A control system and device for efficient and clean intelligent operation of an incinerator throughout the whole process

Abstract

This invention discloses a control system and device for efficient, clean, and intelligent operation of an incinerator throughout the entire process, belonging to the field of municipal solid waste incineration operation control technology. It includes: a combustion process optimization control module, used to optimize boiler feeding and combustion control through a first intelligent prediction model, generating a first optimization result; a flue gas purification process optimization module, used to optimize boiler flue gas purification control through a second intelligent prediction model, generating a second optimization result; and a smart incineration control module, used to predict grate bed thickness, detect fire line and combustion status, calculate waste calorific value in real time, predict NOx, predict conventional pollutants, and predict dioxins through image recognition algorithms and machine learning algorithms, and to control and optimize the entire incinerator process based on the first and second optimization results. This invention achieves real-time diagnostic analysis of stable combustion in a waste incinerator, reducing pollutant emissions and achieving efficient and clean utilization of combustion.

Description

Technical Field

[0001] This invention relates to the field of urban solid waste incineration operation control technology, and more specifically, to a control system and device for efficient, clean and intelligent operation of an incinerator throughout the entire process. Background Technology

[0002] At present, the main technologies for urban domestic waste treatment include sanitary landfill, waste composting, waste incineration and comprehensive utilization. Among them, waste incineration technology has unique advantages in terms of volume reduction, harmlessness and resource recovery.

[0003] Due to the complex composition and unstable calorific value of municipal solid waste, stable control during operation is difficult. Currently, control methods for waste incineration mainly include manual predictive control based on DCS systems and PID control. Most waste incineration plants require operators to perform operations based on experience, such as primary air volume, secondary air volume, primary air ratio, feeder speed, and grate speed. A few plants can automate local equipment such as primary air, secondary air, and feeders, but achieving full-process automation is challenging. Overall, most waste incinerators have low automation levels, lacking a fully automated control system. Manual operation of certain modules is required, leading to poor combustion stability and potential issues such as unstable steam output and excessive levels of pollutants like dioxins and nitrogen oxides, hindering efficient and clean operation. Therefore, there is an urgent need for a fully integrated, efficient, clean, and intelligent operating system that integrates the entire waste incineration process to address these problems. Summary of the Invention

[0004] To address the aforementioned problems, this invention provides a control system for the efficient, clean, and intelligent operation of an incinerator throughout its entire process, comprising:

[0005] The combustion process optimization control module is used to optimize boiler feeding and combustion control through a first intelligent prediction model and generate a first optimization result.

[0006] The flue gas purification process optimization module is used to optimize boiler flue gas purification control through a second intelligent prediction model and generate a second optimization result.

[0007] The intelligent incineration control module is used to predict the grate bed thickness, detect the fire line and combustion status, calculate the calorific value of waste in real time, predict NOx, predict conventional pollutants and dioxins, and optimize the control of the entire incinerator process based on the first optimization result and the second optimization result.

[0008] Preferably, the combustion process optimization control module is further configured to, based on the collected incinerator operating data, use data affecting the incinerator's feeding grate, tilting grate, sliding grate, primary and secondary air, induced draft fan, and steam-water mixture as first feature data, and label them to generate first control variables for boiler feeding and combustion control optimization; simultaneously, based on the labels of the first feature data, obtain the first variable relationship between each first control variable, and assign the first variable relationship to the first feature data to generate a first dataset, train the deep learning model, and construct a first intelligent prediction model.

[0009] Preferably, the combustion process optimization control module is further used to optimize boiler feeding and combustion control based on a first control variable obtained from a first intelligent prediction model, wherein the first control variable includes:

[0010] The control variables for the feeding grate include: pusher speed setting and speed coefficient of each unit grate;

[0011] The control variables for grate turning include: the number of times each unit turns;

[0012] The control variables for the sliding grate include: the number of sliding cycles in each unit and the distribution coefficient of each unit;

[0013] The control variables for primary air volume include: the speed of the primary air fan in each unit;

[0014] The control variables for secondary air volume include: secondary fan frequency and left and right side damper positions;

[0015] The control variables for the induced draft fan include: induced draft fan frequency and induced draft fan inlet damper position;

[0016] The control variables for the soda include: the opening degree of the electric valve of the first-stage desuperheater and the opening degree of the electric valve of the second-stage desuperheater.

[0017] Preferably, the flue gas purification process optimization module is further configured to, based on the collected incinerator operating data, use data affecting the incinerator's semi-dry desulfurization, SNCR denitrification, activated carbon, wet scrubbing, and bag filter dust collection as second feature data, and label them to generate second control variables for boiler flue gas purification control optimization; simultaneously, based on the labels of the second feature data, obtain the second variable relationships between each second control variable, and assign the second variable relationships to the second feature data to generate a second dataset, train the deep learning model, and construct a second intelligent prediction model.

[0018] Preferably, the flue gas purification process optimization module is further used to optimize the control of boiler flue gas purification based on the second control variables obtained from the second intelligent prediction model, wherein the second control variables include:

[0019] The control variables for semi-dry deacidification include: the opening degree of the lime slurry valve of the atomizer and the opening degree of the tap water valve;

[0020] The control variables for SNCR denitrification include: the opening degree of the reducing agent valve and the opening degree of the soft water regulating valve;

[0021] The control variables for activated carbon include: the frequency of the activated carbon feeder;

[0022] The control variables for wet washing include: the opening degree of the sodium hydroxide absorbent pump regulating valve and the opening degree of the sodium hydroxide dehumidifying pump regulating valve;

[0023] The control variables for baghouse dust collectors include: air cannon time interval, pulse valve time interval, and dust removal time interval.

[0024] Preferably, the intelligent incineration control module is further used to obtain a third variable relationship between the first control variable and the second control variable based on the relationship between the first feature data and the second feature data;

[0025] Based on the first variable relationship, the second variable relationship, and the third variable relationship, the first intelligent prediction model and the second intelligent prediction model are used to adjust the first optimization result and the second optimization result, and then the entire process of the incinerator is controlled and optimized.

[0026] Preferably, the intelligent incineration control module is also used to detect the fire line and combustion status through a U-Net fully convolutional neural network.

[0027] Preferably, the intelligent incineration control module is also used to predict dioxins by fitting the nonlinear relationship between incineration conditions and dioxins through an autoregressive moving average model and a support vector machine model.

[0028] This invention discloses a control device for the efficient, clean, and intelligent operation of an incinerator throughout its entire process, comprising:

[0029] The sensor cluster is used to collect physical control parameters at various points in the incinerator in real time, providing real-time operating data of the incinerator.

[0030] An industrial control computer is used to carry the control system, optimize the entire process of the incinerator, and assist the incinerator in operating efficiently and cleanly.

[0031] Data cables are used as a data transmission medium to connect sensor clusters and industrial control computers.

[0032] Preferably, the data storage device is used to store real-time operating data and data generated by the industrial control computer;

[0033] The display device is used to visualize the entire process of the incinerator, and to display real-time operating data as well as the first and second optimization results predicted by the control system.

[0034] The present invention discloses the following technical effects:

[0035] This invention enables real-time diagnostic analysis of stable combustion in waste incinerators, reducing pollutant emissions and achieving efficient and clean utilization of combustion. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 Block diagram of the fully efficient and intelligent cleaning operation system described in this invention;

[0038] Figure 2 This is a schematic diagram of the control principle of the primary wind model predictive control described in this invention;

[0039] Figure 3 This is a comparison chart of the average furnace temperature changes over 5 minutes before and after actual use of the system described in this invention;

[0040] Figure 4 The NO at the tail end of the incinerator before and after the actual use system described in this invention x Comparison chart of changes. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0042] like Figure 1-4 As shown, the present invention provides a highly efficient, clean, and intelligent operation system for an incinerator throughout the entire process, including: a combustion process optimization control module, a flue gas purification process optimization module, and an intelligent incineration control module.

[0043] The combustion process optimization control module is used to optimize boiler feeding and combustion control, including feeding grate, tilting grate, sliding grate, primary and secondary air, induced draft fan, and steam-water system.

[0044] The flue gas purification process optimization module is used to optimize the control of boiler flue gas purification, including semi-dry acid removal, SNCR denitrification, activated carbon, wet scrubbing, and bag filter dust collection.

[0045] The intelligent incineration control module integrates the combustion process and flue gas purification process, realizing the optimization of the entire process control of the incinerator; at the same time, it uses image recognition algorithms and machine learning algorithms to predict the thickness of the grate bed, detect the fire line and combustion status, calculate the calorific value of waste in real time, predict NOx, predict conventional pollutants, and predict dioxins, achieving visualization and quantification of the combustion status, and assisting in the efficient and clean operation of the incinerator.

[0046] In the combustion process optimization control module, the models based on model predictive control include the feeding grate control model, the tilting grate control model, the sliding grate control model, the primary and secondary air volume control model, the induced draft fan control model, and the steam-water control model. Model predictive control is a time-series control method. At each time step, based on the state of the controlled object and the predictive model, the state of the target in the future is predicted. An optimal set of control sequences is then calculated based on a certain performance index (cost function). The first control action of this set of control sequences is passed as the output to the actuator. Simultaneously, the optimization algorithm continues to be executed at the next time step, continuously updating the control sequence. The control variables for the feeding grate control model include: pusher speed setting and grate speed coefficient of each unit; the control variables for the turning grate control model include: number of turnings in each unit; the control variables for the sliding grate control model include: number of slidings in each unit and distribution coefficient of each unit; the control variables for the primary air volume control model include: primary air fan speed in each unit; the control variables for the secondary air volume control model include: secondary air fan frequency and left and right side damper positions; the control variables for the induced draft fan control model include: induced draft fan frequency and induced draft fan inlet damper position; the control variables for the steam-water control model include: opening degree of the electric valve of the primary desuperheater and opening degree of the electric valve of the secondary desuperheater.

[0047] like Figure 2 As shown, taking the primary air volume control model as an example, the control objective of the primary air volume control model is the target load of the boiler, and the control variable is the speed of the primary air fan in each unit. The specific implementation method of the control includes:

[0048] (1) Primary air volume prediction model: The prediction model is the basis of model predictive control. The primary air volume prediction model obtains the current boiler load status information through DCS, and adds the input variables for future boiler load control (the speed of the primary air fan in each unit) to predict the future boiler load status. Optionally, the form of the prediction model includes: state equation or transfer function, step response or impulse response;

[0049] (2) Rolling Optimization: Model predictive control adopts a rolling finite-time optimization strategy. At each sampling time, based on the optimization performance index of the primary air volume at that time, the optimal speed setpoint for a finite period starting from that time is solved. The calculated primary fan speed control action sequence is actually executed at the current time. At the next sampling time, the optimal speed setpoint is re-solved using the above method, and optimization is continuously carried out.

[0050] (3) Feedback correction: Compare the actual output boiler load with the model's predicted boiler load to obtain the model's prediction error. Then, use the model's prediction error to correct the model's prediction value for the primary fan speed, and further obtain a more accurate prediction value for the boiler load output.

[0051] In the flue gas purification process optimization module, the predictive control models include a semi-dry desulfurization control model and an SNCR denitrification control model. The control variables for the semi-dry desulfurization control model include: the opening degree of the atomizer lime slurry valve and the tap water valve; the control variables for the SNCR denitrification control model include: the opening degree of the reducing agent valve and the soft water regulating valve; the control variables for the activated carbon control model include: the activated carbon feeder frequency; the control variables for the wet scrubbing control model include: the opening degree of the sodium hydroxide absorbent pump regulating valve and the opening degree of the sodium hydroxide dehumidifying pump regulating valve; and the control variables for the bag filter dust collector control model include: the air cannon time interval, the pulse valve time interval, and the dust removal time interval. Taking the SNCR denitrification control model as an example, the specific implementation method includes:

[0052] (1) Prediction Model: The SNCR denitrification prediction model uses the current NOx concentration information in the flue gas, along with the opening degree of the reducing agent valve and the opening degree of the soft water regulating valve, to predict the NOx concentration in the flue gas. Optionally, the prediction model can take the form of: state equation or transfer function, step response or impulse response;

[0053] (2) Rolling optimization: At each sampling time, based on the optimized performance index of the reducing agent at that time, the optimal control values ​​for the opening of the reducing agent valve and the opening of the soft water regulating valve for a finite period starting from that time are solved. The calculated control action sequence is actually executed at the current time, and at the next sampling time, the optimal control value is recalculated using the above method, continuously optimizing the process;

[0054] (3) Feedback correction: Compare the actual NOx output value with the NOx value predicted by the model to obtain the prediction error of the model. Then use the prediction error of the model to correct the prediction value of the model and obtain a more accurate NOx output prediction value.

[0055] The intelligent combustion control module integrates the combustion process and flue gas purification process, displaying the real-time control status and effects of boiler feeding control, combustion control, and flue gas purification control, and can switch between starting and stopping operations;

[0056] In the intelligent incineration control module, the image recognition algorithm includes a discharge layer thickness recognition algorithm and a fire line and combustion state detection and recognition algorithm;

[0057] Taking the image recognition-based fire detection algorithm as an example, the specific implementation steps are as follows:

[0058] (1) Images of combustion inside the furnace are collected by industrial cameras and manually labeled to form a dataset. The model is trained in a loop using a U-Net fully convolutional neural network to obtain a recognition model that can accurately identify the fire line.

[0059] (2) Deploy the model on an industrial control computer to realize the data acquisition, processing and recognition functions of the fire line identification model, and store the results;

[0060] (3) Based on historical data, when the model prediction results deviate too much from the actual results, the model will automatically optimize and redeploy.

[0061] In the intelligent incineration control module, machine learning algorithms include real-time calculation algorithms for waste calorific value, NOx prediction, conventional pollutant prediction, and dioxin prediction algorithms.

[0062] Taking the machine learning-based dioxin prediction algorithm as an example, the specific implementation steps are as follows:

[0063] (1) Time series data such as incineration conditions and dioxins are collected through the DCS system. After preprocessing, a dataset is formed. The nonlinear relationship between incineration conditions and dioxins is fitted by the autoregressive moving average model (ARIMA) and support vector machine (SVM) model to obtain a prediction model that can accurately fit dioxins.

[0064] (2) Deploy the model on an industrial control computer to realize the real-time data acquisition, processing, and identification functions of the dioxin prediction model, and store the results;

[0065] (3) Based on historical data, when the model prediction results deviate too much from the actual results, the model will automatically optimize and redeploy.

[0066] In this invention, the real-time input data of the model and algorithm described in any one of the claims is obtained by extracting DCS parameters from physical sensors at each location, thereby achieving real-time control and efficient and clean operation.

[0067] This invention also provides a highly efficient, clean, and intelligent operation device for an incinerator throughout the entire process, comprising:

[0068] Physical sensors at each location are used to acquire control parameters for each DCS and provide real-time data for calculation by each module;

[0069] The industrial control computer has a built-in full-process control module and an intelligent operation system, which enables intelligent identification and prediction of the entire process of the incinerator, controls the incinerator, and assists in the efficient and clean operation of the incinerator.

[0070] Data cables are used to connect physical sensors and industrial control computers at various locations.

[0071] like Figure 3 and Figure 4 The operating system effectively improves the stability of the incinerator operation while reducing pollutant emissions.

[0072] This invention integrates automatic combustion control and flue gas treatment control of a waste incinerator. It unifies and integrates control models for the feeding grate, tilting grate, sliding grate, primary and secondary air volume, induced draft fan, steam-water system, semi-dry desulfurization, SNCR denitrification, activated carbon, wet scrubbing, and baghouse dust collection, achieving fully automated control of the entire incineration process. By applying image recognition and machine learning algorithms, it predicts grate bed thickness, detects the fire line and combustion status, calculates the calorific value of waste in real time, and predicts common pollutants such as NOx and dioxins. This allows operators to monitor the incinerator's operating status and flue gas emission levels in real time, enabling them to adjust load strategies accordingly and achieve clean and efficient operation of the incinerator throughout the entire process.

[0073] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0074] In the description of this invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0075] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A control system for the efficient, clean, and intelligent operation of an incinerator throughout its entire process, characterized in that: include: The combustion process optimization control module is used to optimize boiler feeding and combustion control through a first intelligent prediction model and generate a first optimization result. The flue gas purification process optimization module is used to optimize boiler flue gas purification control through a second intelligent prediction model and generate a second optimization result. The intelligent incineration control module is used to predict the grate bed thickness, detect the fire line and combustion status, calculate the calorific value of waste in real time, predict NOx, predict conventional pollutants and dioxins through image recognition algorithms and machine learning algorithms, and control and optimize the entire process of the incinerator based on the first optimization results and the second optimization results. The combustion process optimization control module is further configured to, based on the collected incinerator operating data, use data affecting the incinerator's feeding grate, tilting grate, sliding grate, primary and secondary air, induced draft fan, and steam-water mixture as first feature data, and label them to generate first control variables for boiler feeding and combustion control optimization; simultaneously, based on the labels of the first feature data, obtain the first variable relationship between each of the first control variables, and assign the first variable relationship to the first feature data to generate a first dataset, train the deep learning model, and construct the first intelligent prediction model; The combustion process optimization control module is further configured to optimize boiler feeding and combustion control based on the first control variables obtained from the first intelligent prediction model, wherein the first control variables include: The control variables for the feeding grate include: pusher speed setting and speed coefficient of each unit grate; The control variables for grate turning include: the number of times each unit turns; The control variables for the sliding grate include: the number of sliding cycles in each unit and the distribution coefficient of each unit; The control variables for primary air volume include: the speed of the primary air fan in each unit; The control variables for secondary air volume include: secondary fan frequency and left and right side damper positions; The control variables for the induced draft fan include: induced draft fan frequency and induced draft fan inlet damper position; The control variables for the carbonated beverage include: the opening degree of the electric valve of the primary desuperheater and the opening degree of the electric valve of the secondary desuperheater; The flue gas purification process optimization module is further configured to, based on the collected incinerator operating data, use data affecting the semi-dry desulfurization, SNCR denitrification, activated carbon, wet scrubbing, and bag filter dust collection of the incinerator as second feature data, and label them to generate second control variables for boiler flue gas purification control optimization; simultaneously, based on the labels of the second feature data, obtain the second variable relationship between each second control variable, and assign the second variable relationship to the second feature data to generate a second dataset, train the deep learning model, and construct the second intelligent prediction model; The flue gas purification process optimization module is further configured to optimize the control of boiler flue gas purification based on the second control variables obtained from the second intelligent prediction model, wherein the second control variables include: The control variables for semi-dry deacidification include: the opening degree of the lime slurry valve of the atomizer and the opening degree of the tap water valve; The control variables for SNCR denitrification include: the opening degree of the reducing agent valve and the opening degree of the soft water regulating valve; The control variables for activated carbon include: the frequency of the activated carbon feeder; The control variables for wet washing include: the opening degree of the sodium hydroxide absorbent pump regulating valve and the opening degree of the sodium hydroxide dehumidifying pump regulating valve; The control variables for baghouse dust collectors include: air cannon time interval, pulse valve time interval, and dust removal time interval.

2. The control system for efficient, clean, and intelligent operation of an incinerator throughout the entire process, as described in claim 1, is characterized in that: The intelligent incineration control module is also used to obtain a third variable relationship between the first control variable and the second control variable based on the relationship between the first feature data and the second feature data; Based on the first variable relationship, the second variable relationship, and the third variable relationship, the first optimization result and the second optimization result are adjusted through the first intelligent prediction model and the second intelligent prediction model to control and optimize the entire process of the incinerator.

3. The control system for efficient, clean, and intelligent operation of an incinerator throughout the entire process, as described in claim 2, is characterized in that: The intelligent incineration control module is also used to detect the fire line and combustion status through a U-Net fully convolutional neural network.

4. The control system for efficient, clean, and intelligent operation of an incinerator throughout the entire process, as described in claim 3, is characterized in that: The intelligent incineration control module is also used to predict dioxins by fitting the nonlinear relationship between incineration conditions and dioxins through an autoregressive moving average model and a support vector machine model.

5. A control device for the efficient, clean, and intelligent operation of an incinerator throughout its entire process, characterized in that: include: A sensor cluster is used to collect physical control parameters at various points in the incinerator in real time, providing real-time operating data of the incinerator. An industrial control computer is used to carry the control system as described in any one of claims 1-4, to control and optimize the entire process of the incinerator, and to assist the incinerator in efficient and clean operation. A data cable is used as a data transmission medium to connect the sensor cluster and the industrial control computer.

6. The control device for efficient, clean, and intelligent operation of an incinerator throughout its entire process, as described in claim 5, is characterized in that: A data storage device is used to store the real-time operating data and the data generated by the industrial control computer; The display device is used to visualize the entire process of the incinerator, and to display the real-time operating data as well as the first and second optimization results predicted by the control system.

Images