(x j ,x i ) represents the distance between the i th first target point and the j th first target point, and a represents a smoothing factor coefficient.

[0116] In actual implementation, the updated quality performance parameter of each of the K first target points can be obtained based on the following formula:

[0117]

[0118] wherein QP i ′ represents the updated quality performance parameter of the i-th first target point, QP j represents the quality performance parameter of the j-th first target point.

[0119] In order to better understand the overall device efficiency improvement method applied to the field of general semiconductors provided by the embodiments of the present application, further explanation is made below, and it should be understood that the following discussion is only exemplary.

[0120] The present application provides an overall device efficiency improvement method applied to the field of general semiconductors, and the specific steps can be as shown in Figure 2

[0121] Step 1: Obtain the running parameters and overall device efficiency parameters of the target device, and calculate the overall device efficiency of the target device.

[0122] In some embodiments, data collection and preprocessing can be performed on the running parameters and overall device efficiency parameters of the target device, which can include the following steps:

[0123] 1. Target device running parameter collection: sensors and Internet of Things devices can be used to collect the running time, downtime, total production number and defective product number of the target device in real time.

[0124] In actual implementation, for the running time and downtime of the target device, device sensors and control systems can be used to record in real time, and the sensors can detect the current, vibration or other running state parameters of the target device to determine whether the target device is running; for the total production number and defective product number, counters and quality detection systems on the automated production line can be used to record the total number of parts produced and the number of defective products, and the quality detection system can use visual detection, laser scanning or other methods including manual detection to determine whether the parts are qualified; for the planned running time, the planned running time of the target device can be obtained from the production planning system, which is usually set by the production scheduling system and represents the expected running time of the target device in a specific period; for the idle speed, the idle speed of the target device can be obtained from the target device specifications or running log, which represents the production speed of the target device in the no-load state.

[0125] 2. Data cleaning: remove outliers and noise data to ensure the accuracy of the running parameters, and statistical methods and machine learning models can be used to detect and correct outliers.​

[0126] 3. Data storage: Store the cleaned running parameters in a database for subsequent analysis and processing. A relational database or a NoSQL database can be used, and the appropriate storage solution can be selected based on the characteristics of the data.

[0127] 4. Data synchronization: Ensure synchronization and consistency of running parameters between different systems using data synchronization tools and technologies such as ETL tools or data middleware.

[0128] In some embodiments, after obtaining the running data of the target device, the overall device efficiency of the target device can be calculated based on the running data of the target device by the following formula:

[0129] Overall device efficiency = availability x quality rate x performance

[0130]

[0131] In actual execution, the cleaned running parameters can be transmitted to the central data processing system, and the distributed stream processing technology of Apache Kafka can be used for data integration. The collected running parameters can be sent to the message queue of Kafka by each device sensor, and Kafka can be used to transmit the running parameters to the central data processing system in a high-throughput and low-latency manner. After receiving the running parameters, Apache Spark Streaming real-time processing framework can be used to process and calculate the data, and the overall device efficiency parameter and overall device efficiency can be obtained by using the above calculation formula.

[0132] Step two: based on the K-nearest neighbor model, obtain the parameter optimization scheme for the target device.

[0133] 1. Through the K-nearest neighbor model, the running parameters of each historical moment in the stable state are selected based on the quality rate parameters and performance parameters of the historical moments.

[0134] In actual execution, the K-nearest neighbor model can be constructed by the running parameters and overall device efficiency parameters of the target device at each historical moment in the stable running state. The K-nearest neighbor model can include an index cluster, and the index cluster can include a plurality of points, each of which can be a multi-dimensional coordinate point. Each point in the index cluster can have a mapping relationship with the uncontrollable parameters and overall device efficiency parameters of a historical moment.

[0135] In some embodiments, the operating parameters of the target device at each historical moment in a stable operating state and the overall device efficiency parameters can be collected. The operating parameters may include controllable parameters and uncontrollable parameters, and data processing such as extreme sample exclusion and normalization is performed. Then, quantization technology is used to store the uncontrollable parameters, and the uncontrollable parameters are represented as multi-dimensional coordinate points, and a quantized inverted index is established. Then, based on the overall device efficiency parameters, the quality performance parameters are obtained, and the points with quality performance parameters greater than a specific threshold are screened out, and the points with quality performance parameters greater than the specific threshold are inserted into the quantized index cluster. The open source Faiss framework can be used to implement the mentioned quantized inverted index technology.

[0136] 2. Using the K-nearest neighbor model, a parameter optimization solution is obtained based on the current operating parameters and overall equipment efficiency parameters.

[0137] In some embodiments, as Figure 3 As shown, the uncontrollable parameters at the current moment can be quantified into multidimensional coordinate points, and K first target points can be obtained from the index cluster. The first target points can be the K coordinate points in the index cluster that are closest to the coordinate points represented by the uncontrollable parameters of the target device at the current moment in terms of the distance between the coordinate values. K is a positive integer greater than 2.

[0138] In some embodiments, the distance (i.e., the first distance) between each of the K first target points and the coordinate point quantified by the uncontrollable parameters at the current moment, the quality performance parameters and the first distance at the historical moment corresponding to each of the K first target points can be obtained, and the first weight of each first target point can be obtained.

[0139] In actual implementation, the first weight of each first target point can be obtained based on the following formula:

[0140]

[0141] Among them, ω i Represents the first weight of the i-th first target point, QP i represents the quality performance parameter of the first target point i, d(x,x i ) represents the first distance between the coordinate point x of the quantized uncontrollable parameter at the current moment and the i-th first target, where i is a positive integer greater than 0 and less than K.

[0142] In some embodiments, the target value of the controllable parameter can be obtained based on the first weights corresponding to the K first target points and the controllable parameters at the historical moment. The controllable parameter can be temperature, humidity, pressure, speed, or other parameter types that can be theoretically controlled.

[0143] In actual implementation, the target value of the controllable parameter can be obtained based on the following formula:

[0144]

[0145] wherein P i represents the controllable parameter value of the i-th first target point, P opt represents the target value of the controllable parameter.

[0146] For example, taking the humidity parameters 0.5 and 0.6 as samples, the quality performance parameters corresponding to the humidity parameters are 0.8 and 0.75 respectively, when the humidity parameter at the current time is 0.55, the first weights of the humidity parameters are ω1=0.16 and ω2=0.15 respectively, and then the target value of the controllable parameter is calculated by The P1 and P2 here can be normalized temperature, pressure and speed and other controllable parameter values.

[0147] In some embodiments, each of the K-neighbor models records a time stamp, the K-neighbor model can obtain the running parameters of the target device at the latest time, and periodically delete old data that is too long in time according to the time stamp, so as to ensure that the K-neighbor model matches the current running state of the target device, and reduce the size of the K-neighbor model, thereby improving the calculation efficiency and accuracy of the K-neighbor model.

[0148] In some embodiments, the K-neighbor model can be updated and optimized based on Laplace smoothing. Specifically, after controlling the target device based on the parameter optimization scheme, the actual value of the overall device efficiency and the first difference value are obtained, the first difference value is the difference between the actual value of the overall device efficiency and the predicted value of the overall device efficiency, and the K-neighbor model is updated based on the first difference value: in the case that the first difference value is less than or equal to a threshold value, for each of the K first target points, a second weight is obtained based on the second distance and the smoothing factor coefficient, wherein the second distance is the distance between each of the first target points and the remaining K-1 first target points; and based on the second weight of the K first target points and the quality performance parameter corresponding to each of the first target points, an updated quality performance parameter is obtained. In the case that the first difference value is greater than the threshold value, the points in the index cluster whose distance from the point corresponding to the controllable parameter at the current time is less than or equal to a preset value are deleted, and the M second target points closest to the point corresponding to the controllable parameter at the current time and the running parameters and the overall device efficiency parameters corresponding to each of the M second target points in the index cluster can also be deleted.

[0149] In actual execution, the second weight of each of the K first target points can be obtained based on the following formula:

[0150]

[0151] wherein ω ija second weight representing the ith first target point, d(x j i represents the distance between the ith first target point and the jth first target point, and a represents a smoothing factor coefficient.

[0152] In actual implementation, the updated quality performance parameter of each of the K first target points can be obtained based on the following formula:

[0153]

[0154] wherein QP i ′ represents the updated quality performance parameter of the ith first target point, QP j represents the quality performance parameter of the jth first target point.

[0155] Step three: obtaining the remaining useful life and maintenance suggestion of the target device based on the RUL model.

[0156] In some embodiments, the pre-trained remaining useful life (RUL) prediction model can be used to reduce the downtime of the target device caused by after-the-fact maintenance and regular maintenance through predictive maintenance, thereby improving the availability of the target device and further improving the overall device efficiency of the target device. Specifically, the RUL prediction model accurately predicts the remaining life of the device according to the operating parameters and historical data of the target device, helping the operation and maintenance personnel to carry out preventive maintenance before the device fails.

[0157] In actual implementation, the RUL model can be a baseline Transformer model trained by unsupervised transfer learning method. This RUL prediction model includes the following key steps: baseline Transformer model training: the baseline Transformer model can be trained by open industrial dataset, and feature extraction and parameter optimization are performed. Through the domain adversarial network and distance preserving regularization loss method, transfer learning of the target device is performed, so that the baseline Transformer model can accurately predict the remaining life of the target device in the unmarked data environment. After the transfer learning is completed, the data feature vectors of the source domain and the target domain are used to train the baseline Transformer model, further improving the accuracy of the remaining life prediction. After the RUL model training is completed, the current operating parameters of the target device are input into the RUL prediction model to obtain the predicted remaining useful life of the target device.

[0158] After obtaining the RUL prediction result of the target device, the predictive maintenance suggestion can be obtained in the case that the normalized remaining useful life is less than a certain threshold, to remind the operation and maintenance personnel to take necessary preventive maintenance measures before the target device fails. This step can include the following processes:​

[0159] 1. Obtain historical time point prediction remaining useful life line chart: The prediction remaining useful life line chart can reflect the overall aging degree of the target device, helping the operation and maintenance personnel to find possible aging failure causes from key time nodes. By observing the trend of the prediction value of the remaining useful life over time, the operation and maintenance personnel can identify potential problems of performance decline of the target device.

[0160] 2. Monitor the running parameter and overall equipment efficiency parameter offset value: Based on the obtained running parameter and overall equipment efficiency parameter of the target device at the historical time point, the normal value range area of the running parameter and overall equipment efficiency parameter is determined by statistical analysis method (such as mean and standard deviation calculation), and the running parameter and overall equipment efficiency parameter of the target device at the current time are compared with the range to generate maintenance information reference. Specifically, the control range (UCL and LCL) of the running parameter and overall equipment efficiency parameter can be calculated by control chart method, and if the running parameter and overall equipment efficiency parameter at the current time exceed the control range, an abnormality is prompted and inspection is recommended.

[0161] 3. Obtain target maintenance scheme based on knowledge graph and large language model: Based on the running parameter, parameter optimization scheme and maintenance suggestion at the current time, search in the knowledge graph database to identify the corresponding fault mode, and use the large language model to generate the target maintenance scheme and the cited original reference link, which are provided to the operation and maintenance personnel as reference.

[0162] In some embodiments, long-term OEE actual improvement evaluation: The availability parameter baseline value can be obtained based on the historical time point, for example, by the formula After the target device is subjected to predictive maintenance based on the maintenance suggestion obtained based on the remaining useful life of the target device, the actual availability parameter mean value of the week, month, quarter and year is counted, and if the actual availability parameter mean value decreases, the unavailable life range of the prediction remaining useful life in the RUL model is adjusted, and the operation and maintenance personnel are prompted about the failure risk of the target device.

[0163] Step four: visual display of parameter optimization scheme, maintenance suggestion and target maintenance scheme.

[0164] In some embodiments, after obtaining the parameter optimization scheme, the controllable parameter target value in the parameter optimization scheme can be calculated with the controllable parameter at the current time to obtain a normalized difference value, and then a reverse normalization step is performed to obtain the maintenance suggestion of the controllable parameter value of the target device. For example, in the pressure parameter, if the normalized pressure difference value is 0.05, the corresponding reverse normalized actual pressure is 375 mTorr. The maintenance suggestion can be visualized according to the size of the difference between the controllable parameter and the target value and the expected improvement rate of the overall device efficiency, so that the operation and maintenance personnel can make corresponding adjustments and decisions. The visualization can directly display the maintenance suggestion and deviation of each controllable parameter through charts, color coding, etc., helping the operation and maintenance personnel to quickly understand and apply these information.

[0165] In some examples, the running parameters of the target device, the overall device efficiency parameters, the parameter optimization scheme and the maintenance suggestion can be dynamically displayed through the html front-end technology integrated with chart.js, helping the operation and maintenance personnel to more intuitively and efficiently manage and maintain the target device, and thus improving the overall device efficiency of the target device.

[0166] In some embodiments, as shown in Figure 4 The running parameters of the target device, the overall device efficiency and the remaining useful life prediction value can be displayed through the visualization tool. The running parameters of the target device, such as running time, downtime, total production, defective product number, temperature and pressure, etc., can be displayed through the visualization tool in the form of line chart, column chart, pie chart, etc. The trend chart of the overall device efficiency at the historical time can be displayed through the visualization tool to show the change of the overall device efficiency of the target device at different time periods, helping the operation and maintenance personnel to identify long-term trends and abnormal situations. The remaining useful life (RUL) prediction results of the target device at the historical time can be displayed through the visualization tool, including the historical RUL prediction line chart and the current prediction value, helping the operation and maintenance personnel to evaluate the health status and maintenance needs of the device.

[0167] In some embodiments, the following can be visualized based on a visualization device, such as a digital instrument panel: 1. Maintenance recommendations for controllable parameters, and the difference between the current controllable parameters and the target value of the controllable parameters, and the corresponding maintenance recommendations, for example, for the pressure of the target device, when the calculated pressure target value is 375 mtorr and the current pressure value is 420 mtorr, the maintenance recommendation is "recommended pressure parameter: reduce 45 mtorr". 2. When the remaining service life at the current time is less than a certain threshold, a maintenance prompt is triggered, and specific maintenance steps and recommendations are provided, including key components that need to be checked and consumables that may need to be replaced. 3. When the operating parameters of the target device exceed the control range, an alarm is issued, and the operator is advised to check and maintain. The maintenance recommendation information can be displayed in a resident window for easy viewing by the operator at any time. At the same time, the maintenance recommendation is notified to the operator through a text prompt warning popup and an SMS push. The text prompt warning popup will display important information at the fixed lowermost end of the visualization interface, and will use yellow prompt text content or red warning content according to the prompt content type; while the SMS push will call the SMS interface to ensure that the operator can receive and handle the relevant notification in a timely manner. Through this multi-channel reminder and recommendation method, the operator can quickly respond and improve maintenance efficiency.

[0168] In some embodiments, the visualization interface supports generating and exporting maintenance reports to help operators record and analyze. The report content can include: maintenance recommendation report, including historical time remaining service life prediction value line chart, operating parameter change trend chart, target maintenance scheme obtained based on knowledge graph and large language model, etc. And the operating data of the target device, the overall device efficiency, the RUL prediction result, etc. can be exported in Excel, PDF, etc. format for further analysis and sharing. Through the visualization device, the operator can obtain the status of the target device in real time, and obtain the maintenance recommendation and the target maintenance scheme in a timely manner, so as to effectively improve the availability parameter of the target device and the overall device efficiency, and ensure that the target device runs in the best state.

[0169] The embodiment of the present application also provides an overall device efficiency improvement device applied to the field of general semiconductors.

[0170] As Figure 5 The overall device efficiency improvement device applied to the field of general semiconductors 500 includes a first acquisition module 510, a second acquisition module 520, and a display module 530.

[0171] The first acquisition module 510 is configured to acquire the operating parameters of the target device at the current time and the overall device efficiency parameters.

[0172] The second acquisition module 520 is configured to acquire a parameter optimization scheme based on the K-neighbor model, the running parameter at the current moment, and the overall device efficiency parameter, and predict the remaining service life of the target device based on the running parameter at the current moment and the overall device efficiency parameter to acquire a maintenance suggestion; the K-neighbor model is constructed based on the running parameter at a plurality of historical moments and the overall device efficiency parameter in a stable running state of the target device.

[0173] The display module 530 is configured to visually display the parameter optimization scheme and the maintenance suggestion.

[0174] According to the overall device efficiency improvement apparatus applied to the field of general semiconductors, the parameter optimization scheme and the maintenance suggestion are acquired in a manner of acquiring a parameter optimization scheme based on the running parameter at the current moment of the target device and the overall device efficiency parameter and the K-neighbor model, and predicting the remaining service life of the target device based on the running parameter at the current moment and the overall device efficiency parameter, so as to replace the manner of determining an optimization scheme by artificial experience, so that the acquired parameter optimization scheme and maintenance suggestion can guarantee the production efficiency improvement effect of the target device.

[0175] In some embodiments, the overall device efficiency improvement apparatus 500 applied to the field of general semiconductors can further include:

[0176] The output module is configured to search a knowledge graph in the field of general semiconductors based on the large language model according to the running parameter at the current moment, the parameter optimization scheme, and the maintenance suggestion, and output a target maintenance scheme.

[0177] In some embodiments, the running parameter includes a controllable parameter and an uncontrollable parameter; and the second acquisition module 520 can include:

[0178] The determination unit is configured to determine K first target points in an index cluster of the K-neighbor model based on the uncontrollable parameter at the current moment; each point in the index cluster corresponds to a running parameter at a historical moment and an overall device efficiency parameter; and K is a positive integer greater than 2.

[0179] The acquisition unit is configured to acquire a parameter optimization scheme based on the controllable parameter at the historical moment corresponding to the K first target points and the overall device efficiency parameter; and the parameter optimization scheme includes a target value of the controllable parameter.

[0180] In some embodiments, the parameter optimization scheme can further include an expected value of the overall device efficiency and an expected improvement rate of the overall device efficiency; and the acquisition unit can include:

[0181] The first acquisition subunit is configured to acquire a target value of the controllable parameter based on the controllable parameter at the historical moment corresponding to the K first target points and the overall device efficiency parameter.

[0182] The second obtaining subunit is configured to obtain a predicted value of the overall device efficiency based on a target value of the controllable parameter.

[0183] The third obtaining subunit is configured to obtain an expected improvement rate of the overall device efficiency based on the overall device efficiency at the current moment and the predicted value of the overall device efficiency.

[0184] In some embodiments, the first obtaining subunit is specifically configured to:

[0185] obtain a first weight of each of the K first target points based on a quality performance parameter at a historical moment corresponding to the K first target points and a first distance, wherein the first distance is a distance between a point corresponding to the current moment's operating parameter and the overall device efficiency parameter in the index cluster and each of the first target points, and wherein the quality performance parameter is obtained based on the overall device efficiency parameter;

[0186] obtain the target value of the controllable parameter based on the first weight of the K first target points and the controllable parameter at the historical moment corresponding to the K first target points.

[0187] In some embodiments, the overall device efficiency improvement apparatus 500 applied to the field of general semiconductors can further include:

[0188] The third obtaining module is configured to obtain an actual value of the overall device efficiency after the target device is controlled with the target value of the controllable parameter as a control target;

[0189] The updating module is configured to update the K nearest neighbor model based on a first difference value, wherein the first difference value is a difference between the actual value of the overall device efficiency and the predicted value of the overall device efficiency.

[0190] In some embodiments, the updating module can include:

[0191] The updating unit is configured to update the quality performance parameter corresponding to the K first target points in a case where the first difference value is less than or equal to a threshold value.

[0192] In some embodiments, the updating module can further include:

[0193] The deleting unit is configured to delete M second target points in the index cluster in a case where the first difference value is greater than the threshold value, wherein the second target point is a point with a distance less than or equal to a preset value between a point corresponding to the controllable parameter at the current moment.

[0194] In some embodiments, the updating unit is specifically configured to:

[0195] For each of the K first target points, obtain a second weight based on a second distance and a smoothing factor coefficient, wherein the second distance is a distance between each of the first target points and the remaining (K-1) first target points.

[0196] Based on the second weight of the K first target points and the quality performance parameter corresponding to each first target point, an updated quality performance parameter is obtained.

[0197] The device for improving overall device efficiency in the field of general semiconductor can be an electronic device or a component in an electronic device, such as an integrated circuit or a chip. The electronic device can be a terminal or other device other than a terminal. For example, the electronic device can be a mobile phone, a tablet computer, a notebook computer, a palm computer, a vehicle-mounted electronic device, a mobile Internet device (MID), an augmented reality (AR) / virtual reality (VR) device, a robot, a wearable device, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), and can also be a server, a network attached storage (NAS), a personal computer (PC), a television (TV), a teller machine, or a self-service machine. The embodiments of the present application are not limited in this regard.

[0198] The device for improving overall device efficiency in the field of general semiconductor can be a device with an operating system. The operating system can be a Windows operating system, an Android operating system, an IOS operating system, or other possible operating systems. The embodiments of the present application are not limited in this regard.

[0199] The device for improving overall device efficiency in the field of general semiconductor 500 provided by the embodiments of the present application can implement each process of the method embodiments, and details are not described herein again to avoid repetition. Figures 1 to 4 The method embodiments implement each process, and details are not described herein again to avoid repetition.

[0200] In some embodiments, as shown in Figure 6 The embodiments of the present application also provide an electronic device 600, which includes a processor 601, a memory 602, and a computer program stored in the memory 602 and executable on the processor 601. The program is executed by the processor 601 to implement each process of the above-mentioned method embodiments for improving overall device efficiency in the field of general semiconductor, and can achieve the same technical effects. Details are not described herein again to avoid repetition.

[0201] It should be noted that the electronic device in the embodiments of the present application includes the mobile electronic device and the non-mobile electronic device described above.

[0202] The embodiments of the present application further provide a non-transitory computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement each process of the above-mentioned embodiments of the whole device efficiency improvement method applied to the field of general semiconductor and achieve the same technical effects. To avoid repetition, details are not described herein.

[0203] The processor is the processor in the electronic device described in the above-mentioned embodiments. The readable storage medium includes a computer readable storage medium, such as a computer readable only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0204] The embodiments of the present application further provide a computer program product, which includes a computer program. The computer program is executed by a processor to implement the above-mentioned whole device efficiency improvement method applied to the field of general semiconductor.

[0205] The processor is the processor in the electronic device described in the above-mentioned embodiments. The readable storage medium includes a computer readable storage medium, such as a computer readable only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0206] The embodiments of the present application further provide a chip, which includes a processor and a communication interface. The communication interface is coupled with the processor. The processor is used to run a program or an instruction to implement each process of the above-mentioned whole device efficiency improvement method applied to the field of general semiconductor and achieve the same technical effects. To avoid repetition, details are not described herein.

[0207] It should be understood that the chip mentioned in the embodiments of the present application can also be referred to as a system level chip, a system chip, a chip system or a system on chip, etc.

[0208] It should be noted that, in the present document, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises a", "comprising", or "comprises" does not, without more constraints, preclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element. Additionally, it should be noted that the terms "one embodiment", "some embodiments", "certain embodiments", "certain examples", or "some examples" as used in the present document are intended to refer to one or more embodiments or examples that do not necessarily have to cover all embodiments or examples of the present application. In other words, use of the above terms does not necessarily refer to the same embodiment or example. Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0209] From the above description of the embodiments, it is apparent that the above-mentioned method of the embodiments can be realized by means of software and a general purpose hardware platform, of course, can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solution of the present application can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, magnetic disc, optical disc), and includes a plurality of instructions for making a terminal (which can be a mobile phone, computer, server, or network device, etc.) execute the method described in the embodiments of the present application.

[0210] The embodiments of the present application are described above in conjunction with the drawings, but the present application is not limited to the above-described specific embodiments, which are merely illustrative rather than restrictive, and a person of ordinary skill in the art can make many forms without departing from the purpose of the present application and the scope protected by the claims.

[0211] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "certain embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Furthermore, the described specific features, structures, materials, or characteristics can be combined in any suitable manner in any one or more embodiments or examples.

[0212] While the embodiments of the application have been shown and described, it is to be understood that the embodiments can be varied, modified, substituted and changed by those skilled in the art without departing from the principles and spirit of the application, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for improving overall equipment efficiency applied in the pan-semiconductor field, characterized in that: include: Obtain the current operating parameters and overall equipment efficiency parameters of the target device; Obtaining a parameter optimization solution based on a K-nearest neighbor model, the current operating parameters, and the overall equipment efficiency parameters, and predicting the remaining useful life of the target equipment based on the current operating parameters and the overall equipment efficiency parameters to obtain maintenance recommendations; The K nearest neighbor model is constructed based on the operating parameters of the target device at multiple historical moments in a stable operating state and the overall device efficiency parameter; The parameter optimization plan and the maintenance suggestion are visually displayed.

2. The overall device efficiency improvement method applied to the pan-semiconductor field according to claim 1 is characterized in that: After visually displaying the parameter optimization plan and the maintenance suggestion, the method further includes: Based on the large language model, the knowledge graph of the pan-semiconductor field is searched according to the current operating parameters, the parameter optimization plan and the maintenance suggestions, and the target maintenance plan is output.

3. The overall device efficiency improvement method applied to the pan-semiconductor field according to claim 1 or 2, characterized in that: The operating parameters include controllable parameters and uncontrollable parameters; The obtaining of a parameter optimization solution based on the K-nearest neighbor model, the current operating parameters, and the overall equipment efficiency parameters includes: Based on the uncontrollable parameters at the current moment, determining K first target points in an index cluster of the K nearest neighbor model; each point in the index cluster corresponds to the operating parameter and the overall equipment efficiency parameter at a historical moment; K is a positive integer greater than 2; Based on the controllable parameters and the overall equipment efficiency parameters at historical moments corresponding to the K first target points, the parameter optimization scheme is obtained; the parameter optimization scheme includes target values ​​of the controllable parameters.

4. The overall device efficiency improvement method applied to the pan-semiconductor field according to claim 3 is characterized in that: The parameter optimization plan also includes: the estimated value and expected improvement rate of overall equipment efficiency; The obtaining of the parameter optimization solution based on the controllable parameters and the overall equipment efficiency parameters at the historical moments corresponding to the K first target points includes: Obtaining target values ​​of the controllable parameters based on the controllable parameters and the overall equipment efficiency parameter at the historical moments corresponding to the K first target points; Obtain an estimated value of overall equipment efficiency based on the target values ​​of controllable parameters; Based on the overall equipment efficiency at a current moment and the estimated value of the overall equipment efficiency, an expected improvement rate of the overall equipment efficiency is obtained.

5. The overall device efficiency improvement method applied to the pan-semiconductor field according to claim 4 is characterized in that: The obtaining of target values ​​of the controllable parameters based on the controllable parameters and the overall equipment efficiency parameters at the historical moments corresponding to the K first target points includes: Obtaining a first weight for each of the K first target points based on the quality performance parameters and first distances corresponding to the K first target points at historical moments; wherein the first distance is the distance between each of the first target points and a point corresponding to the operating parameters and overall equipment efficiency parameter at the current moment in the index cluster; wherein the quality performance parameter is obtained based on the overall equipment efficiency parameter; Based on the first weights of the K first target points and the controllable parameters at historical moments corresponding to the K first target points, a target value of the controllable parameter is acquired.

6. The overall device efficiency improvement method applied to the pan-semiconductor field according to claim 4 is characterized in that: After obtaining an estimated value of overall equipment efficiency based on the target value of the controllable parameter, the method further includes: After the target device is controlled with the target value of the controllable parameter as the control target, obtaining an actual value of the overall device efficiency; The K-nearest neighbor model is updated based on a first difference value, wherein the first difference value is a difference between an actual value of the overall equipment efficiency and an expected value of the overall equipment efficiency.

7. The overall device efficiency improvement method applied to the pan-semiconductor field according to claim 6 is characterized in that: The updating of the K nearest neighbor model based on the first difference includes: When the first difference is less than or equal to a threshold, the quality performance parameters corresponding to the K first target points are updated.

8. The overall device efficiency improvement method applied to the pan-semiconductor field according to claim 7 is characterized in that: The updating of the K nearest neighbor model based on the first difference further includes: When the first difference is greater than the threshold, M second target points in the index cluster are deleted; wherein the second target point is a point whose distance from the point corresponding to the controllable parameter at the current moment is less than or equal to a preset value.

9. The overall device efficiency improvement method applied to the pan-semiconductor field according to claim 7, characterized in that: The updated quality performance parameters include: For each of the K first target points, obtaining a second weight based on a second distance and a smoothing factor coefficient; wherein the second distance is the distance between each of the first target points and the remaining (K-1) first target points; An updated quality performance parameter is acquired based on the second weights of the K first target points and the quality performance parameter corresponding to each of the first target points.

10. A device for improving overall equipment efficiency applied in the field of pan-semiconductors, characterized in that: include: A first acquisition module is used to obtain the current operating parameters and overall equipment efficiency parameters of the target equipment; a second acquisition module, configured to acquire a parameter optimization solution based on a K-nearest neighbor model, the current operating parameters, and the overall equipment efficiency parameters, and to predict the remaining useful life of the target equipment based on the current operating parameters and the overall equipment efficiency parameters, and to acquire maintenance recommendations; The K nearest neighbor model is constructed based on the operating parameters of the target device at multiple historical moments in a stable operating state and the overall device efficiency parameter; A display module is used to visually display the parameter optimization plan and the maintenance suggestion.

Images