Smoke suction and exhaust method, device, equipment and medium

Abstract

The embodiment of the invention provides a smoke suction and exhaust method, device and equipment and a medium. The method comprises the following steps: acquiring multi-modal data of a preset space; determining smoke distribution prediction information of the preset space after the preset time according to the multi-modal data; the air suction capacity of at least one air suction opening of the preset space is determined; according to the smoke distribution prediction information and the air suction capacity of at least one air suction port, at least one target air suction port is determined, air suction equipment of the target air suction port is controlled to perform air suction, the hysteresis quality of smoke suction and exhaust response is reduced, and the suction and exhaust control logic can cope with non-uniformly distributed smoke. And the situation that the energy consumption is too large but the exhaust efficiency is low is avoided, and therefore the user experience is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic control, and particularly to a method, device, equipment and medium for smoke suction and exhaust. Background Art

[0002] Traditional smoke suction and exhaust systems generally adopt a passive smoke exhaust mode triggered by a fixed threshold, and their control trigger logic is only based on the smoke concentration data detected in real time. This lag response mechanism has significant limitations when dealing with complex spatial environments: on the one hand, when the detection device captures that the smoke exceeds the standard, harmful gases often have diffused and accumulated in a local area. At this time, when performing smoke suction and exhaust, the smoke removal cycle will be significantly prolonged; on the other hand, when performing smoke suction and exhaust in areas with complex spatial structures, the smoke may have formed a non-uniform distribution: and the existing technology will have problems of insufficient smoke suction and excessive energy consumption caused by excessive suction and exhaust in the case of dealing with the non-uniform distribution of smoke. Summary of the Invention

[0003] In view of the above problems, embodiments of the present invention are proposed to provide a smoke suction and exhaust method, device, equipment and medium that can overcome the problems of insufficient smoke suction and excessive energy consumption caused by excessive suction and exhaust in the case of dealing with the non-uniform distribution of smoke or at least partially solve the above problems.

[0004] To solve the above problems, embodiments of the present invention disclose a smoke suction and exhaust method, and the method includes: Obtain multimodal data of a preset space; According to the multimodal data, determine the smoke distribution prediction information of the preset space after a preset time; Determine the air suction capacity of at least one air suction port of the preset space; According to the smoke distribution prediction information and the air suction capacity of the at least one air suction port, determine at least one target air suction port and control the air suction equipment of the target air suction port to perform air suction.

[0005] Optionally, the multimodal data includes at least one of smoke concentration data, air flow data and image data, and the determining the smoke distribution prediction information of the preset space after a preset time according to the multimodal data includes: Input at least one of the smoke concentration data, the air flow data and the image data into a preset prediction model, so that the preset prediction model outputs the smoke distribution prediction information of the preset space after a preset time.

[0006] Optionally, the air suction capacity includes at least one of an air suction range and an air suction efficiency; the smoke distribution prediction information includes at least one of smoke concentration prediction information, smoke diffusion trend prediction information and air flow prediction information.

[0007] Optionally, the air suction capacity includes an air suction range and an air suction efficiency. Determining at least one target air suction port based on the predicted smoke distribution information and the air suction capacity of the at least one air suction port, and controlling the air suction device of the target air suction port to perform air suction includes: Determining at least one target air suction port based on at least one of the predicted smoke concentration information, the predicted smoke diffusion trend information, and the predicted air flow information, and the air suction range of the at least one air suction port; Determining target control parameters for the air suction device of the target air suction port based on at least one of the predicted smoke concentration information, the predicted smoke diffusion trend information, and the predicted air flow information, and the air suction efficiency of the at least one target air suction port; Controlling the air suction device corresponding to the target air suction port to perform air suction according to the target control parameters.

[0008] Optionally, determining at least one target air suction port based on at least one of the predicted smoke concentration information, the predicted smoke diffusion trend information, and the predicted air flow information, and the air suction range of the at least one air suction port includes: Dividing the preset space into at least one candidate area; Determining a target area in the at least one candidate area based on at least one of the predicted smoke concentration information, the predicted smoke diffusion trend information, and the predicted air flow information; Determining the overlapping degree between the air suction range of each air suction port and the at least one target area; Determining the target air suction port according to the overlapping degree of each air suction port.

[0009] Optionally, determining target control parameters for the air suction device of the target air suction port based on at least one of the predicted smoke concentration information, the predicted smoke diffusion trend information, and the predicted air flow information, and the air suction efficiency of the at least one target air suction port includes: Determining the working weight of the at least one target air suction port according to the predicted smoke concentration information, the predicted smoke diffusion trend information, and the predicted air flow information; the working weight is used to represent the importance of the at least one target air suction port in the current smoke suction and exhaust process; Determining at least one target control parameter according to the air suction efficiency of the at least one target air suction port and the working weight of the at least one target air suction port.

[0010] Optionally, determining target control parameters for the air suction device of the target air suction port based on at least one of the predicted smoke concentration information, the predicted smoke diffusion trend information, and the predicted air flow information, and the air suction efficiency of the at least one target air suction port includes: Determine the energy consumption data of the at least one target air intake opening; Determine at least one target control parameter according to the energy consumption data of the at least one target air intake opening, the predicted smoke concentration information, the predicted smoke diffusion trend information, the air intake efficiency of the at least one target air intake opening, and a preset multi-objective optimization model.

[0011] Optionally, the determining of the target control parameter for the air intake device of the target air intake opening according to at least one of the predicted smoke concentration information, the predicted smoke diffusion trend information, the predicted air flow information, and the air intake efficiency of the at least one target air intake opening includes: Obtain the user preference information corresponding to the preset space; Determine at least one target control parameter according to the user preference information corresponding to the preset space, the predicted smoke concentration information, the predicted smoke diffusion trend information, the air intake efficiency of the at least one target air intake opening, and a preset multi-objective optimization model.

[0012] On the other hand, an embodiment of the present invention further provides a smoke suction and exhaust device, and the device includes: A data acquisition module, configured to acquire multi-modal data of a preset space; A data prediction module, configured to determine predicted smoke distribution information of the preset space after a preset time according to the multi-modal data; An ability determination module, configured to determine the air intake ability of at least one air intake opening of the preset space; An air intake execution module, configured to determine at least one target air intake opening according to the predicted smoke distribution information and the air intake ability of the at least one air intake opening, and control the air intake device of the target air intake opening to perform air intake.

[0013] Optionally, the multi-modal data includes at least one of smoke concentration data, air flow data, and image data, and the data prediction module includes: A model prediction sub-module, configured to input at least one of the smoke concentration data, the air flow data, and the image data into a preset prediction model, so that the preset prediction model outputs the predicted smoke distribution information of the preset space after a preset time.

[0014] Optionally, the air intake ability includes at least one of an air intake range and an air intake efficiency; the predicted smoke distribution information includes at least one of predicted smoke concentration information, predicted smoke diffusion trend information, and predicted air flow information.

[0015] Optionally, the air intake ability includes an air intake range and an air intake efficiency, and the air intake execution module includes: An air intake opening determination sub-module, configured to determine at least one target air intake opening according to at least one of the smoke concentration prediction information, the smoke diffusion trend prediction information, and the air flow prediction information and the air intake range of the at least one air intake opening; A parameter determination sub-module, configured to determine target control parameters for the air intake device for the target air intake opening according to at least one of the smoke concentration prediction information, the smoke diffusion trend prediction information, and the air flow prediction information and the air intake efficiency of the at least one target air intake opening; An equipment control sub-module, configured to control the air intake device corresponding to the target air intake opening to intake air according to the target control parameters.

[0016] Optionally, the air intake opening determination sub-module includes: A space division unit, configured to divide the preset space into at least one candidate area; A target area determination unit, configured to determine a target area in the at least one candidate area according to at least one of the smoke concentration prediction information, the smoke diffusion trend prediction information, and the air flow prediction information; An overlap degree determination unit, configured to determine the overlap degree between the air intake range of each air intake opening and the at least one target area; A target air intake opening determination unit, configured to determine the target air intake opening according to the overlap degree of each air intake opening.

[0017] Optionally, the parameter determination sub-module includes: A weight determination unit, configured to determine the working weight of the at least one target air intake opening according to the smoke concentration prediction information, the smoke diffusion trend prediction information, and the air flow prediction information; the working weight is used to represent the importance degree of the at least one target air intake opening in the current smoke intake and exhaust process; A weight control unit, configured to determine at least one target control parameter according to the air intake efficiency of the at least one target air intake opening and the working weight of the at least one target air intake opening.

[0018] Optionally, the parameter determination sub-module includes: An energy consumption data determination unit, configured to determine the energy consumption data of the at least one target air intake opening; A target optimization unit, configured to determine at least one target control parameter according to the energy consumption data of the at least one target air intake opening, the smoke concentration prediction information, the smoke diffusion trend prediction information, the air intake efficiency of the at least one target air intake opening, and a preset multi-objective optimization model.

[0019] Optionally, the parameter determination sub-module includes: A user preference acquisition unit, configured to acquire user preference information corresponding to the preset space; A preference parameter determination unit, configured to determine at least one target control parameter according to the user preference information corresponding to the preset space, the smoke concentration prediction information, the smoke diffusion trend prediction information, the air suction efficiency of the at least one target air suction port, and a preset multi-objective optimization model.

[0020] Correspondingly, an embodiment of the present invention discloses an electronic device, including: a processor, a memory, and a computer program stored on the memory and capable of running on the processor. When the computer program is executed by the processor, it implements each step of the above-mentioned embodiment of a smoke suction and exhaust method.

[0021] Correspondingly, an embodiment of the present invention discloses a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements each step of the above-mentioned embodiment of a smoke suction and exhaust method.

[0022] The embodiments of the present invention have the following advantages: By acquiring multi-modal data of a preset space and determining the smoke distribution prediction information of the preset space after a preset time according to the multi-modal data, the early grasp of the smoke diffusion trend is realized, and the lag of the smoke suction and exhaust response is reduced. Then, by determining the air suction capacity of at least one air suction port of the preset space, at least one target air suction port is determined according to the smoke distribution prediction information and the air suction capacity of the at least one air suction port, and the air suction device of the target air suction port is controlled to perform air suction. By determining the target air suction port in the air suction ports and controlling the target air suction port, the smoke suction and exhaust control logic can cope with non-uniformly distributed smoke, avoiding the situation of excessive energy consumption but low exhaust efficiency, thereby improving the user experience. Description of the Drawings

[0023] Figure 1 is a flowchart of the steps of an embodiment of a smoke suction and exhaust method of the present invention; Figure 2 is a schematic diagram of area division of an embodiment of a smoke suction and exhaust method of the present invention; Figure 3 is a structural block diagram of an embodiment of a smoke suction and exhaust device of the present invention. Detailed Embodiments

[0024] To make the above objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the drawings and specific embodiments.

[0025] Existing technologies, when smoke extraction is performed in areas with complex spatial structures, can lead to uneven smoke distribution. Specifically, existing treatment methods allow areas requiring smoke extraction to naturally accumulate more smoke due to air flow characteristics, while the system continues to operate according to static control logic. This static control logic can result in low-concentration areas receiving excessive exhaust, resulting in energy waste, while high-concentration areas cannot receive sufficient exhaust due to total air volume restrictions. This ultimately leads to excessive energy consumption and low exhaust efficiency.

[0026] One of the core concepts of the embodiments of the present invention is to solve the lag of smoke suction and exhaust by actively predicting smoke distribution information based on multimodal data, and dynamically open the air intake according to the prediction results, thereby meeting the problem of uneven distribution of smoke in complex spaces.

[0027] Reference Figure 1 , shows a flowchart of a smoke suction and exhaust method embodiment of the present invention, which may specifically include the following steps: Step 101: Acquire multimodal data of a preset space.

[0028] In daily life, all kinds of production activities are faced with the existence of smoke. The essence of smoke is a dynamic aerosol system formed by the phase change of matter under heat. It generally has the characteristics of thermodynamic drive, irreversible diffusion and hazard carrier. The smoke suction and exhaust method proposed in the present invention focuses on the application scenarios where smoke suction and exhaust are required in closed or semi-closed spaces. Below, this solution will be explained with two typical application scenarios: kitchen scenario and industrial workshop scenario; this method can be carried on a central control device with central control capabilities. The central control device has different forms of expression depending on the scenario. For example, in the kitchen scenario, the central control device can be a central control screen, or a cloud server in a cloud collaboration scenario. In the industrial workshop scenario, it can be a central control device with decision-making capabilities such as a central control server.

[0029] Multimodal data is a data concept that was created to address the perceptual limitations and information blind spots faced by a single data type when solving practical problems. Specifically, it refers to a collection of multiple types of data collected through different technical means at the same time and in the same space. The data types containing at least two data types can be various types, such as numerical types, image types, and text types. All types of data must be marked with unified timestamps and spatial coordinates to ensure that the data association is valid. For the present invention, after obtaining multimodal data, data cleaning and data formatting are required to ensure that data noise is removed and the spatiotemporal characteristics of the multimodal data are unified. Exemplarily, in the present invention, the multimodal data obtained in the kitchen scenario can be various data such as the numerical value of kitchen fume concentration, the thermal map or thermal imaging data of heat sources such as stoves, the air flow direction and speed in the kitchen, and the visual data of the kitchen captured by the camera; the multimodal data obtained in the industrial workshop scenario can be the temperature numerical value of pollution points, such as the temperature of welding points in a welding workshop, the visual data captured by the camera, the equipment log data, the pipeline pressure value of workshop pipelines, the gas chromatography data, the audio data collected by the microphone, and other various multi-source data. The present invention does not limit this.

[0030] The core of obtaining multimodal data is the sensor. The arrangement of the sensor can affect the authenticity and effectiveness of data collection. The specific arrangement methods include position selection, the number of sensors selected, etc., which can be flexibly changed according to different application scenarios. For example, in the kitchen scenario, the arrangement method of the sensor can be as follows: select a thermal imaging camera, a smoke concentration sensor, and an air flow sensor. According to the size of the kitchen and the positions of pollution sources such as stoves and ovens, the number of these sensors can be flexibly changed. For example, in a small kitchen, 2 to 3 sensors of each type are sufficient. The installation positions can be key positions such as directly above the pollution source, in the front side, at the corner, and at the kitchen exit. The core strategy is to give priority to the space corners and sensitive areas where smoke may spread in the scene, such as near the pollution source, while ensuring that the preset space is fully covered, and ensuring that the sensors will not be damaged due to the characteristics of the application scenario. For the industrial workshop scenario, it can be flexibly set on the premise of the above core strategy according to the scene characteristics. The present invention does not limit this.

[0031] Step 102: Determine the smoke distribution prediction information of the preset space after a preset time according to the multimodal data.

[0032] After unifying the spatio-temporal characteristics of the multimodal data, the multimodal data can be used for smoke distribution prediction. The essence of the smoke distribution prediction information is the spatial concentration distribution state of the smoke generated by the pollution source in a preset space at a specific future moment. The preset time represents a specific future moment, and different preset times can be set according to different business scenarios, so as to predict the diffusion trend and concentration information of the smoke at different time scales. For example, in the kitchen scenario, the setting range of the preset time can be set based on the reaction time of the ingredients. For example, in the context of stir-frying ingredients, the preset time can be set within the range of 10 to 60 seconds according to the reaction time of oil pyrolysis; in the industrial workshop scenario, the preset time can be set according to the specific production time for completing a single production task. The method of prediction through multimodal data can be based on the calculation of fluid mechanics formulas or can also be predicted by using existing prediction models.

[0033] In one embodiment, the multimodal data includes at least one of smoke concentration data, airflow data, and image data, and step 102 may include the following sub-steps: Sub-step S11, input at least one of the smoke concentration data, the airflow data, and the image data into a preset prediction model, so that the preset prediction model outputs smoke distribution prediction information for the preset space after a preset time.

[0034] As mentioned above, the method of realizing smoke distribution prediction information through multimodal data can be achieved through a prediction model. In actual application scenarios, the prediction model generally consists of two main parts: a feature extraction part for extracting spatio-temporal feature data of multimodal data, and a distribution prediction part for predicting smoke distribution based on the extracted spatio-temporal feature data. For the selection of the feature extraction part, a feature extraction model based on Vit (Vision Transformer) can be used, such as a CNN (Convolutional Neural Network)-ViT hybrid model or a Vit-Adapter. For the distribution prediction part, corresponding selections need to be made according to different scenarios due to the smoke generation characteristics and the nature of the smoke itself. Exemplarily: in a kitchen scenario, a CNN-LSTM (Long Short-Term Memory) hybrid model can be used for subsequent prediction, while in an industrial workshop scenario, models such as the DeepLabV3+ model or the UNet model can be selected to achieve it. After generating the prediction results, the actual smoke distribution information after a preset time will continue to be obtained, and the accuracy of the preset prediction model will be continuously optimized through the actual smoke distribution information. Combining the prediction model and multimodal data, the smoke distribution situation at the future prediction moment can be accurately predicted, providing reliable data support for subsequent smoke suction and exhaust.

[0035] Step 103, determine the air suction capacity of at least one air suction port in the preset space; In the field of smoke suction and exhaust, in addition to the important information of smoke distribution prediction information, another important piece of information is the suction capacity of the suction device in the preset space. Generally, when sucking and exhausting smoke in a closed or semi-closed space, it is necessary to rely on the suction port. The manifestation form of the suction port can present various forms according to the scene needs. For example, it can be various forms such as a suction port, a ventilation port, a ventilation fan, or a ventilation duct. The suction capacity of the suction port can be quantitatively represented by the performance of the suction device corresponding to the suction port. Exemplarily, the performance of the suction device at different opening degrees can be tested in a laboratory or a simulation environment. Specifically, it can be to test the effective negative pressure range created by the suction device at different opening degrees and use it as the suction range of the suction port; or how much target smoke can be sucked away by the suction device per unit time at different opening degrees and use it as the suction efficiency. In addition to the performance at different opening degrees, the test variables can also be comprehensively determined according to relevant indicators required by the business, such as the airway design and pipe diameter size of the suction port. Correspondingly, key indicators such as the installation position and quantity of the suction port can be comprehensively determined according to the above test results and actual business needs such as the budget dimension.

[0036] Exemplarily, the suction port can be set to include: a top suction port installed directly above or slightly in front of the pollution source; a side / back suction port arranged on the side of the pollution source or on the wall close to the pollution source; a covering suction port on the ceiling, side wall, or corner of the preset space. A certain angle and distance should be maintained between the suction port and the pollution source according to the characteristics of the application scenario to improve the smoke capture efficiency.

[0037] After the installation of the suction port is completed, the suction capacity of the suction port can be obtained by acquiring the working parameters of the current suction port and combining the data in the test results. The implementation method can be determined by setting a mapping table of preset working parameters and relevant suction capacities. The quantization standard of the suction capacity can be flexibly set according to the scene needs. Exemplarily, the suction range weight and the suction efficiency weight can be set according to the application scenario. The quantitative numerical representations of the suction range and the suction efficiency can be represented by normalization according to the test data, and the sum of the products of the corresponding weight data is used as the quantitative representation of the suction capacity. The combination of the suction port and the suction device described in the present invention can dynamically select various implementation methods according to different application scenarios. For example, in the kitchen scenario, the combination of the suction port and the suction device can be various devices such as an oil fume extractor, an air purification device, and an air conditioning device; in the industrial scenario, the implementation methods of the combination of the suction port and the suction device can be various implementation methods such as an industrial exhaust fan, an industrial ventilation duct, and an industrial dust removal system.

[0038] In one embodiment, the air suction capacity includes at least one of an air suction range and an air suction efficiency; the smoke distribution prediction information includes at least one of smoke concentration prediction information, smoke diffusion trend prediction information, and air flow prediction information.

[0039] The smoke concentration prediction information is used to represent the smoke concentration in a preset space at a certain future moment. Its specific manifestation form can be an image display: different colors in the figure represent different concentrations. One way is to use red to represent the area with the highest concentration, green to represent the area with the lowest concentration, and transitional colors such as yellow to represent the intermediate concentration areas. The colors corresponding to the concentrations in the image display can be freely set.

[0040] The smoke diffusion trend prediction information and the air flow prediction information are used to represent the smoke diffusion situation at a certain future moment. For example, it is predicted that within the next 5 minutes, the change process of the smoke concentration diffusion in the preset space can be reflected by color changes or the movement of concentration hotspots. Taking the kitchen scenario as an example, if the smoke is concentrated in the cooking area of the kitchen at the beginning, then in the next few minutes, the prediction map will show that the oil fume concentration gradually spreads to other parts of the kitchen, especially to open areas or areas with weak air flow.

[0041] In the present invention, by selecting at least one of the air suction range and the air suction efficiency as the representation of the air suction capacity, selecting at least one of the smoke concentration prediction information, the smoke diffusion trend prediction information, and the air flow prediction information as the representation of the smoke distribution prediction information, and performing subsequent predictions based on these several pieces of data; the prediction accuracy can be improved and the smoke suction and exhaust efficiency can be improved.

[0042] Step 104, determine at least one target air suction port according to the smoke distribution prediction information and the air suction capacity of the at least one air suction port, and control the air suction device of the target air suction port to perform air suction.

[0043] After determining the smoke distribution prediction information and the air suction capacity of the at least one air suction port, the target air suction port can be determined according to these two indicators, and the air suction device corresponding to the target air suction port can be controlled to perform air suction operations; the operation content includes the opening degree, opening time, etc. of the air suction device corresponding to the air suction port.

[0044] In one embodiment, the air suction capacity includes an air suction range and an air suction efficiency, and step 104 may include the following sub-steps: Sub-step S21, determine at least one target air suction port according to at least one of the smoke concentration prediction information, the smoke diffusion trend prediction information, and the air flow prediction information and the air suction range of the at least one air suction port; One of the confirmation logics for the target air intake is that, through at least one of the smoke concentration prediction information, smoke diffusion trend prediction information, and air flow prediction information, the distribution of smoke can be roughly judged within a certain future time, and according to the smoke distribution, an air intake with a sufficient air intake range can be selected as the target air intake. Exemplarily, an air intake whose air intake range completely covers the area with the highest smoke concentration can be used as the target air intake.

[0045] In one embodiment, sub-step S21 may include the following sub-steps: Sub-step S211, dividing the preset space into at least one candidate area; To facilitate a more specific quantification of the confirmation logic of the target air intake and achieve precise smoke suction and exhaust in the preset space, the preset space can be divided into at least one candidate area through a certain logic. The division logic can be determined according to the importance level of the area. For example, in a kitchen scenario, pollution sources such as the stove can be set as important areas. When determining the target air intake, the air intakes within the important areas or those that can cover the important areas will be preferentially activated; in addition to according to the importance level, the candidate areas can also be divided according to the installation position of the air intake; Exemplarily, one of the area division methods can be to follow that the space grid should be able to map to the coverage range of the air intake: according to the size of the air intake and the area, it is clear that each air intake is responsible for controlling a certain number of areas or each area contains a certain number of air intakes. The division can be carried out according to the air intake capacity of the air intake.

[0046] Sub-step S212, determining the target area in the at least one candidate area according to at least one of the smoke concentration prediction information, smoke diffusion trend prediction information, and air flow prediction information; Through the smoke concentration prediction information, it is determined which areas in the candidate areas need to be subjected to smoke suction and exhaust treatment at a certain future time, and through the smoke diffusion trend prediction information, it can be determined which areas have the risk of smoke diffusion at a certain future time and need to be pre-activated. The candidate areas that need to be subjected to smoke suction and exhaust treatment and the candidate areas that need to be pre-activated for smoke suction and exhaust are used as the target areas.

[0047] Sub-step S213, determining the overlapping degree between the air intake range of each air intake and the at least one target area; After knowing which areas are the target areas, it is possible to determine which air intakes need to be activated according to the coverage degree of the target areas by the air intakes, that is, the overlapping degree, that is, to determine which air intakes are the target air intakes.

[0048] Sub-step S214, determining the target air intake according to the overlapping degree of each air intake.

[0049] Exemplarily, referring to Figure 2, showing a schematic diagram of regional division of an embodiment of a smoke suction and exhaust method according to the present invention: In this example, the preset space can be evenly divided into 16 candidate regions. An air suction port 1 is arranged between region 2 and region 3, an air suction port 2 is arranged between region 5 and region 12, an air suction port 3 is arranged in the middle of region 6, region 7, region 11, and region 10, an air suction port 4 is arranged between region 8 and region 9, and an air suction port 5 is arranged in the middle of region 14 and region 15. In addition to the even division, the regional division can also be carried out according to whether the region contains a pollution source as mentioned above or according to the established positions of the air suction ports, or can be carried out in real time according to the smoke concentration. The present invention does not limit this.

[0050] In the actual scenario, the generation and diffusion mode of smoke are not constant. Therefore, to avoid smoke escape, when determining the target air suction port each time, after determining the target region according to the current smoke concentration prediction information and diffusion trend prediction information, it is also necessary to determine the target air suction port corresponding to the target region. One way to determine the target air suction port can be to use the air suction ports with an overlap degree greater than a certain threshold as the target air suction ports. By dividing the preset space into multiple candidate regions, the accuracy and efficiency of smoke suction and exhaust are improved.

[0051] Sub-step S22, determining the target control parameters of the air suction device for the target air suction port according to at least one of the smoke concentration prediction information, smoke diffusion trend prediction information, and air flow prediction information and the air suction efficiency of the at least one target air suction port; After determining the target air suction port, the target control parameters of the air suction device of the target air suction port can be determined so that the air suction device performs the air suction operation according to this target control parameter. For example, if it is judged according to the smoke diffusion trend prediction information and air flow prediction information that the smoke may spread rapidly at this time, the air suction device corresponding to the target air suction port will be turned on at a higher power; or if it is obtained from the smoke concentration prediction information that the smoke covered by target air suction port A will have a higher concentration while the smoke concentration covered by target air suction port B is relatively small, then at this time, the air suction device corresponding to target air suction port A will be turned on at a higher power, and the air suction device corresponding to target air suction port B will be turned on at a lower power. As for the mapping relationship between the concentration and the opening power or opening degree, it can be set according to actual needs, and the present invention does not limit this.

[0052] In the actual application environment, sub-step S22 can flexibly adjust the specific implementation manner according to business requirements. The following presents three exemplary embodiments: In the first exemplary embodiment, sub-step S22 may include the following steps: Determine the working weight of the at least one target air intake according to the predicted smoke concentration information, the predicted smoke diffusion trend information, and the predicted air flow information; the working weight is used to represent the importance of the at least one target air intake in the current smoke suction and exhaust process; Determine at least one target control parameter according to the air suction efficiency of the at least one target air intake and the working weight of the at least one target air intake.

[0053] Exemplarily, the way to determine the working weight can be: set C i 、T i and F i Three parameters, where C i represents the smoke concentration within the coverage of the i-th target air intake after a certain moment; T i represents the smoke diffusion trend within the coverage of the i-th target air intake after a certain moment, and the smoke diffusion trend can be represented by the smoke concentration change rate; F i : represents the air flow correction factor within the coverage of the i-th target air intake after a certain moment. The setting logic of the correction factor can be that the weaker the air flow intensity, the larger the value, or the greater the influence of the air flow direction on the smoke diffusion trend, the larger the value; each factor will be normalized, and the corresponding normalization method can be any one of the linear normalization (Min-Max) method, the Z-Score standardization method, the maximum absolute value scaling (Max Abs) method, or the decimal scaling method (Decimal Scaling), etc. The present invention does not limit the normalization method; Correspondingly, set the influence coefficients α, β, and γ of the three parameters, where α represents the influence coefficient of the smoke concentration, directly reflecting the severity of pollution, and is the core basis of the suction and exhaust strategy, that is, high-concentration smoke must be processed first, otherwise it will spread rapidly; β represents the influence coefficient of the smoke diffusion trend, which can warn of future potential pollution risks and intervene in advance, and is of great significance for preventing diffusion and energy-saving scheduling; γ represents the influence coefficient of the air flow factor, which can quantitatively represent the controllability of the smoke. For example, if the air flow in a certain area is not smooth, even if the concentration is not high, it may form an accumulation dead corner. The determination of the relevant influence coefficients can be adjusted through simulation experiments, and the corresponding evaluation indicators can be the suction and exhaust delay time, representing how long it takes for the smoke concentration to drop to a safe concentration; the suction and exhaust energy consumption, which can be represented by the total power of the fan per unit time; the smoke residue, which can be represented by the ratio of the remaining smoke concentration to the concentration before the smoke suction and exhaust after a certain time of smoke suction and exhaust.

[0054] Then the expression of the working weight W i is: Wi = α * C i + β * T i + γ * F i .

[0055] According to the smoke concentration prediction information, the smoke diffusion trend prediction information, and the air flow prediction information, the working weight corresponding to each target air intake can be obtained. When setting the control parameters for the target air intake with a larger working weight, the air suction and exhaust capacity is given higher priority over other indicators. That is, the corresponding target control parameter settings will also increase accordingly. For example, the daily opening degree of a certain air intake is at most 80% of the standard opening degree. When the working weight of this air intake is greater than a certain threshold, the opening degree of this air intake will be temporarily allowed to exceed 80% of the standard opening degree to facilitate faster smoke suction and exhaust.

[0056] In addition, when multiple target air intakes perform suction and exhaust operations together, abnormal situations such as duct competition and negative pressure disorder may occur. At this time, according to the working weight, the main control air intake and the auxiliary air intake can be set. The target air intake with a larger working weight is the main control air intake, and the target air intake with a smaller working weight is the auxiliary air intake, which has a lower priority. During the suction and exhaust process, the main control air intake maintains normal suction operation; the auxiliary air intake only operates as a supplement to avoid reverse interference of the suction force. The weight calculation is dynamic and will continuously adjust the working weights of each target air intake in real time as the smoke diffusion process changes. After comprehensively considering the smoke concentration, smoke diffusion trend, and air flow factors covered by each target air intake, the working weight of each target air intake is determined, so that the target air intake with a higher working weight has a larger opening degree, thereby obtaining more suction and exhaust resources, improving the pertinence and efficiency of smoke suction and exhaust, and avoiding competition among multiple target air intakes.

[0057] In the second exemplary embodiment, sub-step S22 may include the following steps: Determine the energy consumption data of the at least one target air intake; According to the energy consumption data of the at least one target air intake, the smoke concentration prediction information, the smoke diffusion trend prediction information, the air suction efficiency of the at least one target air intake, and a preset multi-objective optimization model, determine at least one target control parameter.

[0058] In an actual scenario, in addition to the suction and exhaust efficiency, the suction and exhaust energy consumption is also one of the key points to be considered. Smoke suction and exhaust should satisfy the unity of energy consumption and suction and exhaust efficiency. It is not possible to neglect the suction and exhaust efficiency due to emphasizing energy consumption, nor to neglect energy consumption due to emphasizing the suction and exhaust efficiency, resulting in unnecessary waste of resources.

[0059] The normal air suction efficiency and energy consumption do not show a linear relationship but a complex non-linear relationship. Therefore, it is not possible to simply find the balance point between energy consumption and efficiency. At this time, multi-dimensional indicators such as energy consumption data, predicted smoke concentration information, predicted smoke diffusion trend information, and the air suction efficiency of at least one target air suction port can be input into a preset multi-objective optimization model to find an equilibrium solution. The selection of the multi-objective optimization model can be a multi-objective reinforcement learning model, an evolutionary model, such as a multi-objective optimization model with a decomposition-based multi-objective evolutionary algorithm as the core. The present invention does not limit this. By combining the multi-objective optimization model, the balance between energy conservation and air suction and exhaust efficiency can be achieved, improving the user experience.

[0060] In the third exemplary embodiment, sub-step S22 may further include the following steps: Obtain the user preference information corresponding to the preset space; According to the user preference information corresponding to the preset space, the predicted smoke concentration information, the predicted smoke diffusion trend information, the air suction efficiency of the at least one target air suction port, and the preset multi-objective optimization model, determine at least one target control parameter.

[0061] In addition to the preset conditions, the target parameter settings can also be determined according to the user's usage preference information, such as the device operation model set by the user, the device operation parameters specified by the user, the operation status, etc., so that the smoke suction and exhaust process better meets the user's needs.

[0062] Exemplarily, when the user sets the silent energy-saving model as a priority, unless the current situation belongs to a dangerous situation according to the predicted smoke concentration and smoke diffusion trend, even if the predicted smoke concentration and smoke diffusion trend show that the target air suction port should turn on the corresponding air suction device at high power at this time, but due to the user's set preference data, the air suction device will start at a lower power or in the silent mode at this time, or during the suction and exhaust process, the user manually closes the target air suction port that has participated in the suction and exhaust work. At this time, the target air suction port will be re-judged, and the target air suction port manually closed by the user will not participate in the smoke suction and exhaust work within a certain period of time; considering the user preference data as one of the target control parameters can significantly improve the user's satisfaction with the smoke suction and exhaust process.

[0063] Sub-step S23, control the air suction device corresponding to the target air suction port to perform air suction according to the target control parameter.

[0064] Control the air suction device corresponding to the target air suction port to perform air suction according to the determined target control parameter, and complete the smoke suction and exhaust.

[0065] Based on at least one of the smoke concentration prediction information, smoke diffusion trend prediction information, and air flow prediction information, and the air suction efficiency and air suction range of the at least one target air suction opening, the target air suction opening and the corresponding control parameters are determined respectively, and the determination of the target air suction opening is decoupled from the determination of the control parameters of the target air suction opening, realizing dynamic smoke suction and exhaust, which can meet various application scenarios and improve the accuracy of smoke suction and exhaust.

[0066] By obtaining multi-modal data of a preset space, based on the multi-modal data, the smoke distribution prediction information of the preset space after a preset time is determined, realizing the early grasp of the smoke diffusion trend and reducing the lag of the smoke suction and exhaust response. Then, by determining the air suction capacity of at least one air suction opening of the preset space, based on the smoke distribution prediction information and the air suction capacity of the at least one air suction opening, at least one target air suction opening is determined and the air suction equipment of the target air suction opening is controlled to perform air suction. By determining the target air suction opening in the air suction openings and controlling the target air suction opening, the smoke suction and exhaust control logic can cope with non-uniformly distributed smoke, avoiding the situation of excessive energy consumption but low exhaust efficiency, thereby improving the user experience.

[0067] It should be noted that for the method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the embodiments of the present invention are not limited by the described action sequence, because according to the embodiments of the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily essential for the embodiments of the present invention.

[0068] Refer to Figure 3 , a structural block diagram of an embodiment of a smoke suction and exhaust device of the present invention is shown, which may specifically include the following modules: The data acquisition module 201 is used to acquire multi-modal data of a preset space; The data prediction module 202 is used to determine the smoke distribution prediction information of the preset space after a preset time according to the multi-modal data; The capacity determination module 203 is used to determine the air suction capacity of at least one air suction opening of the preset space; The air suction execution module 204 is used to determine at least one target air suction opening according to the smoke distribution prediction information and the air suction capacity of the at least one air suction opening, and control the air suction equipment of the target air suction opening to perform air suction.

[0069] In one embodiment, the multi-modal data includes at least one of smoke concentration data, air flow data, and image data. The data prediction module includes: A model prediction sub-module, configured to input at least one of the smoke concentration data, the air flow data, and the image data into a preset prediction model, so that the preset prediction model outputs smoke distribution prediction information for the preset space after a preset time.

[0070] In one embodiment, the air suction capacity includes at least one of an air suction range and an air suction efficiency; the smoke distribution prediction information includes at least one of smoke concentration prediction information, smoke diffusion trend prediction information, and air flow prediction information.

[0071] In one embodiment, the air suction capacity includes an air suction range and an air suction efficiency, and the air suction execution module includes: An air suction port determination sub-module, configured to determine at least one target air suction port according to at least one of the smoke concentration prediction information, the smoke diffusion trend prediction information, the air flow prediction information, and the air suction range of the at least one air suction port; A parameter determination sub-module, configured to determine target control parameters for the air suction device for the target air suction port according to at least one of the smoke concentration prediction information, the smoke diffusion trend prediction information, the air flow prediction information, and the air suction efficiency of the at least one target air suction port; An equipment control sub-module, configured to control the air suction device corresponding to the target air suction port to perform air suction according to the target control parameters.

[0072] In one embodiment, the air suction port determination sub-module includes: A space division unit, configured to divide the preset space into at least one candidate area; A target area determination unit, configured to determine a target area from the at least one candidate area according to at least one of the smoke concentration prediction information, the smoke diffusion trend prediction information, and the air flow prediction information; An overlap degree determination unit, configured to determine the overlap degree between the air suction range of each air suction port and the at least one target area; A target air suction port determination unit, configured to determine a target air suction port according to the overlap degree of each air suction port.

[0073] In one embodiment, the parameter determination sub-module includes: A weight determination unit, configured to determine the working weight of the at least one target air suction port according to the smoke concentration prediction information, the smoke diffusion trend prediction information, and the air flow prediction information; the working weight is used to represent the importance of the at least one target air suction port in the current smoke suction and exhaust process; A weight control unit, configured to determine at least one target control parameter according to the air suction efficiency of the at least one target air suction port and the working weight of the at least one target air suction port.

[0074] In one embodiment, the parameter determination sub-module includes: An energy consumption data determination unit for determining the energy consumption data of the at least one target air intake; A target optimization unit for determining at least one target control parameter according to the energy consumption data of the at least one target air intake, the smoke concentration prediction information, the smoke diffusion trend prediction information, the air intake efficiency of the at least one target air intake, and a preset multi-objective optimization model.

[0075] In one embodiment, the parameter determination sub-module includes: A user preference acquisition unit for acquiring user preference information corresponding to the preset space; A preference parameter determination unit for determining at least one target control parameter according to the user preference information corresponding to the preset space, the smoke concentration prediction information, the smoke diffusion trend prediction information, the air intake efficiency of the at least one target air intake, and a preset multi-objective optimization model.

[0076] By acquiring multi-modal data of a preset space and determining, according to the multi-modal data, smoke distribution prediction information of the preset space after a preset time, the advance grasp of the smoke diffusion trend is realized, and the hysteresis of the smoke suction and exhaust response is reduced. Then, by determining the air intake capacity of at least one air intake of the preset space and determining at least one target air intake according to the smoke distribution prediction information and the air intake capacity of the at least one air intake, and controlling the air intake device of the target air intake to perform air intake, by determining the target air intake in the air intakes and controlling the target air intake, the smoke suction and exhaust control logic can cope with non-uniformly distributed smoke, avoiding the situation of excessive energy consumption but low exhaust efficiency, thereby improving the user experience.

[0077] For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple. For the relevant parts, refer to the partial description of the method embodiment.

[0078] The embodiment of the present invention further provides an electronic device, including: a processor, a memory, and a computer program stored on the memory and capable of running on the processor. When the computer program is executed by the processor, it implements each process of the above-mentioned method embodiment of a smoke suction and exhaust method and can achieve the same technical effect. To avoid repetition, it will not be elaborated here. In a kitchen scenario, the electronic device can be various devices such as a range hood, an air purification device, an air conditioner, etc.; in an industrial scenario, the electronic device can be various devices such as an industrial exhaust fan, an industrial ventilation duct, and an industrial dust removal system.

[0079] The embodiments of the present invention also provide a computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, it implements each process of the above-mentioned embodiment of a smoke suction and exhaust method and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0080] Each embodiment in this specification is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. For the same or similar parts among the embodiments, reference can be made to each other.

[0081] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a device, or a computer program product. Therefore, the embodiments of the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0082] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal devices generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0083] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0084] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, so that a series of operation steps are executed on the computer or other programmable terminal device to generate a computer-implemented process. Therefore, the instructions executed on the computer or other programmable terminal device provide for implementing the functions in the process Figure 1One or more processes and / or blocks Figure 1 Steps of the functions specified in one or more blocks.

[0085] Although the preferred embodiments of the embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.

[0086] Finally, it should also be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or terminal device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or terminal device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or terminal device comprising the element.

[0087] The above has introduced in detail a smoke suction and exhaust method, device, equipment and medium provided by the present invention. Specific examples are used in this text to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A method for smoke suction and exhaust, characterized in that, The method includes: Obtaining multimodal data of a preset space; the multimodal data includes at least one of smoke concentration data, air flow data, and image data; Inputting at least one of the smoke concentration data, the air flow data, and the image data into a preset prediction model, so that the preset prediction model outputs smoke distribution prediction information of the preset space after a preset time; Determining the air suction capacity of at least one air suction opening in the preset space; Determining at least one target air suction opening according to the smoke distribution prediction information and the air suction capacity of the at least one air suction opening, and controlling the air suction device of the target air suction opening to perform air suction.

2. The smoke suction and exhaust method according to claim 1, characterized in that, The air suction capacity includes at least one of an air suction range and an air suction efficiency; the smoke distribution prediction information includes at least one of smoke concentration prediction information, smoke diffusion trend prediction information, and air flow prediction information.

3. A smoke suction and exhaust method according to claim 2, characterized in that, The air suction capacity includes an air suction range and an air suction efficiency. The determining at least one target air suction opening according to the smoke distribution prediction information and the air suction capacity of the at least one air suction opening, and controlling the air suction device of the target air suction opening to perform air suction includes: Determining at least one target air suction opening according to at least one of smoke concentration prediction information, smoke diffusion trend prediction information, air flow prediction information, and the air suction range of the at least one air suction opening; Determining target control parameters for the air suction device of the target air suction opening according to at least one of smoke concentration prediction information, smoke diffusion trend prediction information, air flow prediction information, and the air suction efficiency of the at least one target air suction opening; Controlling the air suction device corresponding to the target air suction opening to perform air suction according to the target control parameters.

4. A smoke suction and exhaust method according to claim 3, characterized in that, The determining at least one target air suction opening according to at least one of smoke concentration prediction information, smoke diffusion trend prediction information, air flow prediction information, and the air suction range of the at least one air suction opening includes: Dividing the preset space into at least one candidate area; Determining a target area in the at least one candidate area according to at least one of the smoke concentration prediction information, the smoke diffusion trend prediction information, and the air flow prediction information; Determining the overlapping degree between the air suction range of each air suction opening and the at least one target area; Determining the target air suction opening according to the overlapping degree of each air suction opening.

5. A method for sucking and discharging smoke according to claim 3, characterized in that, The determining target control parameters for the air suction device of the target air suction opening according to at least one of smoke concentration prediction information, smoke diffusion trend prediction information, air flow prediction information, and the air suction efficiency of the at least one target air suction opening includes: Determining the working weight of the at least one target air suction opening according to the smoke concentration prediction information, the smoke diffusion trend prediction information, and the air flow prediction information; the working weight is used to represent the importance of the at least one target air suction opening in the current smoke suction and exhaust process; Determining at least one target control parameter according to the air suction efficiency of the at least one target air suction opening and the working weight of the at least one target air suction opening.

6. A method for sucking and discharging smoke according to claim 3, characterized in that, Determining target control parameters for the air suction device for the target air suction opening according to at least one of the smoke concentration prediction information, the smoke diffusion trend prediction information, and the air flow prediction information and the air suction efficiency of the at least one target air suction opening, includes: Determining the energy consumption data of the at least one target air suction opening; Determining at least one target control parameter according to the energy consumption data of the at least one target air suction opening, the smoke concentration prediction information, the smoke diffusion trend prediction information, the air suction efficiency of the at least one target air suction opening, and a preset multi-objective optimization model.

7. A method for sucking and discharging smoke according to claim 3, characterized in that, Determining target control parameters for the air suction device for the target air suction opening according to at least one of the smoke concentration prediction information, the smoke diffusion trend prediction information, and the air flow prediction information and the air suction efficiency of the at least one target air suction opening, includes: Obtaining user preference information corresponding to the preset space; Determining at least one target control parameter according to the user preference information corresponding to the preset space, the smoke concentration prediction information, the smoke diffusion trend prediction information, the air suction efficiency of the at least one target air suction opening, and a preset multi-objective optimization model.

8. A smoke suction and exhaust device, characterized in that, The device includes: A data acquisition module, configured to acquire multi-modal data of a preset space; the multi-modal data includes at least one of smoke concentration data, air flow data, and image data; A data prediction module, configured to input at least one of the smoke concentration data, the air flow data, and the image data into a preset prediction model, so that the preset prediction model outputs smoke distribution prediction information of the preset space after a preset time; An air suction capacity determination module, configured to determine the air suction capacity of at least one air suction opening of the preset space; An air suction execution module, configured to determine at least one target air suction opening according to the smoke distribution prediction information and the air suction capacity of the at least one air suction opening, and control the air suction device of the target air suction opening to perform air suction.

9. An electronic device, characterized in that, Including: A processor, a memory, and a computer program stored on the memory and capable of running on the processor, where when the computer program is executed by the processor, the steps of a smoke suction and exhaust method according to any one of claims 1-7 are implemented.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and when the computer program is executed by the processor, the steps of a smoke suction and exhaust method according to any one of claims 1-7 are implemented.

Images