Clean room environment control method and system

Abstract

The embodiment of the present specification provides a clean room environment control method and system, the method is executed by a processor, comprising: obtaining target unit data of a target unit in different areas of a clean room, the target unit data comprising at least one of a target unit type, position data and displacement data; determining target activity data based on the target unit data, the target activity data comprising a moving frequency and a moving distance; determining an environmental pollution index based on the area data and the target activity data; determining an execution strategy based on the environmental pollution index; the execution strategy comprising a first clean strategy and a second clean strategy.

Description

TECHNICAL FIELD

[0001] The present specification relates to the field of clean room environment control, and in particular, to a clean room environment control method and system. BACKGROUND

[0002] A clean room, also known as a dust-free workshop, dust-free room or clean room, is an environment with low pollution level. The pollution sources of the clean room include dust, air-borne microorganisms, suspended particles and chemical volatile gases. In order to ensure the environmental quality of the clean room, the air in the clean room needs to be cleaned periodically, and the temperature in the room needs to be adjusted. Generally, the air in the clean room is cleaned by air cleaning equipment (such as a fan), but due to different functions and different personnel in different areas of the clean room, the same cleaning strategy may lead to lack of pertinence in cleaning.

[0003] In view of how to control the environment of the clean room, CN102864952B proposes a clean room and a purification unit thereof, which intelligently controls the temperature, humidity, dust content and static electricity of the clean room to achieve the purpose of cleaning the clean room. However, it does not consider the influence of obstacles, changes in personnel and other factors in the clean room on air flow stability and air pollution, which may make the determined clean room cleaning scheme not accurate enough, and the intelligent control of the clean room sometimes has defects.

[0004] Therefore, it is desirable to provide a clean room environment control method and system to improve the accuracy and effectiveness of clean room cleaning and improve the comprehensiveness of intelligent control of the clean room. SUMMARY

[0005] One of the embodiments of the present specification provides a clean room environment control method. The method is executed by a processor and includes: obtaining target unit data of different areas in a clean room, the target unit data including at least one of target unit type, position data and displacement data; determining target activity data based on the target unit data, the target activity data including moving frequency and moving distance; determining an environmental pollution index based on area data and the target activity data; determining an execution strategy based on the environmental pollution index; the execution strategy including a first cleaning strategy and a second cleaning strategy.

[0006] One of the embodiments of the present specification provides a clean room environment control system, the system comprising: an acquisition module configured to acquire target unit data of a target unit in different areas of a clean room, the target unit data comprising at least one of a target unit type, position data, and displacement data; a first determination module configured to determine target activity data based on the target unit data, the target activity data comprising a moving frequency and a moving distance; a second determination module configured to determine an environment pollution index based on area data and the target activity data; and a third determination module configured to determine an execution strategy based on the environment pollution index, the execution strategy comprising a first cleaning strategy and a second cleaning strategy.

[0007] One of the embodiments of the present specification provides a clean room environment control device, comprising a processor configured to execute the computer instructions or part of the instructions to implement the clean room environment control method of any one of the above embodiments.

[0008] One of the embodiments of the present specification provides a computer readable storage medium, the storage medium storing computer instructions, when the computer reads the computer instructions in the storage medium, the computer executes the clean room environment control method of any one of the above embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0009] The present specification will be further illustrated in the manner of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, the same numbers represent the same structures, wherein:

[0010] Figure 1 is an exemplary block diagram of a clean room environment control system according to some embodiments of the present specification;

[0011] Figure 2 is an exemplary flowchart of a clean room environment control method according to some embodiments of the present specification;

[0012] Figure 3 is an exemplary schematic diagram of determining a clean evaluation value based on a strategy evaluation model according to some embodiments of the present specification;

[0013] Figure 4 is an exemplary schematic diagram of predicting target unit data at a future time point according to some embodiments of the present specification. DETAILED DESCRIPTION

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the drawings needed to be used in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some examples or embodiments of the present specification, and for those skilled in the art, the present specification can also be applied to other similar scenarios without creative labor on the basis of these drawings. Unless it is clear from the language environment or otherwise stated, the same reference numbers in the drawings represent the same structure or operation.

[0015] It should be understood that the "system", "device", "unit" and / or "module" used herein is a method for distinguishing different components, elements, parts, sections or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0016] As shown in the specification and claims, unless the context clearly indicates otherwise, the words "one", "a", "an", and / or "the" do not refer to the singular, but can also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements explicitly identified, and these steps and elements do not constitute an exclusive list, and the method or device can also include other steps or elements.

[0017] Flowcharts are used in the present specification to illustrate the operations performed by the system according to the embodiments of the present specification. It should be understood that the preceding or subsequent operations are not necessarily performed in sequence. On the contrary, each step can be processed in reverse order or simultaneously. At the same time, other operations can be added to these processes, or one or more steps of the operation can be removed from these processes.

[0018] The clean room environment quality is affected by many factors, in addition to the dust carried by air flow, air-borne microorganisms, suspended particles, and chemical volatile gases, the movement of personnel in the clean room, the movement of various equipment, etc. can also bring pollutants to different areas. And, the different distribution of personnel and equipment in different areas of the clean room will also cause differences in the environment quality of the clean room, and the clean room cleaning method needs to consider all-round problems and develop a comprehensive and targeted cleaning plan. CN102864952B only controls the temperature, humidity, dust content and static electricity of the clean room intelligently, but does not consider the environmental quality changes caused by obstacles, personnel distribution and flow in the clean room, and the determined clean solution may have defects.

[0019] Therefore, some embodiments of the present specification can improve the comprehensiveness and effectiveness of the clean room cleaning strategy, and improve the accuracy of intelligent control of the clean room, by obtaining target unit data (i.e., relevant data of personnel or equipment, etc.) and target activity data of target units in different areas of the clean room, determining an environmental pollution index based on the area data and the target activity data, and determining an execution strategy (clean room cleaning strategy) based on the environmental pollution index.

[0020] Figure 1 is an exemplary module diagram of a clean room environment control system according to some embodiments of the present specification. As shown in Figure 1 the clean room environment control system 100 includes an obtaining module 110, a first determining module 120, a second determining module 130, and a third determining module 140.

[0021] The obtaining module 110 is configured to obtain target unit data of target units in different areas of the clean room. The target unit data can include at least one of target unit type, location data, and displacement data.

[0022] The first determining module 120 is configured to determine target activity data based on the target unit data. The target activity data can include movement frequency and movement distance.

[0023] The second determining module 130 is configured to determine an environmental pollution index based on the area data and the target activity data.

[0024] In some embodiments, the second determining module 130 can be further configured to determine a particle residue value of at least one area based on the area data and the target activity data, and determine the environmental pollution index based on the particle residue value of the at least one area.

[0025] The third determining module 140 is configured to determine an execution strategy based on the environmental pollution index. The execution strategy can include a first cleaning strategy and a second cleaning strategy.

[0026] In some embodiments, the third determining module 140 can be further configured to determine a first initial cleaning strategy and a second initial cleaning strategy based on historical execution strategy data, and determine the execution strategy based on the environmental pollution index, the target activity data, the first initial cleaning strategy, and the second initial cleaning strategy.

[0027] In some embodiments, the third determining module 140 can be further configured to predict target unit data at a future time point based on the target unit data, determine a first initial cleaning strategy at the future time point and a second initial cleaning strategy at the future time point based on the target unit data at the future time point, and determine an execution strategy at the future time point based on the target unit data at the future time point, the first initial cleaning strategy at the future time point, and the second initial cleaning strategy at the future time point.

[0028] For more information about the target unit data, target activity data, execution strategy, and other parameters described above, please refer to the relevant descriptions in other parts of this specification (e.g. Figure 2 ).

[0029] It should be noted that the above description of the cleanroom environment control system 100 and its modules is for the convenience of description only, and does not limit the scope of this specification to the embodiments described. It can be understood that, for those skilled in the art, after understanding the principles of the system, any combination of the modules or connection of the sub-systems with other modules can be made without departing from the principles.

[0030] Figure 2 is an exemplary flowchart of a cleanroom environment control method according to some embodiments of the present specification. As shown in Figure 2 , the flowchart 200 can include the following steps. In some embodiments, the flowchart 200 can be executed by a processor.

[0031] Step 210, obtaining target unit data of target units in different areas of the cleanroom, the target unit data including at least one of target unit type, position data, and displacement data.

[0032] In some embodiments, the processor can divide the cleanroom into multiple different areas in various ways. For example, according to the functions of different areas, the cleanroom can be divided into preparation areas, operation areas, etc. For another example, the cleanroom can be divided into different areas by gridding (e.g., nine-square gridding).

[0033] The target unit refers to movable objects and / or personnel in the cleanroom. For example, including workers, handling devices, movable equipment, etc.

[0034] In some embodiments, the processor can obtain the position coordinates (e.g., center point coordinates) of different areas in the cleanroom and obtain the monitoring images corresponding to each position coordinate area, determine the static and / or dynamic target units present in different areas based on image recognition or RFID recognition.

[0035] The target unit data refers to data information related to the target unit. For example, including target unit type, position data, displacement data, etc.

[0036] The target unit type can include personnel, movable equipment, handling devices, etc.

[0037] The position data refers to the position coordinates of the target unit in the cleanroom.

[0038] The displacement data refers to the movement information of the target unit in the cleanroom. For example, including the moving speed, moving direction, etc. of the target unit.

[0039] In some embodiments, the target unit data can be obtained in various ways. The processor can directly obtain the target unit type through image recognition, and obtain the position data and displacement data through target detection or a positioning device provided by the target unit.

[0040] At step 220, the target activity data is determined based on the target unit data, and the target activity data includes the moving frequency and the moving distance.

[0041] The target activity data refers to data information related to the movement of the target unit. For example, it can include the moving frequency of the target unit, the distance of each movement, etc.

[0042] In some embodiments, the processor can determine the target activity data through sensors. For example, the moving frequency and the moving distance of the target unit are obtained through motion sensors, position sensors, cameras, etc.

[0043] At step 230, the environmental pollution index is determined based on the region data and the target activity data.

[0044] The region data refers to the relevant data of the area space occupied by the devices and equipment that cannot be moved in the region. For example, it includes the floor area, volume, etc. of the devices and equipment that cannot be moved in the region. Exemplarily, the region data is [220, 90, 80], indicating that the floor area of the target unit that cannot be moved in the region is 220*90 cm, and the height is 80 cm.

[0045] The environmental pollution index refers to the degree of pollution in the clean room. For example, it includes the concentration of impurities or dust in the air.

[0046] In some embodiments, the processor can analyze and process the region data and the target activity data through modeling or by using various data analysis algorithms, such as regression analysis and discriminant analysis, to determine the environmental pollution index.

[0047] In some embodiments, determining the environmental pollution index based on the region data and the target activity data includes: determining the particle residue value of at least one region based on the region data and the target activity data; and determining the environmental pollution index based on the particle residue value of the at least one region.

[0048] The particle residue value can reflect the degree of particle residue in the air in the clean room. The higher the particle residue value, the greater the particle density that affects the cleanliness of the clean room.

[0049] In some embodiments, the processor can determine the particle residue value in various ways. For example, the particle residue value is directly measured using a particle counter.

[0050] In some embodiments, the processor can determine the particle residual value according to the area data, the moving frequency, and the moving distance, by the following formula (1).

[0051] O = W1 x B + W2 x C + W3 x D (1) Wherein, B, C, and D represent the influence coefficients of the area data, the moving frequency, and the moving distance on the particle residual value in the clean room, respectively; W1 represents the area data; W2 represents the moving frequency; and W3 represents the moving distance.

[0052] In some embodiments, the three influence coefficients B, C, and D can be determined by fitting based on historical monitoring results of the air quality detector and historical area data, historical moving frequency, and historical moving distance corresponding to the historical monitoring results.

[0053] In some embodiments, the processor can set a preset table containing different particle residual values and corresponding environmental pollution indexes, and determine the environmental pollution indexes of different areas by table lookup. In some embodiments, the processor can also determine the average value or the weighted average value of the particle residual values of at least one area as the environmental pollution index. Wherein, the weight of the area with equipment in the clean room is greater than that of the area without equipment.

[0054] In some embodiments, the processor determines the environmental pollution index based on the particle residual value of at least one area, including: processing the particle residual value, the last execution strategy data, the historical target unit data, the historical clean room usage data, and the product data of at least one area based on a pollution index determination model, to determine the environmental pollution index; the pollution index determination model is a machine learning model.

[0055] In some embodiments, the pollution index determination model can be a machine learning model, such as a recurrent neural network model (RNN) and the like.

[0056] The last execution strategy data refers to the specific cleaning parameters of the execution strategy of the clean room closest to the current time point. The cleaning parameters can include air flow intensity, coverage range, execution time, and the like. For the description of the execution strategy, please refer to the description below.

[0057] The historical target unit data refers to the target unit data of the historical time (including the current time) in at least one area. For example, including historical target unit type, historical position data, historical displacement data, and the like.

[0058] The historical clean room usage data refers to the clean room usage data of the historical time (including the current time). For example, including the running situation of the production equipment in the clean room, the usage situation of the detection instrument and / or office facilities, and the like.

[0059] The product data refers to relevant data of a product produced in the clean room. For example, it includes product material, product size, processing method, etc.

[0060] In some embodiments, the processor can obtain the last execution strategy data, the historical target unit data, the historical clean room usage data, and the product data from a storage device inside or outside the clean room environment control system.

[0061] In some embodiments, the contamination index determination model can be trained by using first training samples with a large number of first labels.

[0062] In some embodiments, the first training samples include sample particle residue values of the zones, sample historical execution strategy data, sample historical target unit data, sample historical clean room usage data, and sample product data, which can be obtained based on historical data of the clean room. The first labels can be actual detected environmental contamination indexes corresponding to the first training samples, which can be determined based on the historical data.

[0063] In some embodiments of the present specification, by using the contamination index determination model to determine the environmental contamination index, the accuracy of the predicted environmental contamination index can be improved, which is beneficial to subsequent determination of the execution strategy based on the environmental contamination index.

[0064] In some embodiments of the present specification, by determining the particle residue values of at least one zone and then determining the environmental contamination index based on the particle residue values, the efficiency and accuracy of determining the environmental contamination index can be improved.

[0065] Step 240, determining an execution strategy based on the environmental contamination index; the execution strategy includes a first clean strategy and a second clean strategy.

[0066] The execution strategy refers to a strategy for cleaning the clean room. For example, it includes a first clean strategy, a second clean strategy, etc.

[0067] The first clean strategy refers to a strategy for cleaning the clean room by using turbulent flow and / or radiant flow.

[0068] The first clean strategy can include a first air flow intensity, a first coverage range, a first execution time, etc. The first air flow intensity refers to the intensity of the turbulent flow and / or the radiant flow (such as high, medium, and low levels, etc.); the first coverage range refers to the range covered by the turbulent flow and / or the radiant flow in the clean room; and the first execution time refers to the execution time of the first clean strategy.

[0069] The second cleaning strategy is a strategy for cleaning the clean room by using laminar flow. The second cleaning strategy can include a second air flow intensity, a blowing direction, and a second execution time. The second air flow intensity refers to the intensity of the laminar flow. The blowing direction is the direction in which the fan blows the laminar flow, and the blowing direction can be determined based on the location of the area to be cleaned and historical experience (for example, directly facing the area to be cleaned). The second execution time is the execution time of the second cleaning strategy.

[0070] In some embodiments, the execution of the first cleaning strategy precedes the execution of the second cleaning strategy.

[0071] In some embodiments, the first cleaning strategy and the second cleaning strategy are time series data, the execution time point of the first cleaning strategy is earlier than that of the second cleaning strategy, and the execution of the first cleaning strategy affects the execution of the second cleaning strategy. For example, if the cleaning of the clean room meets the requirements after the execution of the first cleaning strategy, the second cleaning strategy can reduce the cleaning parameters or even not be executed.

[0072] In some embodiments, the cleaning strategies at different times are different, and the historical cleaning strategy affects the cleaning strategy at the future time point. For example, the historical cleaning strategy is effective, and the cleaning intensity of the cleaning strategy at the future time point can be smaller.

[0073] In some embodiments of the present specification, the execution of the first cleaning strategy precedes the execution of the second cleaning strategy, which can reasonably and effectively clean the clean room and reduce energy consumption and cleaning cost.

[0074] In some embodiments, the execution strategy includes cleaning the clean room based on air laminar flow; the air laminar flow includes low-temperature plasma for electrostatic treatment.

[0075] In some embodiments, the processor can generate a low-temperature plasma flow through a low-temperature plasma flow generation device, mix it with the air laminar flow, and deliver it into the clean room for cleaning and electrostatic treatment. The low-temperature plasma flow is composed of low-temperature plasma, which is used to eliminate static electricity and reduce dust adsorption caused by static electricity.

[0076] In some embodiments, the processor can determine the intensity of the low-temperature plasma flow when determining the execution strategy. The intensity of the low-temperature plasma flow is related to the dust concentration in the clean room. The higher the dust concentration, the greater the intensity of the low-temperature plasma flow, which can be determined based on historical experience.

[0077] In some embodiments of the present specification, the air laminar flow including low-temperature plasma is used to clean the clean room, which can better eliminate dust adsorbed by static electricity and improve the cleaning effect and efficiency of the clean room.

[0078] In some embodiments, the processor can determine the execution strategy in multiple ways based on the environmental pollution index. For example, the processor can preset the execution strategy corresponding to different environmental pollution indexes in advance based on historical experience and store them, and directly query the corresponding execution strategy based on the current environmental pollution index when needed.

[0079] In some embodiments, the processor can first determine the first initial cleaning strategy and the second initial cleaning strategy, and then determine the execution strategy. More details can be found in the following description.

[0080] In some embodiments of the present specification, by obtaining the data of the target units in different areas of the clean room, determining the target activity data and the environmental pollution index, and then formulating a reasonable clean room cleaning strategy, the energy consumption and cost can be reduced while maintaining the cleanliness of the clean room and ensuring the quality and progress of the products produced in the clean room.

[0081] In some embodiments, the processor determines the execution strategy based on the environmental pollution index, including: determining the first initial cleaning strategy and the second initial cleaning strategy based on historical execution strategy data; determining the execution strategy based on the environmental pollution index, the target activity data, the first initial cleaning strategy, and the second initial cleaning strategy.

[0082] The historical execution strategy data refers to the execution strategy data for cleaning the clean room at a historical time. For example, it includes the specific cleaning parameters of the execution strategy at the historical time. More details about the cleaning parameters can be found in Figure 2 and related descriptions.

[0083] The first initial cleaning strategy refers to the first cleaning strategy at the current time without real-time data adjustment. Real-time data refers to dynamic data that affects the environmental quality of the clean room. For example, target unit activity data, environmental pollution index, etc.

[0084] The second initial cleaning strategy refers to the second cleaning strategy at the current time without real-time data adjustment.

[0085] Among them, the real-time data adjustment refers to the adjustment of the determined first initial cleaning strategy and second initial cleaning strategy according to the real-time changing data in the clean room, which can be more accurate.

[0086] In some embodiments, the processor can obtain historical target unit data and historical target activity data that are equal to or similar to the current target unit data and target activity data based on the current target unit data and target activity data, and determine the corresponding historical first cleaning strategy and historical second cleaning strategy as the first initial cleaning strategy and the second initial cleaning strategy.

[0087] In some embodiments, the processor can determine the execution strategy in multiple ways based on the environmental pollution index, the target activity data, the first initial cleaning strategy and the second initial cleaning strategy. For example, when the environmental pollution index is greater than the first pollution threshold, the first air flow intensity is increased and the first coverage range is decreased based on the first initial cleaning strategy; similarly, when the environmental pollution index is greater than the second pollution threshold, the second air flow intensity is increased based on the second initial cleaning strategy. For another example, when the average moving frequency of each target unit in the region is greater than the first frequency threshold, the first air flow intensity is increased and the first coverage range is decreased based on the first initial cleaning strategy; when the average moving frequency of each target unit in the region is greater than the second frequency threshold, the second air flow intensity is increased based on the second initial cleaning strategy. In some embodiments, the adjustment range of the first initial cleaning strategy and the second initial cleaning strategy is related to the range exceeding the pollution threshold or the frequency threshold, and the greater the range, the greater the adjustment range.

[0088] In some embodiments, the processor determines the execution strategy based on the environmental pollution index further comprises: determining the strategy execution time based on the region data and the target activity data.

[0089] The strategy execution time can include the execution duration of the first cleaning strategy, the execution duration of the second cleaning strategy, the execution time interval of the first cleaning strategy and the second cleaning strategy, etc.

[0090] In some embodiments, the processor can determine a time preset table containing different region data, target activity data and corresponding strategy execution time, and determine the execution duration of the first cleaning strategy, the execution duration of the second cleaning strategy, and the execution time interval of the first cleaning strategy and the second cleaning strategy by looking up the table. In some embodiments, when determining the time preset table, the execution time interval of the first cleaning strategy and the second cleaning strategy is proportional to the first air flow intensity of the first cleaning strategy.

[0091] In some embodiments of the present specification, the strategy execution time is determined by the region data and the target activity data, which can more accurately control the execution duration and the time interval of the first execution strategy and the second execution strategy, adapt to the real-time environment of the clean room, avoid blind or fixed clean strategy execution time arrangement, improve cleaning efficiency, and reduce energy consumption and cost.

[0092] Figure 3 is an exemplary schematic diagram of determining a cleaning evaluation value based on a strategy evaluation model according to some embodiments of the present specification.

[0093] In some embodiments, the processor can determine at least one candidate execution strategy based on historical execution strategy data.

[0094] The candidate execution strategy refers to an execution strategy that can be a target execution strategy.

[0095] In some embodiments, the processor may determine at least one candidate execution strategy in a variety of ways.

[0096] For example, the processor can cluster historical execution strategies and identify all historical execution strategies in the cluster with the most members as candidate execution strategies.

[0097] For example, the processor can search the database based on cleanroom usage data and target unit data to obtain at least one set of historical cleanroom usage data and historical target unit data with a similarity greater than a similarity threshold, and determine at least one historical execution strategy corresponding to the at least one set of historical cleanroom usage data and historical target unit data as a candidate execution strategy.

[0098] In some embodiments, the processor can determine the cleanliness evaluation value of at least one candidate execution policy based on a policy evaluation model, and determine the target execution policy based on the cleanliness evaluation value. The policy evaluation model can be a machine learning model with a custom structure as described below, or other neural network models, such as deep neural network models.

[0099] In some embodiments, such as Figure 3 As shown, the strategy evaluation model includes an efficiency evaluation layer 320 and a cleanliness evaluation layer 360. The efficiency evaluation layer 320 is used to process at least one candidate implementation strategy 310-1, regional data 310-2, environmental pollution index 310-3, and historical cleanroom usage data 310-4 to determine the cleanliness efficiency 330; the cleanliness evaluation layer 360 is used to process the cleanliness efficiency 330, cleanroom characteristics 340, and cleanliness cost 350 to determine the cleanliness evaluation value 370 of at least one candidate implementation strategy.

[0100] In some embodiments, the efficiency evaluation layer 320 may be a neural network (NN) model, and the cleanliness evaluation layer 360 may be a backpropagation (BP) neural network.

[0101] For more information on regional data 310-2 and the environmental pollution index 310-3, please refer to [link / reference needed]. Figure 2 And related descriptions. Historical cleanroom usage data 310-4 can refer to the usage of cleanrooms over a historical period, such as the duration of use, distribution and movement of target units, etc.

[0102] Cleanliness efficiency of 330 refers to the cleanliness efficiency of a cleanroom. For example, cleanliness efficiency can be expressed as: clean area area / per minute.

[0103] The cleanroom feature 340 refers to feature data related to the cleanroom. For example, the cleanroom feature can include cleanroom purpose, clean standard, energy saving requirement, cleanroom size, etc.

[0104] The clean cost 350 refers to the cost required to complete the cleaning of the cleanroom. For example, the cost of electricity, etc.

[0105] The clean evaluation value can be used to comprehensively evaluate the cleaning effect and cost consumption of different candidate execution strategies on the cleanroom. The greater the clean evaluation value, the more likely the candidate execution strategy becomes the target execution strategy.

[0106] In some embodiments, the input of the efficiency evaluation layer 320 further includes production plan data 310-5 of the cleanroom in a future time period. Wherein, the future time period can be a future period of time after the preset current time point.

[0107] The production plan data refers to data information related to the production plan. For example, the production plan data can include production tasks, production time arrangement, production equipment and personnel input in the future time period, etc.

[0108] In some embodiments of the present specification, by inputting the production plan data of the future time period to the efficiency evaluation layer, different production plan data can be considered to bring different degrees of pollution to the cleanroom, etc., affect the cleaning efficiency of the cleaning strategy, or may need to improve the cleaning efficiency, so that the cleaning efficiency output by the efficiency evaluation layer is more accurate.

[0109] In some embodiments, the input of the efficiency evaluation layer 320 further includes target unit data 310-6 of the future time point.

[0110] The future time point can include multiple time points in the future time period, or can be the last time point in the future time period.

[0111] The displacement data in the target unit data of the future time point is the moving direction and moving speed of the target unit when moving from the current time position to the corresponding position of the future time point.

[0112] In some embodiments of the present specification, by inputting the target unit data of the future time point to the efficiency evaluation layer, different target unit data can be considered to bring different degrees of pollution to the cleanroom, etc., affect the cleaning efficiency of the cleaning strategy, or may need to improve the cleaning efficiency, further improve the accuracy of the cleaning efficiency output by the efficiency evaluation layer.

[0113] In some embodiments, the output of the efficiency evaluation layer can be the input of the clean evaluation layer, and the efficiency evaluation layer and the clean evaluation layer can be obtained by joint training.

[0114] In some embodiments, the second training sample of the strategy evaluation model comprises a sample execution strategy, sample area data, a sample environmental pollution index, sample cleanroom usage data, sample clean cost, and sample cleanroom characteristics. In some embodiments, the second training sample can further comprise sample production plan data and sample target unit data. The second training sample can be obtained based on historical data of the cleanroom.

[0115] The second label is an actual clean evaluation value corresponding to the sample execution strategy of the second training sample, which can be determined according to manual actual scoring of the sample execution strategy.

[0116] In some embodiments, the processor can input the sample execution strategy, the sample area data, the sample environmental pollution index, and the sample cleanroom usage data (which can further comprise sample production plan data and sample target unit data) into an initial efficiency evaluation layer to obtain an initial clean efficiency; input the initial clean efficiency, the sample clean cost, and the sample cleanroom characteristics into an initial clean evaluation layer to obtain an initial clean evaluation value. A loss function is constructed based on the initial clean evaluation value and the second label, and the parameters of the initial efficiency evaluation layer and the initial clean evaluation layer are updated simultaneously using the loss function. Through parameter updating, a trained efficiency evaluation layer and a clean evaluation layer are obtained, that is, a trained strategy evaluation model is obtained.

[0117] In some embodiments, the processor can determine the target execution strategy in multiple ways based on the clean evaluation value.

[0118] The target execution strategy refers to a strategy finally used for clean processing of the cleanroom, including a target first clean strategy, a target second clean strategy, and a target strategy execution time.

[0119] In some embodiments, the processor can determine the candidate execution strategy with the highest clean evaluation value in the at least one candidate execution strategy as the target execution strategy.

[0120] In some embodiments of the present specification, by analyzing the at least one candidate execution strategy using the strategy evaluation model and generating a corresponding clean evaluation value, a more accurate target execution strategy can be determined, and the clean effect and clean efficiency of the cleanroom can be improved; by dividing the strategy evaluation model into different layers to process data respectively, the efficiency and accuracy of data processing can be improved.

[0121] In some embodiments of the present specification, by determining the execution strategy based on historical execution strategy data, an environmental pollution index, and target activity data, the clean strategy for the cleanroom can be flexibly adjusted, the clean efficiency and effect can be effectively improved, and the clean cost can be reduced.

[0122] In some embodiments, the processor can further be configured to predict target unit data at a future time point based on the target unit data, and determine the first initial cleaning strategy at the future time point, the second initial cleaning strategy at the future time point as the execution strategy at the future time point based on the target unit data at the future time point.

[0123] In some embodiments, the processor can predict target unit data at a future time point based on the target unit data in multiple ways, including position data and displacement data (moving speed and moving direction) at the future time point.

[0124] In some embodiments, the processor can predict position data of the target unit based on the following prediction methods.

[0125] Exemplary prediction methods include:

[0126] S1, the processor generates a candidate target unit moving track based on the current position of the target unit and production work arrangement. The production work arrangement can include positions to which the target unit needs to move / be moved due to production work; the candidate moving track refers to a moving track that can be a target moving track of the target unit.

[0127] In some embodiments, when the moving frequency of the target unit is greater than the moving frequency threshold based on the production work arrangement, several candidate moving tracks can be generated accordingly, so as to determine a more accurate target moving track from more candidate moving tracks.

[0128] S2, record a reference moving track of the target unit within a preset time period, and match the reference moving track with an initial track segment in the candidate moving track. The reference moving track can refer to the actual moving track of the target unit recorded by a camera, a sensor, etc. within the preset time period; the initial track segment can be a beginning segment (i.e. the segment from the starting position in the candidate track segment) in the candidate moving track with a length close to that of the reference moving track.

[0129] S3, the candidate moving track corresponding to the initial track segment with the highest matching degree is taken as the target moving track. The target moving track refers to a moving track used to determine the moving direction and position of the target unit within a future time period. For example, the target moving track can include path length, average moving speed of the target unit, etc.

[0130] In some embodiments, the processor can calculate the coincidence degree of the initial track segment and the reference track segment to determine the matching degree. The higher the coincidence degree, the higher the matching degree.

[0131] S4, determining a moving distance of the target unit at the future time point based on the average moving speed of the target unit in the target moving track and the time length from the current time to the future time point, and determining the position data of the target unit at the future time point based on the moving distance.

[0132] In some embodiments, the processor can determine the average moving speed of the target unit in the target moving track as the moving speed in the displacement data of the target unit. The processor can also determine the moving direction in the displacement data of the target unit based on the position data of the current target unit and the position data of the target unit at the future time point.

[0133] Figure 4 is an exemplary schematic diagram of predicting the target unit data at the future time point according to some embodiments of the present specification.

[0134] In some embodiments, as Figure 4 shown, the processor predicts the target unit data at the future time point based on the target unit data, which includes: obtaining a behavior feature sequence 420 of the target unit based on the target unit data 410; determining the target unit data 310-6 at the future time point based on the behavior feature sequence 420 through a target unit data model 470; and the target unit data model 470 is a machine learning model.

[0135] The behavior feature sequence refers to a sequence composed of the target unit data and the operation features of the target unit collected at multiple time points within a preset historical time period. The operation features can include specific operation behaviors (such as the hand movements of personnel, the product lifting movements of equipment, etc.) of the target unit (such as equipment, personnel, etc.).

[0136] In some embodiments, the processor can obtain the operation features of the target unit within the preset historical time period through image recognition, combine the target unit data within the preset historical time period, arrange in chronological order, and form a behavior feature sequence.

[0137] As Figure 4 shown, in some embodiments, the processor determines the target unit data 310-6 at the future time point based on the behavior feature sequence 420 through the target unit data model 470, which includes: obtaining at least one behavior feature sub-sequence 430 based on the behavior feature sequence 420 with a preset interval length and a preset step length; then determining the label data of the at least one feature sub-sequence 430, inputting the label data 440 of the at least one feature sub-sequence 430, the target unit 450, and the production plan data 460 of the target unit into the target unit data model 470, and outputting the target unit data 310-6 at the future time point.

[0138] Wherein, the preset interval length is positively correlated with the number of target units, and the preset step length is negatively correlated with the number of target units.

[0139] The preset interval length refers to a preset interval length of the behavior feature subsequence. For example, if the preset interval length is 10s, the interval length of the behavior feature subsequence extracted from the behavior feature sequence is 10s (such as [0s-10s], [1s-11s], etc.). In some embodiments, the processor can determine the preset interval length based on the number of target units. The more the number of target units, the longer the preset interval length.

[0140] The preset interval step refers to a preset interval step of two adjacent feature subsequences. For example, if the preset interval step is 2s, the first behavior feature subsequence extracted from the behavior feature sequence is [0s-10s], the second behavior feature subsequence is [2s-12s], and so on. In some embodiments, the processor can determine the preset interval step based on the number of target units. The more the number of target units, the smaller the preset interval step.

[0141] The behavior feature subsequence 430 refers to a segment extracted from the behavior feature sequence 420. The behavior feature sequence 420 can extract multiple behavior feature subsequences 430 (such as feature subsequence 1, feature subsequence 2, …, feature subsequence n, etc.).

[0142] In some embodiments, the processor can obtain at least one behavior feature subsequence based on the preset interval length and the preset step, which can be referred to the above description. For example, based on the preset interval length of 10s and the preset interval step of 2s, the behavior feature subsequences are the segments [0s-10s], [2s-12s], [4s-14s], etc. of the behavior feature sequence in turn.

[0143] The label data refers to the operation of the target unit reflected by the behavior feature subsequence and the operation occurrence probability. It includes the operation type of the target unit and the operation type occurrence probability. For example, the label can include “take medicine (90%)” and “computer office (80%)”.

[0144] In some embodiments, each behavior feature subsequence corresponds to a label. For example, the feature subsequences: feature subsequence 1, feature subsequence 2, …, feature subsequence n, can correspond to the label data: label data 1, label data 2, …, label data n, respectively.

[0145] In some embodiments, the processor can determine the label of the behavior feature subsequence in various ways, such as table lookup, vector matching, etc.

[0146] For example, the processor can match the behavior feature subsequence in the vector database based on the behavior feature subsequence, obtain a reference behavior feature subsequence with a similarity satisfying a similarity threshold, and determine a label corresponding to the reference behavior feature subsequence as the label of the current feature subsequence. The vector database includes a plurality of reference behavior feature subsequences and labels corresponding to the reference behavior feature subsequences. The plurality of reference behavior feature subsequences and the labels corresponding to the reference behavior feature subsequences can be obtained by clustering fragments of historical behavior feature sequences. The clustering algorithm includes K-means, hierarchical clustering, and the like.

[0147] In some embodiments, the target unit data model 470 can be a recurrent neural network model (RNN).

[0148] In some embodiments, the target unit data model can be trained by a third training sample with a large number of third labels. An exemplary training method can include, but is not limited to, gradient descent method, and the like.

[0149] In some embodiments, the third training sample can include a sample target unit, sample label data, and production plan data of the sample target unit. The third training sample can be obtained based on historical data of the clean room. The third label can be actual target unit data of the third training sample at a subsequent time point, which can be determined based on historical data annotation.

[0150] In some embodiments of the present specification, the behavior feature subsequence is obtained by a preset interval length and a preset step length of the behavior feature sequence, and the corresponding label data is determined. The target unit data model can be used to more accurately predict the target unit data at the future time point. The preset interval length and the preset step length can be determined based on the number of target units, which can reduce the processing of the data amount and efficiently obtain the operation features of the target units.

[0151] In some embodiments of the present specification, the behavior feature sequence is obtained by the target unit data, and the target unit data model is used to predict the target unit data at the future time point, which can improve the accuracy and efficiency of the prediction, and thus more accurately implement the cleaning strategy and improve the cleaning effect of the clean room.

[0152] In some embodiments, the processor can determine the target activity data at the future time point based on the target unit data at the future time point by the method of step 220. For more details, refer to the related description of step 220.

[0153] In some embodiments, the processor can determine the first initial cleaning strategy at the future time point and the second initial cleaning strategy at the future time point based on the target unit data at the future time point and the target activity data, in combination with the historical execution strategy data. The determination method is similar to that of determining the first initial cleaning strategy at the current time and the second initial cleaning strategy at the current time. For more details, refer to the related description above.

[0154] In some embodiments, the processor may determine the execution strategy for a future time point in a variety of ways based on the target unit data at a future time point, the first initial cleaning strategy at a future time point, and the second initial cleaning strategy at a future time point.

[0155] In some embodiments, the processor can determine the execution strategy for a future time point based on the environmental pollution index at a future time point, the target activity data at a future time point, and the first and second initial cleanliness strategies at the future time point. The determination method is similar to that for determining the execution strategy at the current time, as described above. For example, when the environmental pollution index at a future time point is greater than a first pollution threshold, the processor can increase the intensity of the first airflow and decrease the first coverage area based on the first initial cleanliness strategy at the future time point.

[0156] In some embodiments, the environmental pollution index at future points in time can be determined in a manner similar to step 230, which will not be elaborated here.

[0157] In some embodiments, the processor can also determine at least one candidate execution strategy for a future time point based on cleanroom usage data and target unit data at future time points, combined with historical execution strategy data. In some embodiments, the processor can determine the target execution strategy for a future time point from the candidate execution strategies at least one future time point based on a strategy evaluation model. Similar to determining the execution strategy, more details can be found above (e.g., Figure 3 Related descriptions.

[0158] The environmental pollution index at future time points in the input strategy evaluation model can be used to determine the model prediction based on the pollution index. For more details, please refer to the relevant description in step 230.

[0159] In some embodiments of this specification, by predicting target unit data at future points in time and then determining the execution strategy at those future points in time, the cleaning strategy for the cleanroom at future points in time can be determined in advance, thereby improving the cleaning efficiency of the cleanroom.

[0160] Some embodiments of this specification also disclose a cleanroom environment control device, including: at least one memory and at least one processor, wherein the at least one memory is used to store computer instructions, and the at least one processor executes the computer instructions or some of the instructions to implement the cleanroom environment control method described in any of the above embodiments.

[0161] Some embodiments of this specification also disclose a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the cleanroom environment control method described in any of the above embodiments.

[0162] Having described the basic concepts, it is obvious that the above detailed disclosure is intended to be illustrative only and not restrictive of the present description. Although the present description has been described with reference to specific exemplary embodiments, it will be apparent to those having ordinary skill in the art that a variety of modifications, improvements and / or alterations can be made to the present description. Such modifications, improvements and / or alterations are therefore contemplated and are within the spirit and scope of the exemplary embodiments of the present description.

[0163] Also, the present description can use particular terminology when describing certain embodiments of the present description. For example, the terms "one embodiment," "an embodiment," "some embodiments," or "one alternative" are used interchangeably, and each denotes a certain feature, structure, or characteristic in at least one embodiment of the present description. As such, it is noted that the use of these phrases in various places in the specification are not intended to exclude that an embodiment described in one place can be incorporated with an embodiment described in another place. Further, it is noted that certain features, structures, or characteristics described in one or more embodiments of the present description can be incorporated in other embodiments.

[0164] Further, the order of processing elements or sequence of steps in the description of the present description, unless specifically stated to the contrary, is not intended to explicitly indicate that particular embodiments of the present description are inherently linked or tied to a particular placement of steps or processing elements. Further, the use of numbering in certain places in the description is used only to help clarify the description and is not intended to limit the scope of the claims in any way whatsoever. Although the above disclosure discusses some presently preferred embodiments of the application, it should be apparent that other embodiments of the application can be practiced within the scope of the description and claims. For example, the described systems and methods can be implemented in hardware, software, or a combination of hardware and software. Further, the described systems and methods can be implemented in a server or a mobile device. Further, the described systems and methods can be implemented in a server or a mobile device.

[0165] Similarly, it is noted that, for the sake of brevity, certain features of the description, which are well known and need not be described in detail, can not be described in the present description. Further, it is noted that the description of the present description has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the present description as claimed in the following description. As such, the present description is intended to be illustrative, and not restrictive, of the claimed present description. For example, while the described systems and methods can be implemented in hardware, software, or a combination of hardware and software, the present description is not limited to any particular implementation. Further, the described systems and methods can be implemented in a server or a mobile device. Further, the described systems and methods can be implemented in a server or a mobile device.

[0166] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0167] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0168] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A method for controlling the environment of a cleanroom, characterized in that, The method is executed by a processor and includes: Acquire target unit data for target units in different areas of a cleanroom. The target unit refers to a movable object and / or person within the cleanroom. The target unit data includes at least one of target unit type, location data, and displacement data. Based on the target unit data, target activity data is determined, including movement frequency and movement distance; Based on regional data and the target activity data, an environmental pollution index is determined. The regional data refers to data related to the space occupied by immovable devices and equipment within the region, including the floor area and volume of these devices and equipment. The determination of the environmental pollution index based on the regional data and the target activity data includes: Based on the regional data and the target activity data, determine the particle residue value of at least one region; The environmental pollution index is determined based on the particulate residue values ​​in the at least one region. An execution strategy is determined based on the environmental pollution index. The execution strategy includes a first cleaning strategy and a second cleaning strategy. The first cleaning strategy refers to cleaning the cleanroom through turbulent flow and / or radiative flow, and includes a first airflow intensity, a first coverage area, and a first execution time. The second cleaning strategy refers to cleaning the cleanroom through laminar flow, and includes a second airflow intensity, a blowing direction, and a second execution time. The execution of the first cleaning strategy precedes the execution of the second cleaning strategy. If the cleanroom meets the cleaning requirements after the first cleaning strategy is executed, the cleaning parameters of the second cleaning strategy are reduced accordingly or the second cleaning strategy is not executed.

2. The method according to claim 1, characterized in that, The determination of the execution strategy based on the environmental pollution index includes: Based on the current target unit data and target activity data, obtain historical target unit data and historical target activity data that are equal to or similar to the current target unit data and target activity data; The historical first clean strategy and the historical second clean strategy corresponding to the historical target unit data and the historical target activity data are determined as the first initial clean strategy and the second initial clean strategy, respectively. The execution strategy is determined by adjusting the first initial cleaning strategy and / or the second initial cleaning strategy based on the environmental pollution index and / or the target activity data.

3. The method according to claim 2, characterized in that, The method further includes: Based on the target unit data, predict the target unit data at future time points; Based on the target unit data at the future time point, determine the first initial cleaning strategy and the second initial cleaning strategy at the future time point. Based on the target unit data at the future time point, the first initial cleaning strategy at the future time point, and the second initial cleaning strategy at the future time point, the execution strategy at the future time point is determined.

4. A cleanroom environment control system, characterized in that, The system includes: The acquisition module is used to acquire target unit data of target units in different areas of the clean room. The target unit refers to a movable object and / or person in the clean room. The target unit data includes at least one of target unit type, location data, and displacement data. The first determining module is used to determine target activity data based on the target unit data, wherein the target activity data includes movement frequency and movement distance; The second determining module is used to determine an environmental pollution index based on regional data and the target activity data. The regional data refers to data related to the space occupied by immovable devices and equipment within the region, including the floor area and volume of the immovable devices and equipment within the region. To determine the environmental pollution index, the second determining module is further used to: Based on the regional data and the target activity data, determine the particle residue value of at least one region; The environmental pollution index is determined based on the particulate residue values ​​in the at least one region. The third determining module is used to determine an execution strategy based on the environmental pollution index. The execution strategy includes a first cleaning strategy and a second cleaning strategy. The first cleaning strategy refers to a strategy of cleaning the cleanroom through turbulent flow and / or radiative flow. The first cleaning strategy includes a first airflow intensity, a first coverage area, and a first execution time. The second cleaning strategy refers to a strategy of cleaning the cleanroom through laminar flow. The second cleaning strategy includes a second airflow intensity, a blowing direction, and a second execution time. The execution of the first cleaning strategy precedes the execution of the second cleaning strategy. If the cleanroom meets the cleaning requirements after the first cleaning strategy is executed, the cleaning parameters of the second cleaning strategy are reduced accordingly or the second cleaning strategy is not executed.

5. The system according to claim 4, characterized in that, The third determining module is further used for: Based on the current target unit data and target activity data, obtain historical target unit data and historical target activity data that are equal to or similar to the current target unit data and target activity data; The historical first clean strategy and the historical second clean strategy corresponding to the historical target unit data and the historical target activity data are determined as the first initial clean strategy and the second initial clean strategy, respectively. The execution strategy is determined by adjusting the first initial cleaning strategy and / or the second initial cleaning strategy based on the environmental pollution index and / or the target activity data.

6. The system according to claim 5, characterized in that, The third determining module is also used for: Based on the target unit data, predict the target unit data at future time points; Based on the target unit data at the future time point, determine the first initial cleaning strategy and the second initial cleaning strategy at the future time point. Based on the target unit data at the future time point, the first initial cleaning strategy at the future time point, and the second initial cleaning strategy at the future time point, the execution strategy at the future time point is determined.

7. A cleanroom environment control device, characterized in that, The device includes at least one memory and at least one processor, the at least one memory being used to store computer instructions, and the at least one processor executing the computer instructions or parts thereof to implement the cleanroom environment control method according to any one of claims 1-3.

8. A computer-readable storage medium, characterized in that, The storage medium stores computer instructions, and when the computer reads the computer instructions, the computer executes the cleanroom environment control method as described in any one of claims 1-3.

Images