CN-122024923-A - Method and device for predicting backward flow of oil smoke

CN122024923ACN 122024923 ACN122024923 ACN 122024923ACN-122024923-A

Abstract

The application discloses a method and a device for predicting lampblack backflow, and belongs to the technical field of kitchen appliances. The method comprises the steps of obtaining environmental parameters of a public flue, wherein the environmental parameters comprise at least two of pressure difference between the public flue and a room where a smoke machine is located, target concentration of the public flue and temperature of the public flue, extracting characteristics of various environmental parameters, carrying out characteristic fusion on the characteristics of the various environmental parameters based on time sequence relativity of the various environmental parameters to obtain characteristic vectors, inputting the characteristic vectors into a trained prediction model, and determining the lampblack backflow condition of the public flue according to a prediction result output by the prediction model. The embodiment of the application fully considers the relevance among different environmental parameters and the law of the change of the parameters along with time, so that the feature vector can more accurately reflect the occurrence trend of the oil smoke backward flow, the accuracy of the oil smoke backward flow prediction is improved, the basis can be provided for timely taking precautionary measures, and the pollution of the oil smoke backward flow to the indoor environment is reduced.

Inventors

Request for anonymity
LIANG ZHANGHUA
ZHANG YIHUI
Request for anonymity
Liang Rongxian

Assignees

广东澄一科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260122

Claims (10)

1. The lampblack backflow prediction method is characterized by comprising the following steps of: The method comprises the steps of obtaining environmental parameters of a public flue, wherein the environmental parameters comprise at least two of pressure difference between the public flue and a room where a smoke machine is located, target concentration of the public flue and temperature of the public flue; Extracting the characteristics of various environmental parameters, and carrying out characteristic fusion on the characteristics of various environmental parameters based on the time sequence relativity of the various environmental parameters to obtain characteristic vectors; And inputting the feature vector into a trained prediction model, and determining the lampblack backflow condition of the public flue according to a prediction result output by the prediction model.
2. The method of claim 1, wherein the environmental parameters of the common tunnel are obtained from at least one of: micro pressure sensors are respectively arranged in the common flue and the room where the smoke machine is located, the pressure of the common flue and the pressure in the room where the smoke machine is located are detected according to the micro pressure sensors, and the pressure difference is calculated according to the pressure of the common flue and the pressure in the room where the smoke machine is located; the public flue is provided with a target object monitoring sensor, and the target object concentration of the public flue is obtained through the target object monitoring sensor, wherein the target object concentration comprises one or more of VOCs concentration, particulate matter concentration and fatty acid concentration; And the public flue is provided with a temperature sensor, and the temperature of the public flue is obtained through the temperature sensor.
3. The method of claim 1, wherein said extracting features of each of said environmental parameters comprises: Determining the pressure difference and the change rate of the pressure difference in a target time window as the characteristics of the pressure difference under the condition that the environmental parameters comprise the pressure difference between the public flue and the indoor where the smoke machine is located; determining the integral and/or gradient of the target concentration in a target time window as the characteristic of the target concentration under the condition that the environmental parameter is the target concentration of the public flue; And determining the temperature difference between the temperature of the common flue and the indoor temperature as the characteristic of the temperature of the common flue under the condition that the environmental parameter is the temperature of the common flue.
4. The method according to claim 1, wherein the feature fusion is performed on the features of the environmental parameters based on the time sequence relevance of the environmental parameters to obtain feature vectors, including: calculating time sequence related parameters among the characteristics of the environmental parameters, wherein the time sequence related parameters comprise at least one of time differences of abrupt changes of the environmental parameters and sequence related coefficients of the characteristics of the environmental parameters; and splicing the time sequence related parameters and the characteristics of various environmental parameters to obtain the characteristic vector.
5. The method of claim 1, wherein the predictive model is trained in accordance with: Constructing a training sample set, wherein the training sample set comprises a positive sample and a negative sample, the negative sample is a feature vector corresponding to an environmental parameter collected when lampblack backflow occurs in a public flue, and the positive sample is a feature vector corresponding to the environmental parameter collected in a non-lampblack backflow scene; and training the prediction model by adopting the training sample set.
6. The method of claim 1, wherein the prediction result comprises a soot back-flow probability value; The determining the lampblack backflow condition of the public flue according to the prediction result output by the prediction model comprises the following steps: and under the condition that the oil smoke backflow probability value is larger than a target threshold value, determining that the common flue has an oil smoke backflow risk.
7. The method according to claim 1, wherein the method further comprises: And under the condition that the common flue is determined to have the risk of backward flow of the oil smoke, starting the smoke machine.
8. The method according to claim 1, wherein the method further comprises: receiving feedback information of a user aiming at the prediction result, wherein the feedback information comprises effective prediction and ineffective prediction; Adding the feature vector corresponding to the environment parameter as a positive sample into an incremental training set when the feedback information is effective prediction, and adding the feature vector corresponding to the environment parameter as a negative sample into the incremental training set when the feedback information is ineffective prediction; and performing incremental training on the prediction model by adopting the incremental training set.
9. The method according to claim 1, wherein the method further comprises: counting the lampblack backflow conditions of the public flue at each time interval every day; Determining the same time period with the risk of lampblack backflow in continuous N days as a backflow easy time period; and under the condition that the current time distance is equal to the target time length from the backflow easy-to-occur period, starting the smoke machine.
10. The utility model provides a lampblack backward flow prediction device which characterized in that includes: the system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring environmental parameters of a public flue, wherein the environmental parameters comprise at least two of pressure difference between the public flue and a room where a smoke machine is located, target concentration of the public flue and temperature of the public flue; the extraction module is used for extracting the characteristics of various environmental parameters, and carrying out characteristic fusion on the characteristics of various environmental parameters based on the time sequence relativity of the various environmental parameters to obtain characteristic vectors; And the prediction module is used for inputting the characteristic vector into a trained prediction model, and determining the lampblack backflow condition of the public flue according to a prediction result output by the prediction model.

Description

Method and device for predicting backward flow of oil smoke Technical Field The application belongs to the technical field of kitchen appliances, and particularly relates to a method and a device for predicting lampblack backflow. Background In multi-storey and high-rise residential buildings, in order to realize the intensification of pipeline layout and rationalization of space utilization, the smoke exhaust pipeline of the smoke machine is usually connected to a common flue uniformly arranged in the building, and the smoke generated by each resident is intensively discharged outdoors through the common flue. However, when the range hood is not in operation, external oil smoke is easy to reversely flow into the room through the public flue, so that the oil smoke flows backward. The phenomenon not only pollutes indoor air quality and affects freshness and comfort of living environment, but also brings remarkable negative influence on user experience. At present, the phenomenon of lampblack backflow is not accurately predicted by an effective method. Disclosure of Invention The present application aims to solve at least one of the technical problems existing in the prior art. Therefore, the application provides a method and a device for predicting the backward flow of oil smoke so as to improve the accuracy of the backward flow prediction of the oil smoke. In a first aspect, the present application provides a method for predicting a backward flow of oil smoke, including: The method comprises the steps of obtaining environmental parameters of a public flue, wherein the environmental parameters comprise at least two of pressure difference between the public flue and a room where a smoke machine is located, target concentration of the public flue and temperature of the public flue; extracting the characteristics of various environmental parameters, and carrying out characteristic fusion on the characteristics of various environmental parameters based on the time sequence relativity of the various environmental parameters to obtain characteristic vectors; and inputting the feature vector into a trained prediction model, and determining the lampblack backflow condition of the public flue according to a prediction result output by the prediction model. The oil smoke backward flow prediction method comprises the steps of obtaining environmental parameters of a public flue, wherein the environmental parameters comprise at least two of pressure difference between the public flue and a room where a smoke machine is located, target concentration of the public flue and temperature of the public flue, extracting characteristics of various environmental parameters, carrying out characteristic fusion on the characteristics of the various environmental parameters based on time sequence relevance of the various environmental parameters to obtain characteristic vectors, inputting the characteristic vectors into a trained prediction model, and determining oil smoke backward flow conditions of the public flue according to prediction results output by the prediction model. According to the embodiment of the application, through different environment parameters, the environment characteristics related to the oil smoke backflow can be obtained from different dimensions, the characteristics of various environment parameters are extracted, the characteristic vector is obtained by carrying out characteristic fusion based on the time sequence relevance of the various environment parameters, the relevance among the different environment parameters and the law of the change of the parameters along with time are fully considered, the characteristic vector can more accurately reflect the occurrence trend of the oil smoke backflow, the data analysis and prediction capability of the model are utilized to output the prediction result and determine the oil smoke backflow condition, the accuracy of the oil smoke backflow prediction is improved, the basis can be provided for timely taking precautionary measures, and the pollution of the oil smoke backflow to the indoor environment is reduced. According to one embodiment of the application, the environmental parameters of the common tunnel are obtained according to at least one of the following ways: the pressure of the common flue and the pressure of the room where the smoke machine is positioned are respectively provided with a micro pressure sensor, the pressure of the common flue and the pressure of the room where the smoke machine is positioned are detected according to the micro pressure sensors, calculating to obtain pressure difference according to the pressure of the common flue and the pressure of the room where the smoke machine is positioned; The public flue is provided with a target object monitoring sensor, and the target object concentration of the public flue is obtained through the target object monitoring sensor, wherein the target object concentration comprises one or more of