Method, system and device for controlling overall movement of waste incinerator combustion grate

Abstract

A method for controlling the overall movement of a waste incinerator combustion grate, comprising: collecting a furnace flame image and incinerator data in real time; processing the furnace flame image to obtain flame feature data; processing incinerator data by means of a waste bed thickness soft-sensing model to obtain a thickness prediction result, wherein the waste bed thickness soft-sensing model is constructed by means of training a neural network model, and training data comprises a furnace chamber temperature, an oxygen content, and a flue gas composition; obtaining a flame center position by means of the flame feature data, obtaining a combustion intensity by combining a final flame center position with the thickness prediction result, obtaining an overall motion cycle parameter of the grate by means of the flame center position and the combustion intensity, and sending the overall motion cycle parameter of the grate to a control system; and controlling the waste incinerator grate by means of the control system according to the overall motion cycle parameter of the grate. The method processes data collected in real time, thereby allowing for real-time adjustment of a waste incinerator grate.

Description

A control method, system and device for overall movement of a combustion grate of a waste incinerator TECHNICAL FIELD

[0001] The present application relates to the technical field of waste incinerators, and in particular to a control method, system and device for overall movement of a combustion grate of a waste incinerator. BACKGROUND

[0002] In traditional waste incinerators, the control method for the grate is usually based on the principle of reciprocating grates, and multiple parameters need to be frequently adjusted manually to control the operation of the grate. Specifically, the movement cycle parameters of each grate as a whole need to be adjusted every time the combustion grate operates a cycle, including the number of turning times, the number of sliding times, the interval time of turning action and the interval time of sliding action, etc. The adjustment of these parameters directly affects the effect of waste incineration and the stability of the grate operation. In traditional waste incinerators, the adjustment of these parameters is usually performed manually by experienced operators. The operator needs to adjust these parameters constantly to achieve the best combustion effect according to the type, humidity, combustion temperature of the waste, and other factors, combined with actual observation and experience. However, this manual adjustment has some problems. First, the amount of parameter adjustment is large, and the operator needs to spend a lot of time and effort to adjust. Second, due to human factors and subjective judgments, errors and unstable situations may occur during the adjustment process. Finally, the results of parameter adjustment may be affected by other factors, such as changes in waste and changes in environmental conditions, so that the adjusted parameters may not be suitable for all situations. SUMMARY

[0003] The main purpose of the present application is to provide a control method, system and device for overall movement of a combustion grate of a waste incinerator, which processes real-time collected data through a learning model, so that the system can adjust and control the overall movement of the grate of the waste incinerator in real time according to the cycle parameters calculated by the learning model, without the need for manual adjustment of the conditions corresponding to each parameter, thereby reducing errors and avoiding unstable situations.

[0004] To achieve the above-mentioned purpose, the present application provides a control method for overall movement of a combustion grate of a waste incinerator, comprising:

[0005] real-time collection of furnace flame images and incinerator data, the incinerator data including: furnace temperature, oxygen content, flue gas composition;

[0006] processing the furnace flame images to obtain flame feature data, the flame feature data including: flame brightness, flame color, flame shape, flame area;

[0007] The thickness prediction result is obtained by processing the incinerator data through a layer thickness soft measurement model, wherein the layer thickness soft measurement model is constructed through neural network model training, and the training data includes layer thickness, furnace temperature, oxygen content, and flue gas composition;

[0008] The flame center position is obtained through the flame feature data, the combustion intensity is obtained by combining the final flame center position with the thickness prediction result, the overall movement cycle parameter of the grate is obtained through the flame center position and the combustion intensity, and the overall movement cycle parameter of the grate is sent to a control system;

[0009] The control system controls the garbage incinerator grate according to the overall movement cycle parameter of the grate.

[0010] In some embodiments, the processing of the furnace flame image to obtain flame feature data includes:

[0011] The denoising processing is performed on the furnace flame image, the image enhancement is performed on the furnace flame image after the denoising processing, the enhanced image is converted into a corresponding color space, the color space is divided into regions, and the number of pixels in each region is counted to obtain a color histogram, and the statistical characteristics of the color are obtained by calculating the mathematical expectation and the center distance of the color value and the position of the pixels in the color space; the flame feature data is obtained by combining the color histogram and the statistical characteristics of the color, wherein the statistical characteristics of the color include the average value of the color, the variance of the color, and the skewness of the color.

[0012] In some embodiments, the thickness prediction result is obtained by processing the incinerator data through a layer thickness soft measurement model, wherein the layer thickness soft measurement model is constructed through neural network model training, and the training data includes layer thickness, furnace temperature, oxygen content, and flue gas composition, including:

[0013] The data quality of the layer thickness and the incinerator data is judged, the noise is removed, the data is smoothed, and the data is processed by moving average, and the extracted layer data includes average thickness data, maximum thickness data, and thickness change rate data;

[0014] The initial layer thickness soft measurement model is constructed by training the neural network model according to the layer data and the furnace temperature, the oxygen content, and the flue gas composition data, and the layer thickness soft measurement model is obtained by optimizing the parameters of the initial layer thickness soft measurement model, wherein the optimization parameters include: pressure difference, air supply fan opening change, air supply flow, air temperature, grate area, and air density;

[0015] And the real-time collected incinerator data is calculated through the material layer thickness soft measurement model to obtain the thickness prediction result.

[0016] In some embodiments, the calculation of the flame feature data to obtain the flame center position, and the calculation of the thickness prediction result in combination with the final flame center position to obtain the combustion intensity, includes:

[0017] According to the preset threshold, the hearth flame image is threshold segmented, and the centroid of the segmented flame binary image is calculated to obtain an initial flame center position, the flame feature data is calculated, and the calculation result is compared and verified in combination with the initial flame center position to obtain a final flame center position.

[0018] At the same time, the thickness prediction result is calculated in combination with the final flame center position to obtain the combustion intensity.

[0019] In some embodiments, the motion cycle parameters of the grate as a whole are obtained through the flame center position and the combustion intensity, including:

[0020] A fuzzy rule base is established, and the final flame center position and the combustion intensity are input for fuzzy control calculation to obtain fuzzy motion cycle parameters of the grate as a whole, at the same time, the fuzzy motion cycle parameters of the grate as a whole are calculated through a preset maximum membership degree method to obtain a peak value of the membership function of the fuzzy motion cycle parameters of the grate as a whole, and the specific motion cycle parameters of the grate as a whole are obtained through conversion of the peak value.

[0021] In some embodiments, the control system controls the garbage incinerator grate according to the motion cycle parameters of the grate as a whole, including:

[0022] The control system automatically sets the turning frequency, sliding speed, sliding frequency, interval time of turning action and interval time of sliding action of the garbage incinerator grate, and calculates the total speed of the garbage incinerator grate, at the same time, the control system monitors the data in the furnace in real time, and detects the abnormality according to the real-time monitoring data, when the monitored data exceeds the preset value, the control system adjusts the motion cycle parameters of the grate as a whole in real time.

[0023] The application also provides a garbage incinerator combustion grate overall motion control method system, including:

[0024] The acquisition module is used for real-time acquisition of hearth flame images and incinerator data, and the incinerator data includes hearth temperature, oxygen content and flue gas composition.

[0025] A processing module is configured to process the furnace flame image to obtain flame feature data, the flame feature data including flame brightness, flame color, flame shape, and flame area; the processing module is further configured to establish a material layer thickness soft measurement model according to the incinerator data, and to process the incinerator data by the material layer thickness soft measurement model to obtain a thickness prediction result, wherein the material layer thickness soft measurement model is constructed by neural network model training, and the training data includes material layer thickness, furnace temperature, oxygen content, and flue gas composition;

[0026] A calculation module is configured to obtain a flame center position by the flame feature data, to obtain a combustion intensity by combining the final flame center position with the thickness prediction result, to obtain a motion cycle parameter of the overall grate by the flame center position and the combustion intensity, and to send the motion cycle parameter of the overall grate to a control system;

[0027] A control module is configured to control the garbage incinerator grate by the control system according to the motion cycle parameter of the overall grate.

[0028] The application further provides a garbage incinerator grate overall motion control method and device, which comprises:

[0029] A memory is configured to store a program;

[0030] A processor is configured to execute the program to realize each step of the garbage incinerator grate overall motion control method.

[0031] The application has the following beneficial effects:

[0032] Real-time acquisition of the furnace flame image and the incinerator data can provide an accurate data basis for analyzing and judging the working state of the furnace, processing the flame image to obtain flame feature data, better understanding the combustion in the furnace, and thus effectively optimizing and adjusting the working of the furnace. A material layer thickness soft measurement model based on the incinerator data is established, and the incinerator data is processed and predicted through the model, so that the thickness change of the material layer can be accurately mastered, and thus the garbage incineration process in the furnace can be better controlled. In combination with the flame feature data and the thickness prediction result, the motion cycle parameter of the overall grate is obtained, so that the motion law of the grate can be better understood, and thus reasonable regulation and control can be performed. The motion cycle parameter of the overall grate is sent to the control system, and multiple parameters such as the number of turning, the number of sliding, the interval time between turning and sliding, and the like in one motion cycle do not need to be set multiple times, but all the grate actions in one cycle are considered as a complete body for control, so that the grate of the garbage incinerator can be controlled according to one overall motion cycle parameter, more accurate and efficient overall grate motion control is realized, and the operation efficiency and safety of the garbage incinerator are improved. BRIEF DESCRIPTION OF DRAWINGS

[0033] Fig. 1 is a flow chart of the control method of the overall motion of the garbage incinerator grate in the embodiment of the present application;

[0034] Fig. 2 is a flow chart of processing the flame feature data and the thickness prediction result in the embodiment of the present application;

[0035] Fig. 3 is a system structure diagram of the control method of the overall motion of the garbage incinerator grate in the embodiment of the present application. DETAILED DESCRIPTION

[0036] The embodiments of the present application are described in detail below. It should be emphasized that the following description is only exemplary, but is not intended to limit the scope of the present application and its applications.

[0037] It should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the embodiments of the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0038] In addition, the terms "first", "second", "third", etc. are used only for descriptive purposes and are not to be construed as indicating or implying relative importance or an ordered ranking of the indicated technical features. Thus, features defined with "first", "second" or "third" can explicitly or implicitly include one or more of such features. In the description of embodiments of the present application, the meaning of "a plurality" is two or more, unless explicitly specified otherwise.

[0039] Referring to FIG. 1, the present application proposes a control method for overall movement of a grate of a waste incinerator, comprising:

[0040] Real-time acquisition of a furnace flame image and incinerator data, wherein the incinerator data includes: bed thickness, furnace temperature, oxygen content, flue gas composition;

[0041] Processing the furnace flame image to obtain flame feature data, wherein the flame feature data includes: flame brightness, flame color, flame shape, flame area, wherein the combustion conditions (such as bed thickness, oxygen content, furnace temperature, etc.) directly affect the characteristics of the flame, such as too thick fuel layer leading to incomplete local combustion, thereby affecting the color and brightness of the flame. The characteristics of the flame can also reflect the state of the combustion conditions, such as a reddish flame color indicating a low combustion temperature and insufficient oxygen; while a bright white flame may indicate that the combustion is more complete;

[0042] Establishing a bed thickness soft measurement model according to the incinerator data, wherein the bed thickness soft measurement model is constructed by neural network model training, and a thickness prediction result is obtained by processing the incinerator data through the bed thickness soft measurement model, wherein the bed thickness soft measurement model is trained by the bed thickness, furnace temperature, oxygen content, and flue gas composition of the incinerator data, and the bed thickness soft measurement model is used for thickness prediction of the waste incinerator. The bed thickness soft measurement model extracts features from the bed thickness, furnace temperature, oxygen content, and flue gas composition to obtain specific incinerator feature data, and performs feature prediction on the incinerator feature data to obtain the corresponding thickness prediction result;

[0043] Calculating the flame feature data through a preset periodic calculation model to obtain a flame center position, calculating the thickness prediction result through the periodic calculation model combined with the final flame center position to obtain a combustion intensity, performing data analysis on the flame center position and the combustion intensity to obtain a movement cycle parameter of the grate as a whole, and sending the movement cycle parameter of the grate as a whole to a control system;

[0044] The control system controls the grate of the waste incinerator according to the movement cycle parameter of the grate as a whole.

[0045] The overall movement of the above-mentioned grate is relative to the independent movement of the traditional turning and sliding. When the traditional combustion grate is working, the incinerator is provided with multiple combustion sections, each combustion section has a turning grate and a sliding grate. Their movements are relatively independent. When specifically controlled, the number of turns, the number of slides, the interval time between turns, and the interval time between turns and slides are set in one movement cycle. In the embodiment, the method considers all grate movements in one cycle as a complete body for control, so that more accurate and efficient overall grate movement control can be realized, and the operation efficiency and safety of the waste incinerator are improved.

[0046] The application provides a control method for overall movement of a combustion grate of a waste incinerator. The method collects real-time furnace flame images and incinerator data, provides real-time combustion and material layer state information in the furnace, processes the furnace flame images to obtain flame feature data, more accurately describes the flame state, establishes a material layer thickness soft measurement model based on the incinerator data, processes the incinerator data through the model to obtain thickness prediction results, realizes real-time monitoring and prediction of the material layer thickness, and helps the control system optimize the combustion process and combustion efficiency. The flame feature data and the thickness prediction results are calculated to obtain movement cycle parameters of the overall grate, the control system controls the grate of the waste incinerator according to the movement cycle parameters of the overall grate, realizes control of the grate movement, improves the efficiency and stability of the combustion in the furnace, can automatically adjust the grate, and improves the operation efficiency and safety of the waste incinerator.

[0047] Based on the above-mentioned control method for overall movement of a combustion grate of a waste incinerator, in one embodiment, the cycle calculation model comprises:

[0048] The artificial neural network used adopts a fully connected multilayer perceptron structure to establish a learning model, including an input layer, multiple hidden layers and an output layer, each hidden layer contains different numbers of neurons, and is used for extracting feature information at different levels.

[0049] The forward propagation calculation of the neural network can be represented as:

[0050] Hidden layer: Z (e) = W (e) A (e-1) + B (e) A (e) = S(Z (e) )

[0051] Wherein, e: represents the e layer, and the calculation starts from the input layer.

[0052] W (e) : the weight matrix from the e layer to the e-1 layer.

[0053] A (e-1): Activation value (output) of the e-1 layer;

[0054] B (e) : Bias vector of the e layer;

[0055] S: Activation function, such as Sigmoid, ReLU, etc.

[0056] Each time the data is processed or acted upon:

[0057] Input layer: receives the information of the original data and passes it to the next hidden layer.

[0058] Hidden layer: calculates the weighted input of each hidden layer neuron through the linear combination of weight matrix and bias term; uses the activation function for nonlinear transformation to generate the activation value of each neuron; the role of the hidden layer is to extract the features of the input data, and different levels of hidden layers can extract different levels of abstract features.

[0059] Output layer: calculates the final output result through the output value of the hidden layer, which is used to complete specific tasks such as classification or regression; the activation function of the output layer maps the output value to the appropriate range to ensure that the result meets the requirements of the task.

[0060] In one embodiment, the furnace flame image is processed to obtain flame feature data, including:

[0061] The furnace flame image is denoised, the denoised furnace flame image is image enhanced, the enhanced image is converted into a corresponding color space, the color space is regionally divided, and the number of pixels in the image falling in each interval is counted to obtain a color histogram, and the statistical characteristics of the color are obtained by calculating the mathematical expectation of the color value and the position of the pixels in the color space and the center distance, the statistical characteristics of the color include the average value of the color, the variance of the color, and the skewness of the color.

[0062] The embodiment effectively eliminates the noise in the furnace flame image through the denoising processing of the learning model, enhances the flame image through the learning model, improves the visual quality of the furnace flame image, makes the flame clearer and easier to analyze, improves the effect of subsequent processing, converts the enhanced image to the corresponding color space, can better represent the color information of the flame, is helpful for subsequent color analysis and feature extraction, divides the image into regions, and counts the number of pixels in each region to obtain the color histogram of the flame image. Based on the distribution of each color in the flame image provided by the color histogram, the statistical features of the color, such as the mean, variance and skewness, are extracted by calculating the mathematical expectation of the color value and the center distance of the position in the color space. Finally, by combining the learning model, the color histogram and the statistical features of the color, different feature information is comprehensively utilized to obtain comprehensive and accurate flame feature data, and the accuracy and reliability of the flame analysis and the flame feature data are improved.

[0063] In one embodiment, the thickness prediction result is obtained by processing the incinerator data through the material layer thickness soft measurement model, wherein the material layer thickness soft measurement model is constructed by training a neural network model, and the training data includes material layer thickness, furnace temperature, oxygen content and flue gas composition, including:

[0064] The material layer thickness and incinerator data are subjected to noise removal and smoothing data processing, and the processed data is subjected to feature extraction. The extracted material layer data includes average thickness data, maximum thickness data and thickness change rate data, and a plurality of parameters such as pressure difference between the grate and the grate, air supply fan opening change, air supply flow, air temperature, grate area, air density, etc.

[0065] The neural network model is trained according to the furnace temperature, oxygen content and flue gas composition data to construct an embryonic measurement model with an initial state, the embryonic measurement model is trained and set according to the material layer data, an initial material layer thickness soft measurement model is obtained, and the initial material layer thickness soft measurement model is optimized according to the pressure difference, the air supply fan opening change, the air supply flow, the air temperature, the grate area and the air density to obtain the material layer thickness soft measurement model. The embodiment considers the problem of air supply flow change caused by short-time rapid change of the fan in modeling, so that the soft measurement mathematical model has good anti-interference performance.

[0066] The real-time collected incinerator data (such as furnace temperature, oxygen content, flue gas composition) are calculated through the material layer thickness soft measurement model to obtain the thickness prediction result.

[0067] The embodiment denoises and smooths the incinerator data, eliminates noise and fluctuations in the data, and improves the accuracy and reliability of the data. The average thickness data, maximum thickness data, and thickness change rate data are extracted from the processed incinerator data to obtain the overall situation and trend of the incinerator data, thereby providing a basis for subsequent analysis and prediction. The incinerator data is used as input data, and the relationship between the material layer thickness and other related variables is calculated to establish a material layer thickness soft measurement model. The real-time collected incinerator data is input into the material layer thickness soft measurement model, and the thickness prediction result is obtained by model calculation. Real-time monitoring and prediction of the material layer thickness are realized, and data support is provided for subsequent system control and optimization.

[0068] Based on the previous embodiment, in one embodiment, the specific material layer thickness soft measurement model establishment process includes:

[0069] The material layer thickness soft measurement model is established by using a fully connected multi-layer perceptron structure and a neural network model, including a sensing layer, a material layer processing layer, and an output layer.

[0070] Specific material layer data is obtained, including average thickness, maximum thickness, and thickness change rate data.

[0071] The data is processed, and the processed data is used as input features to construct an input data matrix of the neural network.

[0072] The neural network structure is designed, and the number of neurons and activation functions of the material layer processing layer, and the activation function of the output layer are defined.

[0073] The loss function for material layer thickness prediction is determined, such as mean square error (MSE).

[0074] The training data is used to train the material layer thickness soft measurement model, and the weights and bias terms are updated through the forward propagation algorithm. The validation data set is used to evaluate the performance of the model, and the model hyperparameters are adjusted to improve the prediction accuracy.

[0075] Each layer processes and affects the data as follows:

[0076] Input layer: receives average thickness, maximum thickness, and thickness change rate data.

[0077] Material layer processing layer: calculates the weighted input of each hidden layer neuron through linear combination of weight matrix and bias term, uses activation function for non-linear transformation to generate activation value of each neuron. The material layer processing layer processes the extracted input features, learns and extracts feature information.

[0078] Output layer: Calculate the final layer thickness prediction result by the output value of the layer processing layer. The activation function of the output layer maps the output value to the appropriate range to get the final prediction result.

[0079] At the same time, the algorithm used by each layer is:

[0080] Sensing layer: O = W * X + D, a = h(O);

[0081] Output layer: P = r * a + d, y = q(P);

[0082] X: input data matrix, including average thickness, maximum thickness and thickness change rate data; W: weight matrix of layer processing layer to input layer; D: bias term of layer processing layer; a: activation value of layer processing layer; h: activation function of layer processing layer; r: weight matrix of output layer to layer processing layer; d: bias term of output layer; y: prediction result; q: activation function of output layer.

[0083] In one embodiment, the flame feature data and the thickness prediction result are processed, including:

[0084] According to the preset threshold, the furnace flame image is threshold segmented, wherein the furnace flame image includes the furnace flame image when the furnace is running and the furnace flame image when the furnace is slagging, and the centroid of the segmented flame binary image is calculated to obtain the initial flame center position, the flame feature data is calculated, and the calculation result is compared and verified with the initial flame center position to obtain the final flame center position.

[0085] At the same time, the thickness prediction result is calculated combined with the final flame center position to obtain the combustion intensity.

[0086] In this embodiment, the flame image is threshold segmented according to the preset threshold to separate the flame from the background and obtain a binary image, improving the extraction accuracy of the flame feature. Then, the centroid of the segmented flame binary image is calculated to obtain the initial center position of the flame. The flame feature data is calculated to extract the key features of the flame, such as color and shape. The calculation result is compared and verified with the initial flame center position to ensure that the extracted flame feature matches the actual flame, improving the accuracy and reliability of the result. And through comparison and verification, the final flame center position is obtained. Combined with the final flame center position, the thickness prediction result is calculated to obtain the combustion intensity. The evaluation and monitoring of the flame combustion state are realized.

[0087] In one embodiment, the movement cycle parameters of the grate as a whole are obtained through the flame center position and the combustion intensity, including:

[0088] The fuzzy rule base is established, and the final flame center position and combustion intensity are input for fuzzy control calculation to obtain the fuzzy overall movement cycle parameters of the grate. Meanwhile, the peak value of the membership function of the fuzzy overall movement cycle parameters of the grate is obtained by preset maximum membership degree method calculation, and the specific overall movement cycle parameters of the grate are obtained by conversion of the peak value.

[0089] The movement cycle parameters further include:

[0090] In one movement cycle of the combustion grate, all actions are regarded as a whole, i.e., the total time T occupied by the complete turning and sliding actions of the combustion grate in one cycle, and the total distance L moved by them, so that their total speed V=L / T.

[0091] The current combustion intensity of the waste combustion is obtained through the measurement of the material layer thickness and the combustion flame, so that the appropriate cycle speed V can be determined.

[0092] After V is determined, the turning frequency, sliding frequency, interval time between turning and interval time between turning and sliding of the combustion grate in each cycle are determined.

[0093] Based on the turning frequency, sliding frequency, interval time between turning and interval time between turning and sliding of the combustion grate in each cycle, key factor analysis is performed to obtain the movement frequency, movement amplitude and movement direction.

[0094] The turning frequency, sliding frequency, interval time between turning and interval time between turning and sliding are combined into corresponding movement cycle parameters, and the movement frequency, movement amplitude and movement direction are combined into auxiliary cycle parameters, which are used to assist the subsequent adjustment of the movement cycle parameters.

[0095] Movement frequency: the action speed of the grate, i.e., the number of reciprocating actions completed per unit time.

[0096] Movement amplitude: the displacement size of each action of the grate.

[0097] Movement direction: the direction of the grate action, usually involving pushing the waste forward to make it enter the next stage, or pulling it back to help mix the waste.

[0098] In a multi-stage reciprocating incineration grate, the movement cycle parameters refer to a set of parameters that control the frequency, direction and amplitude of the grate action, which determine the movement of the waste on the grate, and further affect the combustion efficiency and waste treatment capacity.

[0099] The embodiment can improve the adaptability and robustness of the overall movement cycle parameters of the grate by establishing a fuzzy rule base and inputting the final flame center position and combustion intensity for fuzzy control calculation to obtain the overall movement cycle parameters of the grate. Meanwhile, the preset maximum membership degree method is used to calculate the fuzzy overall movement cycle parameters of the grate, so that the peak value of the membership function of the fuzzy overall movement cycle parameters of the grate can be obtained, which is more conducive to determining the fuzzy degree and weight of the overall movement cycle parameters of the grate and provides a basis for subsequent parameter conversion. Finally, the peak value of the membership function is converted, so that the fuzzy overall movement cycle parameters of the grate can be converted into specific numerical values. Thus, the precise control and adjustment of the overall movement cycle parameters of the grate can be realized, and the combustion efficiency and energy utilization efficiency of the furnace can be improved.

[0100] Based on the overall movement cycle parameters of the grate obtained in the previous embodiment, when applied to calculate the movement cycle parameters of the multi-stage reciprocating incineration grate, the method further includes:

[0101] The number of grates of the multi-stage reciprocating incineration grate and the movement time of the upper and lower furnace slag of each stage of grates are obtained, the flame center position and the combustion intensity in the movement time are obtained, the preliminary movement cycle parameters of the multi-stage reciprocating incineration grate are calculated in combination with the number of grates and the movement time, the preliminary movement cycle parameters are verified according to the flame center position and the combustion intensity, and when the verification is passed, the preliminary movement parameters are calculated according to the obtained overall movement cycle parameters of the grate to obtain the movement cycle parameters of the multi-stage reciprocating incineration grate.

[0102] In one embodiment, the control system controls the grate of the waste incinerator according to the overall movement cycle parameters of the grate, including:

[0103] The control system automatically sets the turning frequency, sliding speed, sliding frequency, interval time of turning action and interval time of sliding action of the waste incinerator grate, and calculates the total running speed of the waste incinerator grate. At the same time, the control system monitors the data in the furnace in real time, and detects abnormalities according to the real-time monitoring data. When the monitored data exceeds the preset value, the control system adjusts the overall movement cycle parameters of the grate in real time.

[0104] The embodiment automatically sets the turning number, sliding speed, sliding number, interval time of turning action and interval time of sliding action of the grate according to the motion cycle parameters of the whole grate through the control system, without manually setting the conditions corresponding to each parameter, so as to realize accurate control of the grate of the waste incinerator. This helps to optimize the combustion process and improve the combustion efficiency and waste incineration processing capacity. Based on the application of the learning model, the control system can monitor the data in the furnace in real time, such as temperature, pressure, gas composition, etc. Through abnormal detection of the real-time monitoring data, the abnormal conditions in the furnace, such as excessively high temperature and abnormal pressure, can be found in time, so that corresponding control measures can be taken to ensure the safe operation of the hearth. When the monitored data exceeds the preset value, the control system can adjust the motion cycle parameters of the whole grate in real time. Through the application of the learning model, the motion cycle parameters of the whole grate can be adjusted according to the real-time monitoring data and the preset control strategy, so as to realize dynamic control and adjustment of the combustion process. Thus, the stable operation of the hearth can be maintained, and the combustion efficiency and waste incineration processing effect can be improved.

[0105] Referring to FIG. 2, the application further provides a control method system for the whole motion of the grate of a waste incinerator, comprising:

[0106] A collection module is configured to collect the hearth flame image and incinerator data in real time, wherein the incinerator data includes hearth temperature, oxygen content and flue gas composition.

[0107] A processing module is configured to process the hearth flame image to obtain flame feature data, wherein the flame feature data includes flame brightness, flame color, flame shape and flame area; the processing module is further configured to establish a soft measurement model for the thickness of the material layer according to the incinerator data, and process the incinerator data through the soft measurement model for the thickness of the material layer to obtain a thickness prediction result, wherein the soft measurement model for the thickness of the material layer is constructed through a neural network model training, and the training data includes the thickness of the material layer, hearth temperature, oxygen content and flue gas composition.

[0108] A calculation module is configured to obtain the center position of the flame through the flame feature data, obtain the combustion intensity by combining the final center position of the flame with the thickness prediction result, obtain the motion cycle parameters of the whole grate through the center position of the flame and the combustion intensity, and send the motion cycle parameters of the whole grate to a control system.

[0109] A control module is configured to control the grate of the waste incinerator through the control system according to the motion cycle parameters of the whole grate.

[0110] The control method and system for the overall movement of the grate of the waste incinerator can collect the furnace flame image and the incinerator data in real time, can provide accurate data basis, is used for analyzing and judging the working state of the furnace, processes the flame image through the learning model, obtains the flame feature data, can better understand the combustion condition in the furnace, thereby effectively optimizing and adjusting the work of the furnace, establishes the soft measurement model of the material layer thickness based on the incinerator data, and processes and predicts the incinerator data through the model, can accurately master the thickness change condition of the material layer, thereby better controlling the waste incineration process in the furnace, the flame feature data and the material layer thickness prediction result are calculated through the learning model, the movement cycle parameter of the grate is obtained, the movement law of the grate can be better understood, thereby reasonable regulation and control and control are carried out, the movement cycle parameter of the grate is sent to the control system, only one movement cycle parameter of the grate is needed to control the grate of the waste incinerator, it is not necessary to adjust the condition corresponding to each parameter manually, so that the grate movement can be controlled more accurately and efficiently, and the operation efficiency and safety of the waste incinerator are improved.

[0111] The application further provides a control method and device for the overall movement of the grate of a waste incinerator.

[0112] The memory is used for storing programs.

[0113] The processor is used for executing programs, and realizes each step of the control method for the overall movement of the grate of the waste incinerator.

[0114] In the embodiment, the processor and the memory can be connected through a bus or other means. The memory can include a volatile memory, such as a random access memory; the memory can also include a non-volatile memory, such as a read-only memory, a flash memory, a hard disk or a solid state disk. The processor can be a general-purpose processor, such as a central processing unit, a digital signal processor, an application-specific integrated circuit, or one or more integrated circuits configured to implement the embodiments of the application.

[0115] The above further describes the present application in conjunction with specific / preferred embodiments, and cannot be deemed to limit the specific implementation of the present application to these descriptions. For those skilled in the art to which the present application belongs, without departing from the concept of the present application, they can make several substitutions or variations to the described embodiments, and these substitutions or variations shall be deemed to fall within the protection scope of the present application. In the description of the present application, the description of the terms "an embodiment", "some embodiments", "a preferred embodiment", "an example", "a specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are contained in at least one embodiment or example of the present application. In the present application, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In the case of no mutual contradiction, those skilled in the art can combine and combine the different embodiments or examples described in the present application and the features of the different embodiments or examples. Although the embodiments of the present application and their advantages have been described in detail, it should be understood that various changes, substitutions and modifications can be made herein without departing from the scope of protection of the patent application.

Claims

1. A method of controlling the overall movement of a grate in a waste incinerator, characterized in that The method comprises the following steps: Real-time acquisition of furnace flame image and incinerator data, wherein the incinerator data comprises: furnace temperature, oxygen content, flue gas composition; Processing the furnace flame image to obtain flame feature data, wherein the flame feature data comprises: flame brightness, flame color, flame shape, flame area; Processing the incinerator data by a material layer thickness soft measurement model to obtain thickness prediction results, wherein the material layer thickness soft measurement model is constructed by neural network model training, and the training data comprises material layer thickness, furnace temperature, oxygen content and flue gas composition; Obtaining the flame center position by the flame feature data, obtaining the combustion intensity by combining the final flame center position with the thickness prediction results, obtaining the motion cycle parameters of the overall grate by the flame center position and the combustion intensity, and sending the motion cycle parameters of the overall grate to a control system; Controlling the garbage incinerator grate by the control system according to the motion cycle parameters of the overall grate.

2. The control method of the garbage incinerator combustion grate overall movement according to claim 1, characterized in that, The processing of the furnace flame image to obtain flame feature data comprises: Carrying out noise removal processing on the furnace flame image, carrying out image enhancement on the furnace flame image after noise removal processing, converting the enhanced image into a corresponding color space, dividing the color space into regions, counting the number of pixels in each region, obtaining a color histogram, calculating the mathematical expectation of the color value and position of the pixels in the color space, and obtaining the statistical characteristics of the color; combining the color histogram and the statistical characteristics of the color to obtain the flame feature data, wherein the statistical characteristics of the color comprise the average value of the color, the variance of the color and the skewness of the color.

3. The control method of the garbage incinerator combustion grate overall movement according to claim 1, characterized in that, The processing of the incinerator data by the material layer thickness soft measurement model to obtain thickness prediction results, wherein the material layer thickness soft measurement model is constructed by neural network model training, and the training data comprises material layer thickness, furnace temperature, oxygen content and flue gas composition, comprises: Carrying out data quality judgment, removing noise, smoothing data, data moving average processing on the material layer thickness data and the incinerator data, and extracting features from the processed data, wherein the extracted material layer data comprises average thickness data, maximum thickness data and thickness change rate data; Training a neural network model according to the material layer data, furnace temperature, oxygen content and flue gas composition data to construct an initial material layer thickness soft measurement model, and optimizing the parameters of the initial material layer thickness soft measurement model to obtain the material layer thickness soft measurement model, wherein the optimization parameters comprise: pressure difference, air supply fan opening change, air supply flow, air temperature, grate area, air density; And calculating the thickness prediction results by the material layer thickness soft measurement model from the real-time acquired incinerator data.

4. The control method of the garbage incinerator combustion grate overall motion according to claim 1, characterized by, The obtaining of the flame center position by the flame feature data, the combination of the final flame center position with the thickness prediction results, and the obtaining of the combustion intensity comprise: According to the preset threshold value, the hearth flame image is threshold segmented, the flame binary image obtained by the segmentation is centroid calculated, the initial flame center position is obtained, the flame feature data is calculated, and the calculation result is compared and verified in combination with the initial flame center position, and the final flame center position is obtained; At the same time, the thickness prediction result is calculated in combination with the final flame center position, and the combustion intensity is obtained.

5. The control method of the garbage incinerator combustion grate overall movement according to claim 4, characterized in that, Through the flame center position and the combustion intensity, the motion cycle parameters of the grate as a whole are obtained, including: A fuzzy rule base is established, and the final flame center position and the combustion intensity are input for fuzzy control calculation to obtain the fuzzy motion cycle parameters of the grate as a whole. At the same time, the fuzzy motion cycle parameters of the grate as a whole are calculated by a preset maximum membership degree method to obtain the peak value of the membership function of the fuzzy motion cycle parameters of the grate as a whole. Through conversion of the peak value, the specific motion cycle parameters of the grate as a whole are obtained.

6. The control method of the garbage incinerator combustion grate overall movement according to claim 5, characterized in that, The control system controls the grate of the waste incinerator according to the motion cycle parameters of the grate as a whole, including: The control system automatically sets the turning frequency, sliding speed, sliding frequency, interval time of turning action and interval time of sliding action of the grate of the waste incinerator, calculates the total running speed of the grate of the waste incinerator, and simultaneously monitors the data in the furnace in real time and detects abnormalities according to the real-time monitoring data. When the monitored data exceeds the preset value, the control system adjusts the motion cycle parameters of the grate as a whole in real time.

7. A control method system for the overall movement of the combustion grate of a waste incinerator, characterized in that, Including: The acquisition module is used for real-time acquisition of hearth flame image and incinerator data, and the incinerator data includes hearth temperature, oxygen content and flue gas composition; The processing module is used for processing the hearth flame image to obtain flame feature data, and the flame feature data includes flame brightness, flame color, flame shape and flame area. The processing module is also used for establishing a material layer thickness soft measurement model according to the incinerator data, and obtaining a thickness prediction result by processing the incinerator data through the material layer thickness soft measurement model. The material layer thickness soft measurement model is constructed by neural network model training, and the training data includes material layer thickness, hearth temperature, oxygen content and flue gas composition; The calculation module is used for obtaining a flame center position through the flame feature data, obtaining a combustion intensity by combining the final flame center position with the thickness prediction result, and obtaining the motion cycle parameters of the grate as a whole through the flame center position and the combustion intensity, and sending the motion cycle parameters of the grate as a whole to the control system; The control module is used for controlling the grate of the waste incinerator according to the motion cycle parameters of the grate as a whole through the control system.

8. A method of controlling the overall movement of a grate in a waste incinerator, characterized in that Including: The memory is used for storing programs; The processor is used for executing the programs to realize the steps of the control method for realizing the motion of the combustion grate of the waste incinerator according to any one of claims 1-6.

Images