CN117422982A - Flame vision-based intelligent control method and device for organic waste incineration - Google Patents
Flame vision-based intelligent control method and device for organic waste incineration Download PDFInfo
- Publication number
- CN117422982A CN117422982A CN202311417480.7A CN202311417480A CN117422982A CN 117422982 A CN117422982 A CN 117422982A CN 202311417480 A CN202311417480 A CN 202311417480A CN 117422982 A CN117422982 A CN 117422982A
- Authority
- CN
- China
- Prior art keywords
- incineration
- organic waste
- flame
- hearth
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000010815 organic waste Substances 0.000 title claims abstract description 291
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000004056 waste incineration Methods 0.000 title claims abstract description 13
- 230000000694 effects Effects 0.000 claims abstract description 17
- 238000003062 neural network model Methods 0.000 claims description 99
- 238000012937 correction Methods 0.000 claims description 68
- 230000006870 function Effects 0.000 claims description 42
- 238000012360 testing method Methods 0.000 claims description 39
- 238000012549 training Methods 0.000 claims description 38
- 238000002360 preparation method Methods 0.000 claims description 36
- 238000002372 labelling Methods 0.000 claims description 31
- 238000002485 combustion reaction Methods 0.000 claims description 30
- 238000012795 verification Methods 0.000 claims description 18
- 238000001914 filtration Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 7
- 239000005416 organic matter Substances 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 230000009286 beneficial effect Effects 0.000 abstract description 17
- 239000002699 waste material Substances 0.000 description 15
- 238000010586 diagram Methods 0.000 description 7
- 230000002349 favourable effect Effects 0.000 description 4
- 238000005259 measurement Methods 0.000 description 3
- 239000000779 smoke Substances 0.000 description 3
- 230000001960 triggered effect Effects 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000007499 fusion processing Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000010850 non-combustible waste Substances 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Abstract
The invention relates to the technical field of intelligent control, and discloses an organic waste incineration intelligent control method and device based on flame vision. The method comprises the following steps: collecting a current incineration image when the hearth incinerates the organic waste to be treated in the process of incinerating the organic waste to be treated in the hearth; inputting the current incineration image into a flame recognition model trained in advance until convergence to obtain a recognition result; determining the current burning flame state of the organic waste to be treated based on the identification result, and comparing the current burning flame state with a predetermined target burning flame state to obtain a comparison result; and generating corresponding incineration control parameters of the hearth according to the comparison result, and executing the incineration control operation on the organic waste to be treated based on the incineration control parameters. Therefore, the invention not only can realize the control accuracy of the incineration of the organic waste, but also is beneficial to improving the incineration effect of the organic waste and the incineration safety of the organic waste.
Description
Technical Field
The invention relates to the technical field of intelligent control, in particular to an organic waste incineration intelligent control method and device based on flame vision.
Background
Along with the rapid development of economy and the continuous improvement of the living standard of people, certain wastes are generated in life and production, and how to treat the wastes to ensure the normal running of life and production is a major consideration problem for related personnel. Currently, based on the consideration of the damage degree to the environment and the treatment cost, for some wastes, such as organic wastes which have little damage to the environment and are easy to burn, the most common treatment mode is also a burning mode, and the treatment mode mainly depends on human experience and subjective judgment, so that the conditions of inaccurate judgment and unstable burning effect are easily caused, and accidents such as personnel burn and the like are more seriously caused.
In order to solve the above-mentioned problems of the manual control of the incineration of organic waste, the prior art also proposes a treatment scheme for determining the incineration state of organic waste in a manner of combining the measurement value of a measuring element (such as a sensor) with manual observation, and further controlling the furnace for incinerating organic waste, and the treatment scheme introduces the measurement value of the measuring element, but the determination of the incineration state based on the measurement value of the measuring element has a certain hysteresis, and still requires manual observation, and in the aspect of the incineration control of organic waste, the problem of poor control accuracy and thus poor incineration effect still exists.
Therefore, how to provide an intelligent control manner for organic waste incineration to improve the control accuracy of the organic waste incineration and further improve the incineration effect of the organic waste is important.
Disclosure of Invention
The invention provides an intelligent control method and device for organic matter incineration based on flame vision, which can realize the control accuracy of organic waste incineration, thereby being beneficial to improving the incineration effect of organic waste.
In order to solve the technical problems, the first aspect of the invention discloses an intelligent control method for organic matter incineration based on flame vision, which comprises the following steps:
collecting a current incineration image of the organic waste to be treated when the hearth incinerates the organic waste to be treated in the process of incinerating the organic waste to be treated in the hearth;
inputting the current incineration image into a flame recognition model trained in advance until convergence, and obtaining a recognition result of the flame recognition model;
determining the current incineration flame state of the organic waste to be treated based on the identification result, and comparing the current incineration flame state with a predetermined target incineration flame state to obtain a comparison result;
And generating an incineration control parameter corresponding to the hearth according to the comparison result, and executing the incineration control operation for the organic waste to be treated based on the incineration control parameter.
As an alternative embodiment, in the first aspect of the present invention, the flame identification model is trained by:
acquiring a plurality of historical incineration image information acquired by an image acquisition device when the organic waste is historically incinerated for a plurality of times;
constructing an initial data set based on all the historical incineration image information, and executing data expansion operation on the initial data set to obtain an expanded data set;
performing combustion state labeling operation on the extended data set based on flame parameters in each incineration image information of the extended data set to obtain a target data set with labeling results, wherein the labeling results of each incineration image information of the extended data set comprise combustion states corresponding to the flame parameters in each incineration image information; wherein the flame parameters include flame size and/or flame color, the flame size including flame size and/or flame force size;
dividing the target data set into a training data set, a verification data set and a test data set according to the determined dividing proportion;
Constructing an initial neural network model, setting initial model parameters of the neural network model, inputting the training data set and the verification data set into the neural network model to obtain a prediction result of the neural network model, determining a loss function according to the prediction result of the neural network model and a corresponding labeling result, reversely transmitting the loss function to each layer of network of the neural network model, calculating the partial conductance of each layer of network parameters of the neural network model by errors of each layer of network, updating network parameters of each layer of network in the neural network model according to a preset gradient descent algorithm, repeatedly executing the network parameters of each layer of network in the neural network model, inputting the training data set and the verification data set into the neural network model to obtain the prediction result of the neural network model, determining a loss function according to the prediction result of the neural network model and the corresponding labeling result, reversely transmitting the loss function to each layer of network of the neural network model, calculating the partial conductance of each layer of network parameters of the neural network model by errors of each layer of network model until the neural network parameters of each layer of network meet the loss function of the neural network model according to the preset gradient descent algorithm, and repeatedly executing the step of updating the neural network model;
And testing the neural network model with the loss function meeting the preset loss condition based on the test data set to obtain a test result, and determining the neural network model with the loss function meeting the preset loss condition as a flame recognition model trained to be converged when the test result indicates that the neural network model with the loss function meeting the preset loss condition passes the test.
In an optional implementation manner, in a first aspect of the present invention, the generating, according to the comparison result, the incineration control parameter corresponding to the furnace chamber includes:
determining a factor to be corrected corresponding to the organic waste to be treated burned by the hearth and a correction direction of the factor to be corrected according to the comparison result;
determining target equipment to be controlled according to the factors to be corrected and the correction directions of the factors to be corrected;
and generating control parameters of the target equipment according to the comparison result, the factors to be corrected and the correction directions of the factors to be corrected, and taking the control parameters as the corresponding incineration control parameters of the hearth.
As an optional implementation manner, in the first aspect of the present invention, the determining, according to the factor to be corrected and the correction direction of the factor to be corrected, the target device to be controlled includes:
Determining an incineration control parameter range of at least one incineration control device corresponding to the hearth;
judging whether target incineration control equipment capable of meeting the correction requirement of the factors to be corrected exists in all the incineration control equipment according to the incineration control parameter range of each incineration control equipment, the factors to be corrected and the correction directions of the factors to be corrected;
when the judgment result is yes, determining target incineration control equipment capable of meeting the correction requirement of the factor to be corrected in all the incineration control equipment as target equipment to be controlled;
the correction requirement at least comprises a fixed correction requirement corresponding to the factor to be corrected, and the fixed correction requirement comprises a correction safety requirement and a correction effect requirement.
As an alternative embodiment, in the first aspect of the present invention, the method further includes:
determining the organic waste to be treated which is currently required to be incinerated in the hearth;
the determining the organic waste to be treated, which is required to be incinerated currently, of the hearth comprises the following steps:
determining all the organic wastes to be screened, and identifying multi-dimensional information of all the organic wastes to be screened, wherein the multi-dimensional information of all the organic wastes to be screened comprises type information, processing priority, stacking time, stacking quantity and hazard degree to surrounding environment;
According to the type information in the multidimensional information of all the organic wastes to be screened, performing filtering operation on the organic wastes which do not meet the incineration requirement in all the organic wastes to be screened so as to update all the organic wastes to be screened;
and determining the organic waste to be treated, which is currently required to be incinerated in the hearth, according to all the updated organic waste to be screened.
In a first aspect of the present invention, the determining the organic waste to be treated, which is currently required to be incinerated in the furnace, according to all the updated organic waste to be screened includes:
according to preset grouping conditions and updated multi-dimensional information of all the organic wastes to be screened, grouping the updated all the organic wastes to be screened to obtain a plurality of organic waste groups;
and screening one organic waste group from all the organic waste groups according to the current state information of the hearth, the multidimensional information of each organic waste group and the current scene information of the scene of the hearth, and determining the organic waste group in the one organic waste group as the organic waste to be treated, which is currently required to be incinerated in the hearth.
As an alternative embodiment, in the first aspect of the present invention, the method further includes:
after determining the organic waste to be incinerated which is currently required by the hearth, determining the alternative organic waste which is required by the hearth to be incinerated after the organic waste to be incinerated is incinerated;
in the process of incinerating the organic waste to be treated by the hearth, determining an incineration preparation condition of the alternative organic waste and an incineration preparation parameter corresponding to the incineration preparation condition based on the incinerated condition, the current incineration condition and the expected incineration condition of the organic waste to be treated;
and in the process of incinerating the organic waste to be treated in the hearth, if the current condition meets the incineration preparation condition, controlling the incineration preparation equipment corresponding to the hearth to execute the pre-incineration preparation operation corresponding to the alternative organic waste according to the incineration preparation parameter.
The invention discloses an intelligent control device for organic matter incineration based on flame vision, which comprises the following components:
the acquisition module is used for acquiring a current incineration image when the hearth incinerates the organic waste to be treated in the process of incinerating the organic waste to be treated in the hearth;
The recognition module is used for inputting the current incineration image into a flame recognition model trained to be converged in advance to obtain a recognition result of the flame recognition model, and determining the current incineration flame state of the organic waste to be treated based on the recognition result;
the comparison module is used for comparing the current incineration flame state with a predetermined target incineration flame state to obtain a comparison result;
and the incineration control module is used for generating the incineration control parameters corresponding to the hearth according to the comparison result and executing the incineration control operation on the organic waste to be treated based on the incineration control parameters.
As an alternative embodiment, in the second aspect of the present invention, the flame identification model is trained by:
acquiring a plurality of historical incineration image information acquired by an image acquisition device when the organic waste is historically incinerated for a plurality of times;
constructing an initial data set based on all the historical incineration image information, and executing data expansion operation on the initial data set to obtain an expanded data set;
performing combustion state labeling operation on the extended data set based on flame parameters in each incineration image information of the extended data set to obtain a target data set with labeling results, wherein the labeling results of each incineration image information of the extended data set comprise combustion states corresponding to the flame parameters in each incineration image information; wherein the flame parameters include flame size and/or flame color, the flame size including flame size and/or flame force size;
Dividing the target data set into a training data set, a verification data set and a test data set according to the determined dividing proportion;
constructing an initial neural network model, setting initial model parameters of the neural network model, inputting the training data set and the verification data set into the neural network model to obtain a prediction result of the neural network model, determining a loss function according to the prediction result of the neural network model and a corresponding labeling result, reversely transmitting the loss function to each layer of network of the neural network model, calculating the partial conductance of each layer of network parameters of the neural network model by errors of each layer of network, updating network parameters of each layer of network in the neural network model according to a preset gradient descent algorithm, repeatedly executing the network parameters of each layer of network in the neural network model, inputting the training data set and the verification data set into the neural network model to obtain the prediction result of the neural network model, determining a loss function according to the prediction result of the neural network model and the corresponding labeling result, reversely transmitting the loss function to each layer of network of the neural network model, calculating the partial conductance of each layer of network parameters of the neural network model by errors of each layer of network model until the neural network parameters of each layer of network meet the loss function of the neural network model according to the preset gradient descent algorithm, and repeatedly executing the step of updating the neural network model;
And testing the neural network model with the loss function meeting the preset loss condition based on the test data set to obtain a test result, and determining the neural network model with the loss function meeting the preset loss condition as a flame recognition model trained to be converged when the test result indicates that the neural network model with the loss function meeting the preset loss condition passes the test.
In a second aspect of the present invention, as an optional implementation manner, the specific manner of generating, by the incineration control module, the incineration control parameters corresponding to the furnace chamber according to the comparison result includes:
determining a factor to be corrected corresponding to the organic waste to be treated burned by the hearth and a correction direction of the factor to be corrected according to the comparison result;
determining target equipment to be controlled according to the factors to be corrected and the correction directions of the factors to be corrected;
and generating control parameters of the target equipment according to the comparison result, the factors to be corrected and the correction directions of the factors to be corrected, and taking the control parameters as the corresponding incineration control parameters of the hearth.
As an optional implementation manner, in the second aspect of the present invention, the determining, by the incineration control module, a specific manner of the target device to be controlled according to the factor to be corrected and the correction direction of the factor to be corrected includes:
Determining an incineration control parameter range of at least one incineration control device corresponding to the hearth;
judging whether target incineration control equipment capable of meeting the correction requirement of the factors to be corrected exists in all the incineration control equipment according to the incineration control parameter range of each incineration control equipment, the factors to be corrected and the correction directions of the factors to be corrected;
when the judgment result is yes, determining target incineration control equipment capable of meeting the correction requirement of the factor to be corrected in all the incineration control equipment as target equipment to be controlled;
the correction requirement at least comprises a fixed correction requirement corresponding to the factor to be corrected, and the fixed correction requirement comprises a correction safety requirement and a correction effect requirement.
As an alternative embodiment, in the second aspect of the present invention, the apparatus further includes:
the object determining module is used for determining the organic waste to be processed which is currently required to be incinerated in the hearth;
and the specific mode of determining the organic waste to be treated, which is required to be incinerated currently by the hearth, by the object determining module comprises the following steps:
Determining all the organic wastes to be screened, and identifying multi-dimensional information of all the organic wastes to be screened, wherein the multi-dimensional information of all the organic wastes to be screened comprises type information, processing priority, stacking time, stacking quantity and hazard degree to surrounding environment;
according to the type information in the multidimensional information of all the organic wastes to be screened, performing filtering operation on the organic wastes which do not meet the incineration requirement in all the organic wastes to be screened so as to update all the organic wastes to be screened;
and determining the organic waste to be treated, which is currently required to be incinerated in the hearth, according to all the updated organic waste to be screened.
In a second aspect of the present invention, as an optional implementation manner, the specific manner of determining, by the object determining module, the organic waste to be treated that is currently required to be incinerated in the furnace according to all the updated organic waste to be screened includes:
according to preset grouping conditions and updated multi-dimensional information of all the organic wastes to be screened, grouping the updated all the organic wastes to be screened to obtain a plurality of organic waste groups;
And screening one organic waste group from all the organic waste groups according to the current state information of the hearth, the multidimensional information of each organic waste group and the current scene information of the scene of the hearth, and determining the organic waste group in the one organic waste group as the organic waste to be treated, which is currently required to be incinerated in the hearth.
As an alternative embodiment, in the second aspect of the present invention, the apparatus further includes:
the incineration preparation control module is used for determining all the organic wastes to be screened, identifying the multi-dimensional information of all the organic wastes to be screened, wherein the multi-dimensional information of all the organic wastes to be screened comprises type information, processing priority, stacking time, stacking quantity and hazard degree to the surrounding environment; according to the type information in the multidimensional information of all the organic wastes to be screened, filtering the organic wastes which do not meet the incineration requirement in all the organic wastes to be screened so as to update all the organic wastes to be screened; and determining the organic waste to be treated, which is currently required to be incinerated in the hearth, according to all the updated organic waste to be screened.
The third aspect of the invention discloses another intelligent control device for organic matter incineration based on flame vision, which comprises:
a memory storing executable program code;
a processor coupled to the memory;
the processor invokes the executable program codes stored in the memory to execute the flame vision-based intelligent control method for organic matter incineration disclosed in the first aspect of the invention.
A fourth aspect of the present invention discloses a computer storage medium storing computer instructions for executing the file deduplication implementation method disclosed in the first aspect of the present invention when the computer instructions are called.
Compared with the prior art, the invention has the following beneficial effects:
in the invention, in the process of burning the organic waste to be treated in the hearth, the current burning image of the organic waste to be treated in the hearth is collected; inputting the current incineration image into a flame recognition model trained to be converged in advance to obtain a recognition result of the flame recognition model; determining the current burning flame state of the organic waste to be treated based on the identification result, and comparing the current burning flame state with a predetermined target burning flame state to obtain a comparison result; and generating corresponding incineration control parameters of the hearth according to the comparison result, and executing the incineration control operation on the organic waste to be treated based on the incineration control parameters. Therefore, the invention can automatically identify and intelligently analyze the incineration image when the organic waste is incinerated in the hearth based on the artificial intelligence model to obtain the current incineration flame state, and the self-adaptive adjustment of the incineration control parameters of the organic waste is performed on the hearth based on the comparison result of the current incineration flame state and the target incineration flame state, thereby not only realizing the control accuracy of the incineration of the organic waste, but also being beneficial to improving the incineration effect of the organic waste, and being beneficial to improving the incineration safety of the organic waste.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of intelligent control of organic waste incineration based on flame vision, which is disclosed in the embodiment of the invention;
FIG. 2 is a schematic diagram of a training and testing implementation flow of a flame identification model according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an intelligent control device for burning organic waste based on flame vision according to the embodiment of the invention;
FIG. 4 is a schematic structural diagram of another intelligent control device for burning organic waste based on flame vision according to the embodiment of the invention;
fig. 5 is a schematic structural diagram of another intelligent control device for burning organic waste based on flame vision according to the embodiment of the invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The terms first, second and the like in the description and in the claims and in the above-described figures are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, apparatus, article, or article that comprises a list of steps or elements is not limited to only those listed but may optionally include other steps or elements not listed or inherent to such process, method, article, or article.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the invention. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
The invention discloses an intelligent control method and device for burning organic wastes based on flame vision, which can automatically identify and intelligently analyze burning images of burning organic wastes in a hearth based on an artificial intelligent model to obtain the current burning flame state, and realize the self-adaptive adjustment of burning control parameters of burning the organic wastes in the hearth based on the comparison result of the current burning flame state and the target burning flame state, thereby not only realizing the control accuracy of burning the organic wastes, but also being beneficial to improving the burning effect of the organic wastes and the burning safety of the organic wastes. The following will describe in detail.
Example 1
Referring to fig. 1, fig. 1 is a schematic flow chart of intelligent control of burning organic waste based on flame vision according to an embodiment of the invention. The method described in fig. 1 can be used in an incineration control device, which is used for realizing intelligent control on organic waste incineration in a furnace, and optionally, the incineration control device can be integrated in a local control device or a cloud control device, and the embodiment of the invention is not limited. As shown in fig. 1, the method may include:
101. in the process of burning the organic waste to be treated in the hearth, the current burning image of the organic waste to be treated in the hearth is collected.
In the embodiment of the invention, a plurality of corresponding image acquisition devices (such as cameras) are arranged in the hearth, and different image acquisition devices are used for acquiring incineration images at different angles. In practical application, before or during incineration, the incineration control device can determine the image acquisition device needing to acquire images according to the conditions (such as stacking height, stacking area, stacking volume and the like) of the organic waste to be treated, so that the resource utilization rate of the image acquisition device is improved. After the image acquisition device acquires the incineration image, the acquired incineration image can be transmitted to the incineration control device for storage and flame identification, and the image acquisition device can further store the incineration image in a storage space corresponding to the image acquisition device.
Further, after the plurality of image collecting devices collect the incineration images of different angles, the fusion processing may be performed on the incineration images of different angles to obtain a fused incineration image, which is used as an incineration image required to be input into the flame recognition model, or the incineration image of a key angle may be selected from the incineration images of different angles to be used as an incineration image required to be input into the flame recognition model, or the incineration image with the image quality meeting the preset requirements (such as the definition requirement and no shielding) may be selected from the incineration images of different angles to be used as an incineration image required to be input into the flame recognition model, or the incineration image determined based on the combination of at least two modes in the foregoing manner may be used as an incineration image required to be input into the flame recognition model, for example, the multiple incineration images of the key angle may be selected from the incineration images of different angles, and then the fusion processing may be performed on the incineration images of the key angle to obtain the fused incineration image, which is required to be input into the flame recognition model. Therefore, the embodiment of the invention provides a plurality of ways for determining the incineration image needing to be input into the flame identification model, which is beneficial to improving the determination flexibility of the incineration image, is beneficial to self-adaptively selecting a proper determination way based on different incineration scenes or different incineration demands, and improves the matching degree of the incineration image and different incineration scenes or different incineration demands.
102. And inputting the current incineration image into a flame recognition model trained to be converged in advance to obtain a recognition result of the flame recognition model.
The flame recognition model is obtained based on a large number of sample training, further, after training to convergence, the flame recognition model trained to convergence can be verified, and after verification, the flame recognition model trained to convergence is used as the flame recognition model actually required, so that the accuracy and reliability of the flame recognition model can be improved. Specifically, after inputting the current incineration image into the flame recognition model trained to be converged in advance, the flame recognition model may perform the following operations:
performing image segmentation operation on the current incineration image to segment an area image of an area where flame is located from the current incineration image;
and performing intelligent recognition processing on flames in the regional image to obtain flame conditions in the region.
103. And determining the current burning flame state of the organic waste to be treated based on the identification result, and comparing the current burning flame state with a predetermined target burning flame state to obtain a comparison result.
The target incineration flame state can be intelligently determined based on the related information corresponding to the organic waste to be treated, or can be preset. Furthermore, when the target incineration flame state is intelligently determined, the incineration control device can be determined based on the waste types of the organic waste to be treated, the waste proportion of different types, the incineration duration, the actual incineration requirement and the like, so that the determination flexibility of the target incineration flame state is improved. For example, the target incineration flame state may be a full combustion state.
104. And generating corresponding incineration control parameters of the hearth according to the comparison result, and executing the incineration control operation on the organic waste to be treated based on the incineration control parameters.
In the embodiment of the present invention, for example, when the current incineration flame state is an insufficient combustion state and the target incineration flame state is a sufficient combustion state, the incineration control device may generate the incineration control parameters corresponding to the hearth according to the flame state difference condition (such as the flame color difference condition, the flame size difference condition, etc.) between the current incineration flame state and the target incineration flame state.
In the embodiment of the invention, the incineration control parameters corresponding to the hearth can comprise the incineration control parameters of the hearth, can also comprise the incineration control parameters of the incineration auxiliary equipment corresponding to the hearth, and can also be a combination of the two. Further, the incineration auxiliary equipment corresponding to the hearth can be started incineration auxiliary equipment or incineration auxiliary equipment needing to be newly started, and the embodiment of the invention is not limited. For example, if the combustion is insufficient due to low flow rate of air in the furnace, the combustion control parameter of the furnace may be to open the air inlet, the combustion auxiliary device corresponding to the furnace may be a blower, and the combustion control parameter of the blower may include the output power of the blower.
Therefore, the embodiment of the invention can automatically identify and intelligently analyze the incineration image when the organic waste is incinerated in the hearth based on the artificial intelligent model to obtain the current incineration flame state, and the incineration control parameters of the organic waste is incinerated in the hearth based on the comparison result of the current incineration flame state and the target incineration flame state are adaptively adjusted, so that the control accuracy of the incineration of the organic waste can be realized without depending on artificial observation and judgment, the incineration effect of the organic waste is improved, and the incineration safety of the organic waste is improved.
In an alternative embodiment, the model training and testing process for obtaining the flame identification model may refer to fig. 2, and fig. 2 is a schematic diagram of an implementation flow of training and testing of the flame identification model according to an embodiment of the present invention. Specifically, as shown in fig. 2, the training and testing process of the flame identification model may include the following operations:
201. and acquiring a plurality of historical incineration image information acquired by the image acquisition device when the organic waste is historically incinerated for a plurality of times.
When the organic waste is burnt in a history way, a plurality of pieces of history burning image information acquired by the image acquisition device can be stored in the flame form database, and when model training is needed, the corresponding burning image information can be read from the flame form database.
202. And constructing an initial data set based on all the historical incineration image information, and executing data expansion operation on the initial data set to obtain an expanded data set.
It should be noted that, if the data size (may also be referred to as the sample size) of the initial data set is sufficient, the data expansion operation is not required to be performed, that is: the data expansion operation is preferable, the data amount can be increased through the data expansion operation, and the learning objects of the neural network model can be expanded to a certain extent, so that the neural network model learns flames in various forms as much as possible, and the training accuracy of the neural network model, the flame identification accuracy in the application process and the comprehensiveness are improved.
203. And performing combustion state labeling operation on the extended data set based on flame parameters in each incineration image information of the extended data set to obtain a target data set with a labeling result.
The labeling result of each incineration image information of the expansion data set comprises a combustion state corresponding to flame parameters in each incineration image information. Optionally, the flame parameters may include flame size and/or flame color, and still further optionally, the flame size may include flame size and/or flame force size.
In the embodiment of the invention, the combustion state labeling operation performed on the extended data set can be specifically performed by a corresponding labeling program, or performed by a corresponding labeling person meeting the requirements, or a combination of the two. Furthermore, after the labeling is completed, the labeling result can be checked by corresponding qualified check personnel so as to ensure the accuracy and reliability of the labeling result.
In the embodiment of the invention, when the combustion state labeling operation is executed, the analysis and judgment of the combustion state are mainly performed depending on the flame condition in the incineration image information, and the corresponding combustion states are labeled according to the analysis and judgment results, for example, the combustion states (or the incineration states) with a plurality of levels are labeled, wherein the higher the level is, the more complete the combustion is, the lower the level is, and the less complete the combustion is.
It should be noted that the level of the combustion state may be classified according to the requirements, for example, into two combustion levels including full combustion and insufficient combustion.
204. And dividing the target data set into a training data set, a verification data set and a test data set according to the determined dividing proportion.
In some alternative embodiments, the target data set may be divided into only a training data set and a validation data set, the test data set being optional. And when the data sets are divided, the training data set and the verification data set occupy a larger area than the test data set, and if the data amount in the target data set is insufficient, partial data overlap can exist between the data sets. However, in practical applications, if the data volume in the target data set is insufficient, the data set expansion operation may be further considered.
It should be noted that, during the actual training process, data in the test data set may be added to the training data set to increase the number and diversity of the training data sets.
205. And constructing an initial neural network model and setting initial model parameters of the neural network model.
The initial neural network model is any artificial intelligent model capable of realizing image classification segmentation and image recognition.
206. Inputting the training data set and the verification data set into the neural network model to obtain a prediction result of the neural network model, and comparing the prediction result of the neural network model with a corresponding labeling result to obtain a training comparison result.
207. And determining a loss function corresponding to the neural network model based on the training comparison result, and judging whether the corresponding loss function meets a preset loss condition.
When the determination result in step 207 is no, the execution of step 208 may be triggered, that is: continuing training the neural network model after adjusting the model parameters; when the judgment result in step 207 is yes, the neural network model with the corresponding loss function satisfying the preset loss condition may be directly determined as the final required flame identification model, or step 209 may be triggered, and preferably the latter may be triggered, so that the accuracy and reliability of the final obtained flame identification model may be ensured.
In the embodiment of the present invention, the loss function meeting the preset loss condition may specifically be that the loss value is smaller than or equal to the preset loss threshold.
208. And updating model parameters of the neural network model according to the training comparison result and a preset gradient descent algorithm.
In the embodiment of the present invention, after the execution of step 208 is completed, steps 206 to 207 are repeatedly executed until the loss function corresponding to the neural network model meets the preset loss condition.
In the embodiment of the present invention, updating the model parameters of the neural network model according to the training comparison result and the preset gradient descent algorithm may include:
Determining a loss function according to the prediction result and the corresponding labeling result of the neural network model, reversely transmitting the loss function to each layer of network of the neural network model, calculating the partial guide of the error of each layer of network of the neural network model to each layer of network parameter, and updating the network parameter of each layer of network in the neural network model according to a preset gradient descent algorithm. Wherein the network parameters of each layer of network include weights and paraphrasing.
209. And testing the neural network model with the loss function meeting the preset loss condition based on the test data set to obtain a test result, and determining the neural network model with the loss function meeting the preset loss condition as a flame recognition model trained until convergence when the test result indicates that the corresponding neural network model with the loss function meeting the preset loss condition passes the test.
Therefore, the optional embodiment can realize training of the neural network model based on the training data set and the verification data set so as to train to obtain the flame identification model, and the intelligent identification mode based on computer vision is beneficial to improving the comprehensiveness and accuracy of the incineration flame identification. In addition, after training, the test of the flame identification model can be further realized, which is beneficial to further improving the identification accuracy and the identification reliability of the finally obtained flame identification model.
In still another optional embodiment, generating the incineration control parameters corresponding to the hearth according to the comparison result may include:
determining a factor to be corrected corresponding to the organic waste to be treated in the incineration of the hearth according to the comparison result, and determining the correction direction of the factor to be corrected;
determining target equipment to be controlled according to the factors to be corrected and the correction directions of the factors to be corrected;
and generating control parameters of target equipment according to the comparison result, the factors to be corrected and the correction directions of the factors to be corrected, and taking the control parameters as corresponding incineration control parameters of the hearth.
Optionally, the factor to be corrected may have a plurality of influencing factors, and when determining the target device to be controlled, the factor to be corrected may be further matched with the corresponding target device based on the plurality of influencing factors existing in the factor to be corrected, which is not only beneficial to improving the accuracy of the target device, but also improves the matching degree between the target device and the actual correction requirement.
Therefore, the optional embodiment can also adaptively determine the target equipment to be controlled based on the determined factor to be corrected and the correction direction of the factor to be corrected, and further realize adaptive determination of the incineration control parameters based on the multidimensional information, thereby being beneficial to improving the generation efficiency, the generation accuracy and the generation reliability of the incineration control parameters.
In this optional embodiment, further optionally, determining the target device to be controlled according to the factor to be corrected and the correction direction of the factor to be corrected may include:
determining at least one incineration control device corresponding to the hearth and an incineration control parameter range of each incineration control device;
judging whether target incineration control equipment capable of meeting the correction requirement of the factors to be corrected exists in all the incineration control equipment according to the incineration control parameter range, the factors to be corrected and the correction direction of the factors to be corrected of each incineration control equipment;
and when the judgment result is yes, determining target incineration control equipment which can meet the correction requirement of the factor to be corrected in all the incineration control equipment as target equipment to be controlled.
Alternatively, the target device to be controlled may be one or a combination of more of an induced draft fan, a blower, a grate, a pusher, and the like. Still further, after determining the target device to be controlled, it may further determine whether there is an incineration control device linked with the target device to be controlled, and if so, determine the incineration control device linked with the target device to be controlled as the target device to be controlled, which is favorable to improving the comprehensiveness and accuracy of the determined target device to be controlled, and further is favorable to improving the accuracy of the incineration control. For example, if the initially determined target device to be controlled is a blower to supplement air to increase oxygen content and thus increase combustion sufficiency, but sufficient combustion may generate a large amount of smoke in a period of time, the incineration control device associated with the target device to be controlled may be a device that actively absorbs smoke, such as an exhaust fan, to exhaust the smoke.
The correction requirement at least comprises a fixed correction requirement corresponding to the factor to be corrected, the fixed correction requirement comprises a correction safety requirement and a correction effect requirement, and further, the correction effect requirement at least comprises a corrected incineration flame state requirement, and further comprises a correction efficiency requirement and/or a correction cost requirement.
It can be seen that this alternative embodiment also enables, when determining the target device to be controlled, preferably the target incineration control device that meets the correction requirements, facilitating the rational utilization of the incineration control device in case the incineration control requirements are met.
In yet another alternative embodiment, the method may further comprise:
and determining the organic waste to be treated, which is currently required to be incinerated in the hearth.
The determining the organic waste to be treated, which is required to be incinerated currently, of the furnace chamber may include:
determining all the organic wastes to be screened, and identifying the multi-dimensional information of all the organic wastes to be screened, wherein the multi-dimensional information of all the organic wastes to be screened comprises type information, processing priority, stacking time, stacking quantity and hazard degree to the surrounding environment;
according to the type information in the multidimensional information of all the organic wastes to be screened, performing filtering operation on the organic wastes which do not meet the incineration requirement in all the organic wastes to be screened so as to update all the organic wastes to be screened;
And determining the organic waste to be treated which is currently required to be incinerated in the hearth according to all the updated organic waste to be screened.
It can be seen that, before the incineration operation is performed, the optional embodiment can also filter the organic waste to be screened (such as filtering out non-combustible waste, waste with a serious negative influence on the current scene, etc.), and then further determine the organic waste to be treated, which is currently required to be incinerated, after filtering, so as to be beneficial to reducing the influence of the waste which does not meet the incineration requirement on the overall incineration effect.
In this optional embodiment, further optionally, determining, according to all the updated organic wastes to be screened, the organic waste to be treated that is currently required to be incinerated in the furnace chamber may include:
according to preset grouping conditions and updated multidimensional information of all the organic wastes to be screened, grouping the updated organic wastes to obtain a plurality of organic waste groups;
and screening one organic waste group from all the organic waste groups according to the current state information of the hearth, the multidimensional information of each organic waste group and the current scene information of the scene of the hearth, and determining the organic waste group in one organic waste group as the organic waste to be treated, which is currently required to be incinerated in the hearth.
It can be seen that, this optional embodiment can also group all the organic wastes to be screened after the filtration (for example, divide the wastes with small differences in fire points into one group, or divide the wastes with the same kind into one group, etc.), then based on the grouping result, realize the determination of the organic wastes to be treated that need to be incinerated currently, for example, select the organic waste group that affects the current scene information least when incinerating, or select the organic waste group that has the shortest incineration duration, or select the organic waste group that has the highest incineration emergency degree, be favorable to improving the matching degree of the determined organic wastes to be treated that need to be incinerated currently and the current incineration demand, and the grouping mode can also realize the rational utilization of the incineration control resources, reduce the waste of unnecessary incineration control resources. In addition, the determination flexibility of the organic waste to be treated, which is required to be incinerated at present, of the hearth can be improved.
In yet another alternative embodiment, the method may further comprise:
after determining the organic waste to be incinerated which is currently required to be incinerated in the hearth, determining the alternative organic waste to be incinerated in the hearth after the organic waste to be incinerated is incinerated, for example, selecting the organic waste with the highest matching degree in the matching degree of the residual incineration resources of the hearth after the organic waste to be incinerated is incinerated and the residual organic waste;
In the process of incinerating the organic waste to be treated in the hearth, determining incineration preparation conditions of the alternative organic waste and incineration preparation parameters corresponding to the incineration preparation conditions based on the incinerated condition, the current incineration condition and the expected incineration condition of the organic waste to be treated;
in the process of incinerating the organic waste to be treated in the hearth, if the current condition meets the incineration preparation condition, controlling the incineration preparation equipment corresponding to the hearth to execute the pre-incineration preparation operation corresponding to the alternative organic waste according to the incineration preparation parameter.
Therefore, after determining the organic waste to be incinerated which is required to be incinerated currently, the alternative organic waste to be incinerated after the organic waste to be incinerated is incinerated, the incineration preparation conditions of the alternative organic waste and the intelligent determination of the incineration preparation parameters corresponding to the incineration preparation conditions of the hearth can be realized, the accuracy before the incineration of the object to be incinerated of the next wave is improved in advance in the process of incinerating the organic waste to be incinerated by the hearth, and the incineration efficiency of the object to be incinerated of the next wave is improved.
Example two
Referring to fig. 3, fig. 3 is a schematic structural diagram of an intelligent control device for burning organic waste based on flame vision according to an embodiment of the present invention. The device described in fig. 3 is used for realizing intelligent control on the organic waste burned in the furnace, and optionally, the device can be integrated in a local control device or a cloud control device, and the embodiment of the invention is not limited. As shown in fig. 3, the apparatus may include:
The acquisition module 301 is configured to acquire a current incineration image when the organic waste to be treated is incinerated in the hearth in a process of incinerating the organic waste to be treated in the hearth;
the recognition module 302 is configured to input a current incineration image into a flame recognition model trained to converge in advance, obtain a recognition result of the flame recognition model, and determine a current incineration flame state of the organic waste to be treated based on the recognition result;
the comparison module 303 is configured to compare the current incineration flame state with a predetermined target incineration flame state to obtain a comparison result;
the incineration control module 304 is configured to generate an incineration control parameter corresponding to the furnace according to the comparison result, and perform an incineration control operation for the organic waste to be treated based on the incineration control parameter.
Therefore, the device described in fig. 3 can automatically identify and intelligently analyze the incineration image when the organic waste is incinerated in the hearth based on the artificial intelligent model to obtain the current incineration flame state, and the incineration control parameters of the organic waste is incinerated in the hearth based on the comparison result of the current incineration flame state and the target incineration flame state are adaptively adjusted, so that the control accuracy of the incineration of the organic waste can be realized without depending on artificial observation and judgment, the incineration effect of the organic waste is improved, and the incineration safety of the organic waste is improved.
In an alternative embodiment, the flame identification model is trained by:
acquiring a plurality of historical incineration image information acquired by an image acquisition device when the organic waste is historically incinerated for a plurality of times;
constructing an initial data set based on all the historical incineration image information, and executing data expansion operation on the initial data set to obtain an expanded data set;
performing combustion state labeling operation on the extended data set based on flame parameters in each incineration image information of the extended data set to obtain a target data set with labeling results, wherein the labeling results of each incineration image information of the extended data set comprise combustion states corresponding to the flame parameters in each incineration image information; wherein the flame parameters include flame size and/or flame color, the flame size including flame size and/or flame intensity size;
dividing the target data set into a training data set, a verification data set and a test data set according to the determined dividing proportion;
constructing an initial neural network model, setting initial model parameters of the neural network model, inputting a training data set and a verification data set into the neural network model to obtain a prediction result of the neural network model, determining a loss function according to the prediction result of the neural network model and a corresponding labeling result, reversely transmitting the loss function to each layer of network of the neural network model, calculating the partial guide of errors of each layer of network of the neural network model to each layer of network parameters, updating the network parameters of each layer of network of the neural network model according to a preset gradient descent algorithm, repeatedly executing the steps of inputting the training data set and the verification data set into the neural network model to obtain the prediction result of the neural network model, determining the loss function according to the prediction result of the neural network model and the corresponding labeling result, reversely transmitting the loss function to each layer of network of the neural network model, calculating the partial guide of errors of each layer of network of the neural network model to each layer of network parameters of the neural network, and updating the network parameters of each layer of network of the neural network model according to the preset gradient descent algorithm until the loss function corresponding to the neural network model meets preset loss conditions;
And testing the neural network model with the loss function meeting the preset loss condition based on the test data set to obtain a test result, and determining the neural network model with the loss function meeting the preset loss condition as a flame recognition model trained until convergence when the test result indicates that the corresponding neural network model with the loss function meeting the preset loss condition passes the test.
Therefore, the optional embodiment can also realize training of the neural network model based on the training data set and the verification data set so as to train to obtain the flame identification model, and the intelligent identification mode based on computer vision is beneficial to improving the comprehensiveness and accuracy of the incineration flame identification. In addition, after training, the test of the flame identification model can be further realized, which is beneficial to further improving the identification accuracy and the identification reliability of the finally obtained flame identification model.
In another alternative embodiment, the specific manner of generating the incineration control parameters corresponding to the hearth by the incineration control module 304 according to the comparison result includes:
determining a factor to be corrected corresponding to the organic waste to be treated in the incineration of the hearth according to the comparison result, and determining the correction direction of the factor to be corrected;
Determining target equipment to be controlled according to the factors to be corrected and the correction directions of the factors to be corrected;
and generating control parameters of target equipment according to the comparison result, the factors to be corrected and the correction directions of the factors to be corrected, and taking the control parameters as corresponding incineration control parameters of the hearth.
Therefore, the optional embodiment can also adaptively determine the target equipment to be controlled based on the determined factor to be corrected and the correction direction of the factor to be corrected, and further realize adaptive determination of the incineration control parameters based on the multidimensional information, thereby being beneficial to improving the generation efficiency, the generation accuracy and the generation reliability of the incineration control parameters.
In this alternative embodiment, further optionally, the determining, by the incineration control module 304, the specific manner of the target device to be controlled according to the factor to be corrected and the correction direction of the factor to be corrected includes:
determining at least one incineration control device corresponding to the hearth and an incineration control parameter range of each incineration control device;
judging whether target incineration control equipment capable of meeting the correction requirement of the factors to be corrected exists in all the incineration control equipment according to the incineration control parameter range, the factors to be corrected and the correction direction of the factors to be corrected of each incineration control equipment;
And when the judgment result is yes, determining target incineration control equipment which can meet the correction requirement of the factor to be corrected in all the incineration control equipment as target equipment to be controlled.
The correction requirement at least comprises a fixed correction requirement corresponding to the factor to be corrected, and the fixed correction requirement comprises a correction safety requirement and a correction effect requirement.
It can be seen that this alternative embodiment also enables, when determining the target device to be controlled, preferably the target incineration control device that meets the correction requirements, facilitating the rational utilization of the incineration control device in case the incineration control requirements are met.
In yet another alternative embodiment, as shown in fig. 4, the apparatus further comprises:
the object determining module 305 is configured to determine an organic waste to be treated, which is currently required to be incinerated in the furnace.
The specific ways of determining the organic waste to be treated for the incineration treatment currently required by the hearth by the object determining module 305 include:
determining all the organic wastes to be screened, and identifying the multi-dimensional information of all the organic wastes to be screened, wherein the multi-dimensional information of all the organic wastes to be screened comprises type information, processing priority, stacking time, stacking quantity and hazard degree to the surrounding environment;
According to the type information in the multidimensional information of all the organic wastes to be screened, performing filtering operation on the organic wastes which do not meet the incineration requirement in all the organic wastes to be screened so as to update all the organic wastes to be screened;
and determining the organic waste to be treated which is currently required to be incinerated in the hearth according to all the updated organic waste to be screened.
It can be seen that, before the incineration operation is performed, the optional embodiment can also filter the organic waste to be screened (such as filtering out non-combustible waste, waste with a serious negative influence on the current scene, etc.), and then further determine the organic waste to be treated, which is currently required to be incinerated, after filtering, so as to be beneficial to reducing the influence of the waste which does not meet the incineration requirement on the overall incineration effect.
In this optional embodiment, further optionally, the specific manner of determining, by the object determining module 305, the organic waste to be treated that is currently required to be incinerated in the furnace according to all the updated organic waste to be screened includes:
according to preset grouping conditions and updated multidimensional information of all the organic wastes to be screened, grouping the updated organic wastes to obtain a plurality of organic waste groups;
And screening one organic waste group from all the organic waste groups according to the current state information of the hearth, the multidimensional information of each organic waste group and the current scene information of the scene of the hearth, and determining the organic waste group in one organic waste group as the organic waste to be treated, which is currently required to be incinerated in the hearth.
It can be seen that, this optional embodiment can also group all the organic wastes to be screened after the filtration (for example, divide the wastes with small differences in fire points into one group, or divide the wastes with the same kind into one group, etc.), then based on the grouping result, realize the determination of the organic wastes to be treated that need to be incinerated currently, for example, select the organic waste group that affects the current scene information least when incinerating, or select the organic waste group that has the shortest incineration duration, or select the organic waste group that has the highest incineration emergency degree, be favorable to improving the matching degree of the determined organic wastes to be treated that need to be incinerated currently and the current incineration demand, and the grouping mode can also realize the rational utilization of the incineration control resources, reduce the waste of unnecessary incineration control resources. In addition, the determination flexibility of the organic waste to be treated, which is required to be incinerated at present, of the hearth can be improved.
In yet another alternative embodiment, as shown in fig. 4, the apparatus further comprises:
the incineration preparation control module 306 is configured to determine, after determining that the organic waste to be incinerated is currently required to be incinerated in the hearth, an alternative organic waste to be incinerated in the hearth after the organic waste to be incinerated is incinerated; and determining the incineration preparation conditions of the alternative organic waste and the incineration preparation parameters corresponding to the incineration preparation conditions based on the incineration condition, the current incineration condition and the expected incineration condition of the organic waste to be treated in the process of incinerating the organic waste to be treated in the hearth; and in the process of incinerating the organic waste to be treated in the hearth, if the current condition meets the incineration preparation condition, controlling the incineration preparation equipment corresponding to the hearth to execute the pre-incineration preparation operation corresponding to the alternative organic waste according to the incineration preparation parameter.
Therefore, after determining the organic waste to be incinerated which is required to be incinerated currently, the alternative organic waste to be incinerated after the organic waste to be incinerated is incinerated, the incineration preparation conditions of the alternative organic waste and the intelligent determination of the incineration preparation parameters corresponding to the incineration preparation conditions of the hearth can be realized, the accuracy before the incineration of the object to be incinerated of the next wave is improved in advance in the process of incinerating the organic waste to be incinerated by the hearth, and the incineration efficiency of the object to be incinerated of the next wave is improved.
Example III
Referring to fig. 5, fig. 5 is a schematic structural diagram of another intelligent control device for burning organic waste based on flame vision according to an embodiment of the present invention. The device described in fig. 5 is used for realizing intelligent control on the organic waste burned in the furnace, and optionally, the device can be integrated in a local control device or a cloud control device, and the embodiment of the invention is not limited. As shown in fig. 5, the apparatus may include:
a memory 401 storing executable program codes;
a processor 402 coupled with the memory 401;
the processor 402 invokes executable program codes stored in the memory 401 to perform some or all of the steps in the flame vision-based intelligent control method for organic waste incineration described in the first embodiment of the present invention.
Example IV
The embodiment of the invention discloses a computer storage medium which stores computer instructions for executing part or all of the steps in the intelligent control method for burning organic waste based on flame vision, which is described in the first embodiment of the invention, when the computer instructions are called.
The apparatus embodiments described above are merely illustrative, wherein the modules illustrated as separate components may or may not be physically separate, and the components shown as modules may or may not be physical, may be located in one place, or may be distributed over a plurality of network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above detailed description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course by means of hardware. Based on such understanding, the foregoing technical solutions may be embodied essentially or in part in the form of a software product that may be stored in a computer-readable storage medium including Read-Only Memory (ROM), random-access Memory (Random Access Memory, RAM), programmable Read-Only Memory (Programmable Read-Only Memory, PROM), erasable programmable Read-Only Memory (Erasable Programmable Read Only Memory, EPROM), one-time programmable Read-Only Memory (OTPROM), electrically erasable programmable Read-Only Memory (EEPROM), compact disc Read-Only Memory (Compact Disc Read-Only Memory, CD-ROM) or other optical disc Memory, magnetic disc Memory, tape Memory, or any other medium that can be used for computer-readable carrying or storing data.
Finally, it should be noted that: the embodiment of the invention discloses an intelligent control method and device for burning organic waste based on flame vision, which are disclosed as preferred embodiments of the invention, and are only used for illustrating the technical scheme of the invention, but not limiting the technical scheme; although the invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art will understand that; the technical scheme recorded in the various embodiments can be modified or part of technical features in the technical scheme can be replaced equivalently; such modifications and substitutions do not depart from the spirit and scope of the corresponding technical solutions.
Claims (10)
1. An intelligent control method for burning organic waste based on flame vision is characterized by comprising the following steps:
collecting a current incineration image of the organic waste to be treated when the hearth incinerates the organic waste to be treated in the process of incinerating the organic waste to be treated in the hearth;
inputting the current incineration image into a flame recognition model trained in advance until convergence, and obtaining a recognition result of the flame recognition model;
determining the current incineration flame state of the organic waste to be treated based on the identification result, and comparing the current incineration flame state with a predetermined target incineration flame state to obtain a comparison result;
And generating an incineration control parameter corresponding to the hearth according to the comparison result, and executing the incineration control operation for the organic waste to be treated based on the incineration control parameter.
2. The intelligent control method for burning organic waste based on flame vision according to claim 1, wherein the flame recognition model is trained by:
acquiring a plurality of historical incineration image information acquired by an image acquisition device when the organic waste is historically incinerated for a plurality of times;
constructing an initial data set based on all the historical incineration image information, and executing data expansion operation on the initial data set to obtain an expanded data set;
performing combustion state labeling operation on the extended data set based on flame parameters in each incineration image information of the extended data set to obtain a target data set with labeling results, wherein the labeling results of each incineration image information of the extended data set comprise combustion states corresponding to the flame parameters in each incineration image information; wherein the flame parameters include flame size and/or flame color, the flame size including flame size and/or flame force size;
Dividing the target data set into a training data set, a verification data set and a test data set according to the determined dividing proportion;
constructing an initial neural network model, setting initial model parameters of the neural network model, inputting the training data set and the verification data set into the neural network model to obtain a prediction result of the neural network model, determining a loss function according to the prediction result of the neural network model and a corresponding labeling result, reversely transmitting the loss function to each layer of network of the neural network model, calculating the partial conductance of each layer of network parameters of the neural network model by errors of each layer of network, updating network parameters of each layer of network in the neural network model according to a preset gradient descent algorithm, repeatedly executing the network parameters of each layer of network in the neural network model, inputting the training data set and the verification data set into the neural network model to obtain the prediction result of the neural network model, determining a loss function according to the prediction result of the neural network model and the corresponding labeling result, reversely transmitting the loss function to each layer of network of the neural network model, calculating the partial conductance of each layer of network parameters of the neural network model by errors of each layer of network model until the neural network parameters of each layer of network meet the loss function of the neural network model according to the preset gradient descent algorithm, and repeatedly executing the step of updating the neural network model;
And testing the neural network model with the loss function meeting the preset loss condition based on the test data set to obtain a test result, and determining the neural network model with the loss function meeting the preset loss condition as a flame recognition model trained to be converged when the test result indicates that the neural network model with the loss function meeting the preset loss condition passes the test.
3. The intelligent control method for burning organic waste based on flame vision according to claim 1, wherein the generating the burning control parameters corresponding to the hearth according to the comparison result comprises:
determining a factor to be corrected corresponding to the organic waste to be treated burned by the hearth and a correction direction of the factor to be corrected according to the comparison result;
determining target equipment to be controlled according to the factors to be corrected and the correction directions of the factors to be corrected;
and generating control parameters of the target equipment according to the comparison result, the factors to be corrected and the correction directions of the factors to be corrected, and taking the control parameters as the corresponding incineration control parameters of the hearth.
4. The intelligent control method for burning organic waste based on flame vision according to claim 3, wherein the determining the target device to be controlled according to the factor to be corrected and the correction direction of the factor to be corrected comprises:
determining an incineration control parameter range of at least one incineration control device corresponding to the hearth;
judging whether target incineration control equipment capable of meeting the correction requirement of the factors to be corrected exists in all the incineration control equipment according to the incineration control parameter range of each incineration control equipment, the factors to be corrected and the correction directions of the factors to be corrected;
when the judgment result is yes, determining target incineration control equipment capable of meeting the correction requirement of the factor to be corrected in all the incineration control equipment as target equipment to be controlled;
the correction requirement at least comprises a fixed correction requirement corresponding to the factor to be corrected, and the fixed correction requirement comprises a correction safety requirement and a correction effect requirement.
5. The intelligent control method for flame vision-based organic waste incineration according to any one of claims 1 to 4, further comprising:
Determining the organic waste to be treated which is currently required to be incinerated in the hearth;
the determining the organic waste to be treated, which is required to be incinerated currently, of the hearth comprises the following steps:
determining all the organic wastes to be screened, and identifying multi-dimensional information of all the organic wastes to be screened, wherein the multi-dimensional information of all the organic wastes to be screened comprises type information, processing priority, stacking time, stacking quantity and hazard degree to surrounding environment;
according to the type information in the multidimensional information of all the organic wastes to be screened, performing filtering operation on the organic wastes which do not meet the incineration requirement in all the organic wastes to be screened so as to update all the organic wastes to be screened;
and determining the organic waste to be treated, which is currently required to be incinerated in the hearth, according to all the updated organic waste to be screened.
6. The intelligent control method for burning organic waste based on flame vision according to claim 5, wherein the determining the organic waste to be burned currently required by the furnace according to all the updated organic waste to be screened comprises:
According to preset grouping conditions and updated multi-dimensional information of all the organic wastes to be screened, grouping the updated all the organic wastes to be screened to obtain a plurality of organic waste groups;
and screening one organic waste group from all the organic waste groups according to the current state information of the hearth, the multidimensional information of each organic waste group and the current scene information of the scene of the hearth, and determining the organic waste group in the one organic waste group as the organic waste to be treated, which is currently required to be incinerated in the hearth.
7. The intelligent control method for the incineration of organic waste based on flame vision according to claim 6, further comprising:
after determining the organic waste to be incinerated which is currently required by the hearth, determining the alternative organic waste which is required by the hearth to be incinerated after the organic waste to be incinerated is incinerated;
in the process of incinerating the organic waste to be treated by the hearth, determining an incineration preparation condition of the alternative organic waste and an incineration preparation parameter corresponding to the incineration preparation condition based on the incinerated condition, the current incineration condition and the expected incineration condition of the organic waste to be treated;
And in the process of incinerating the organic waste to be treated in the hearth, if the current condition meets the incineration preparation condition, controlling the incineration preparation equipment corresponding to the hearth to execute the pre-incineration preparation operation corresponding to the alternative organic waste according to the incineration preparation parameter.
8. An intelligent control device for organic matter incineration based on flame vision, which is characterized by comprising:
the acquisition module is used for acquiring a current incineration image when the hearth incinerates the organic waste to be treated in the process of incinerating the organic waste to be treated in the hearth;
the recognition module is used for inputting the current incineration image into a flame recognition model trained to be converged in advance to obtain a recognition result of the flame recognition model, and determining the current incineration flame state of the organic waste to be treated based on the recognition result;
the comparison module is used for comparing the current incineration flame state with a predetermined target incineration flame state to obtain a comparison result;
and the incineration control module is used for generating the incineration control parameters corresponding to the hearth according to the comparison result and executing the incineration control operation on the organic waste to be treated based on the incineration control parameters.
9. An intelligent control device for burning organic waste based on flame vision, which is characterized by comprising:
a memory storing executable program code;
a processor coupled to the memory;
the processor invokes the executable program code stored in the memory to perform the flame vision-based intelligent control method of organic waste incineration as claimed in any one of claims 1 to 7.
10. A computer storage medium storing computer instructions which, when invoked, are operable to perform the flame vision-based intelligent control method of organic waste incineration as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311417480.7A CN117422982A (en) | 2023-10-27 | 2023-10-27 | Flame vision-based intelligent control method and device for organic waste incineration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311417480.7A CN117422982A (en) | 2023-10-27 | 2023-10-27 | Flame vision-based intelligent control method and device for organic waste incineration |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117422982A true CN117422982A (en) | 2024-01-19 |
Family
ID=89528014
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311417480.7A Pending CN117422982A (en) | 2023-10-27 | 2023-10-27 | Flame vision-based intelligent control method and device for organic waste incineration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117422982A (en) |
-
2023
- 2023-10-27 CN CN202311417480.7A patent/CN117422982A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103597413A (en) | Method for the computer-supported generation of a data-driven model of a technical system, in particular of a gas turbine or wind turbine | |
| TW201947463A (en) | Model test method and device | |
| CN109902018B (en) | Method for acquiring test case of intelligent driving system | |
| CN110929918A (en) | 10kV feeder line fault prediction method based on CNN and LightGBM | |
| CN110139067A (en) | A kind of wild animal monitoring data management information system | |
| CN110942228A (en) | Energy consumption control method and device, computer equipment and storage medium | |
| CN109711707B (en) | Comprehensive state evaluation method for ship power device | |
| CN109376061A (en) | A kind of information processing method and system | |
| CN110334627A (en) | The device and system that the behavior of personnel is detected | |
| CN113405106A (en) | Artificial intelligence control method for waste incineration process | |
| CN115798022B (en) | Artificial intelligence recognition method based on feature extraction | |
| CN108229138A (en) | A kind of password dictionary generation method based on deep learning algorithm | |
| CN110966731B (en) | Method for regulating operating parameters | |
| CN117422982A (en) | Flame vision-based intelligent control method and device for organic waste incineration | |
| CN115964503B (en) | Safety risk prediction method and system based on community equipment facilities | |
| KR102368834B1 (en) | Attribute based welfare target prediction system and method thereof | |
| CN113313204A (en) | Deep learning-based garbage incineration state identification method and incineration control method | |
| CN115346084A (en) | Sample processing method, sample processing apparatus, electronic device, storage medium, and program product | |
| CN114169623A (en) | Power equipment fault analysis method and device, electronic equipment and storage medium | |
| CN110955286B (en) | Poultry egg monitoring method and device | |
| KR20230063905A (en) | Heatwave Probability Prediction System based on Recurrent Neural Network | |
| CN113807541A (en) | Fairness repair method, system, equipment and storage medium for decision system | |
| CN114301637B (en) | Intrusion detection method and system for medical Internet of things | |
| CN115001941B (en) | Method and device for determining communication network management faults | |
| CN116579337B (en) | False news detection method integrating evidence credibility |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |