A multimodal data-driven modeling method for municipal solid waste incineration controlled objects

Abstract

The present invention provides a multimodal data-driven modeling method for a controlled object of municipal solid waste incineration. The method relates to the technical field of MSWI processes, and includes the following steps: obtaining a global quantization value of an initial flame combustion state; obtaining virtual data under multiple operating conditions; inputting the global quantization value of the initial flame combustion state into an improved DFR combustion state quantization model in combination with real process data to obtain a final global quantization value of the flame combustion state; obtaining real operating data and expanding the real process data using VSG technology to obtain expanded data; mixing the virtual data and expanded data under multiple operating conditions to obtain multi-operating condition time series data under multiple scenarios; constructing a multi-input multi-output controlled object model based on the multi-operating condition time series data under multiple scenarios and the final global quantization value of the flame combustion state, wherein the multi-input multi-output controlled object model includes a stacked four-layer LSTM. The present invention solves the problem in the prior art that the controlled object modeling sample size of the MSWI process is small, resulting in low modeling accuracy.

Description

Technical Field

[0001] The present invention relates to the field of MSWI process technology, and in particular to a multimodal data driven municipal solid waste incineration controlled object modeling method. Background Art

[0002] The MSWI process mechanism is complex and unclear, and its controlled plant model must be constructed based on multimodal data consisting of process data and flame images. Although some ACC systems are equipped with perishable optical equipment that can detect the flame line length in real time, this measurement value does not fully represent the flame combustion state. Excellent experts in my country can achieve precise flame combustion control in some cases through visual perception of flames. Building a comprehensive database for quantifying flame combustion states and achieving both local and global quantification based on the unevenly distributed or even missing multi-condition flame images found in industrial sites, combined with furnace structure and expert knowledge, remains an open problem. While historical data on actual MSWI processes is voluminous, the conditions they can represent are difficult to describe and unevenly distributed, resulting in poor data availability. Therefore, controlled plant modeling faces the problem of small sample sizes, which requires the use of relevant technologies. Furthermore, challenges remain: how to match process data and image data under multiple conditions, how to use numerical simulation software to obtain more comprehensive virtual (mechanistic) data than in-situ conditions, and how to fuse these data to construct a controlled plant model.

[0003] The flame combustion state within the incinerator is the primary objective of combustion process control. However, domestic video acquisition and transmission to a control center based on expert visual perception make it difficult to use as a controlled variable and support modeling. This necessitates the development of a comprehensive flame combustion image library and quantitative combustion state modeling to supplement the missing controlled variable detection values. In addition to the flame combustion state, the controlled variables of the MSWI process also include furnace temperature, oxygen content, and steam flow. While these controlled variables and their corresponding operational variables can be acquired in real time, the operating conditions they represent are difficult to describe and unevenly distributed. This necessitates numerical simulation, multi-condition analysis, and virtual data generation to obtain virtual data under multiple operating conditions. The MSWI process involves real image and process data, as well as generated virtual data. Research is needed on hybrid-driven controlled object modeling that integrates multi-source data to construct a multi-input, multi-output controlled object model that is interpretable and consistent with process mechanisms. Summary of the Invention

[0004] In order to overcome the shortcomings of the existing technology, the purpose of the present invention is to provide a multimodal data-driven municipal solid waste incineration controlled object modeling method. The present invention solves the problem in the existing technology that the controlled object modeling sample volume of the MSWI process is small and the available data is poor, resulting in low modeling accuracy.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A multimodal data-driven modeling method for a municipal solid waste incineration controlled object includes:

[0007] Obtain the global quantitative value of the initial flame combustion state;

[0008] Obtain virtual data under multiple working conditions;

[0009] In combination with real process data, the initial flame combustion state global quantization value is input into the improved DFR combustion state quantization model to obtain the final flame combustion state global quantization value;

[0010] Acquire real operating data and use VSG technology to expand real process data to obtain expanded data;

[0011] Mixing the virtual data under the multiple working conditions and the expanded data to obtain multi-working condition time series data under multiple scenarios;

[0012] A multi-input multi-output controlled object model is constructed according to the multi-operating condition time series data under the multiple scenarios and the final global quantization value of the flame combustion state, wherein the multi-input multi-output controlled object model includes a stacked four-layer LSTM.

[0013] Preferably, the obtaining of virtual data under multiple working conditions includes:

[0014] Conduct MSW sampling and analyze MSW samples to obtain process input factors;

[0015] Obtain a multi-factor and multi-level orthogonal experimental table based on the multi-operating condition orthogonal experimental design and the process input factors;

[0016] Determine the boundary conditions for numerical simulation;

[0017] According to the numerical simulation boundary conditions, FLIC was used to simulate the solid-phase combustion of MSW on the grate to obtain the solid-phase combustion products.

[0018] Use FLUENT to simulate the gas phase combustion of MSW in the furnace and obtain the gas phase combustion products;

[0019] A solid-gas phase coupled combustion simulation is performed based on the solid-phase combustion products and the gas-phase combustion products. Based on the input conditions of the solid-gas phase coupled combustion simulation, ASPEN Plus is used to simulate the material and energy balance of the entire process, and virtual data under multiple operating conditions are obtained according to a multi-factor and multi-level orthogonal experimental table.

[0020] Preferably, the solid phase combustion products include:

[0021] Flue gas temperature and flue gas composition.

[0022] Preferably, the gaseous combustion products include:

[0023] Gas temperature field, velocity field and concentration field.

[0024] Preferably, the combining of the real process data and inputting the initial flame combustion state global quantization value into the improved DFR combustion state quantization model to obtain the final flame combustion state global quantization value includes:

[0025] Acquire real process data and fuse multiple frames of flame video according to the time scale of the real process data to obtain a global quantized value of the initial flame combustion state;

[0026] The initial flame combustion state global quantization value is input into the improved DFR combustion state quantization model and aligned with the real process data to perform data combination to obtain the final flame combustion state global quantization value.

[0027] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0028] The present invention provides a multimodal data-driven municipal solid waste incineration controlled object modeling method, comprising:

[0029] Obtaining an initial global quantization value of the flame combustion state; obtaining virtual data under multiple working conditions; combining the actual process data, inputting the initial global quantization value of the flame combustion state into an improved DFR combustion state quantization model to obtain a final global quantization value of the flame combustion state; obtaining actual operation data and expanding the actual process data using VSG technology to obtain expanded data; mixing the virtual data under multiple working conditions and the expanded data to obtain multi-working condition time series data under multiple scenarios; constructing a multi-input multi-output controlled object model based on the multi-working condition time series data under multiple scenarios and the final global quantization value of the flame combustion state, wherein the multi-input multi-output controlled object model includes a stacked four-layer LSTM. The present invention expands the real data using VSG technology and mixes it with the multi-working condition virtual data to form multi-scenario time series data, thereby increasing the data diversity during model training, helping to capture complex combustion process characteristics, increasing the modeling sample volume, and enhancing the accuracy of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0031] Figure 1This is a flow chart of a multimodal data-driven municipal solid waste incineration controlled object modeling method provided by the present invention. DETAILED DESCRIPTION

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0034] like Figure 1 As shown, the present invention provides a multimodal data driven municipal solid waste incineration controlled object modeling method, comprising:

[0035] Step 100: Obtaining a global quantized value of the initial flame combustion state;

[0036] Step 200: Acquire virtual data under multiple working conditions;

[0037] Step 300: In combination with real process data, the initial flame combustion state global quantization value is input into the improved DFR combustion state quantization model to obtain the final flame combustion state global quantization value;

[0038] Step 400: Acquire real operating data and use VSG technology to expand the real process data to obtain expanded data;

[0039] Step 500: Mixing the virtual data under the multiple working conditions and the expanded data to obtain multi-working condition time series data under multiple scenarios;

[0040] Step 600: Construct a multi-input multi-output controlled plant model according to the multi-operating condition time series data under the multiple scenarios and the final global quantized value of the flame combustion state, wherein the multi-input multi-output controlled plant model includes a stacked four-layer LSTM.

[0041] Specifically, based on the expert experience of field engineers, the geometric segmentation theory is used to qualitatively analyze the combustion characteristics (flame height, brightness and length) presented in the flame image. The quantitative value obtained by the first module is one of the outputs of the modeling of the third module. Then, the typical images are manually annotated to obtain a limited small sample data set. The flame image data is automatically calibrated using the visual Transformer. GAN is used in combination with the flame imaging mechanism to obtain atypical images. A complete flame combustion image quantization database is established. The calibrated complete flame combustion images are matched with the operational variables of the same time scale to obtain a training data set. A multi-input and multi-output deep forest regression (DFR) model that mixes multi-layer stacking learning and selective attention mechanism is constructed. Feature fusion is used in the output layer to realize the mapping between the local quantitative value and the comprehensive quantitative value of the flame combustion state. Based on the component attributes of MSW and M The combustion characteristics of the SWI process are analyzed for key factors, and multi-condition operation scenarios are designed based on domain expert knowledge. On this basis, a multi-factor and multi-level orthogonal experimental design is used to obtain an orthogonal test table. Considering the physical and chemical properties of solid-phase combustion, gas-phase fluid dynamics, and full-process mass-energy balance reactions contained in the MSWI process, a multi-numerical simulation software coupling strategy is used to construct a full-process simulation model. Experiments are carried out based on the orthogonal test table to obtain virtual (mechanism) data under multiple conditions. For the MSWI multi-condition virtual (mechanism) data obtained by multi-software coupling and the real data of benchmark conditions selected from industrial sites, a data fusion mechanism based on domain expert knowledge and virtual sample generation (VSG) technology are used to obtain a multi-scenario sample library for modeling. The serial integrated model structure is designed according to the process mechanism, and a multi-input and multi-output timing characteristic controlled object model is constructed based on the obtained sample library.

[0042] Furthermore, expert knowledge is used to calibrate flame combustion images. Based on domain expert knowledge and operational experience, flame combustion images are quantized based on geometric theory. The flame combustion state is quantized into three local quantization values ​​of height, brightness, and length and one global quantization value of the flame combustion state. Then, the combustion image is manually labeled according to the four quantization targets to obtain a typical flame combustion image quantization dataset; the flame combustion image automatic calibration model takes the obtained typical image dataset as input, pre-processes the image through image linear block and linear mapping, and then trains and obtains a flame combustion image automatic calibration model that combines Transformer encoder and multi-layer perceptron; unlabeled flame combustion image For example, automatic calibration is performed to extract the frame rate of the flame video to obtain the flame combustion image data to be labeled, which is input into the automatic calibration model to obtain the calibrated flame combustion image quantization dataset; the mechanism knowledge and generative adversarial network (GAN) pseudo-label the atypical flame image quantization set, combined with the mapping mechanism of three-dimensional spatial imaging in the furnace, and the typical image dataset is used to obtain pseudo-labeled atypical images. Combined with the unlabeled flame image, the improved cycle consistency GAN and the generated image evaluation criterion are used to obtain the atypical image dataset; a complete flame combustion state quantization database is constructed, and the typical image dataset, the automatic calibration image dataset and the atypical image dataset are merged to obtain a complete flame combustion state quantization value database.

[0043] A dataset was constructed with the quantized value of the flame combustion state as the dependent variable (true value) and the wind and cloth operation variables as the independent variables; the DFR model structure was determined, including the type, number and number of stacked layers of the forest algorithm in each layer; the attention mechanism DFR model was selected to construct a multi-input and multi-output improved DFR model with the three local quantized values ​​of flame height, brightness and length as output, and the selective attention mechanism was used between layers to improve the feature expression ability; the flame height, brightness and length values ​​were fused, and a linear fusion strategy was adopted based on the three local quantized values ​​of height, brightness and length, and the final global quantized value of the flame combustion state was obtained by combining the operation process state information.

[0044] Furthermore, the obtaining of virtual data under multiple working conditions includes:

[0045] Conduct MSW sampling and analyze MSW samples to obtain process input factors;

[0046] Obtain a multi-factor and multi-level orthogonal experimental table based on the multi-operating condition orthogonal experimental design and the process input factors;

[0047] Determine the boundary conditions for numerical simulation;

[0048] According to the numerical simulation boundary conditions, FLIC was used to simulate the solid-phase combustion of MSW on the grate to obtain the solid-phase combustion products.

[0049] Use FLUENT to simulate the gas phase combustion of MSW in the furnace and obtain the gas phase combustion products;

[0050] A solid-gas phase coupled combustion simulation is performed based on the solid-phase combustion products and the gas-phase combustion products. Based on the input conditions of the solid-gas phase coupled combustion simulation, ASPENPlus is used to simulate the material and energy balance of the entire process and virtual data under multiple working conditions are obtained according to a multi-factor and multi-level orthogonal experimental table.

[0051] Specifically, for MSW sampling, the fermented and dehydrated MSW is manually sampled based on the four-part method proposed in the "Methods for Sampling and Analysis of Municipal Waste" to obtain MSW samples for analysis; for MSWI process input factor analysis, the obtained MSW samples are tested offline to analyze the MSW calorific value (low calorific value), industrial analysis (moisture, ash, volatile matter, fixed carbon) and elemental analysis (C, H, O, N, S) data as well as the feasible intervals of MSW composition factors; for MSWI process operation factor analysis, based on the three states of low load, full load and overload, the possible operating conditions during the operation of the MSWI process are analyzed based on the knowledge of domain experts, and the feasible intervals of the operation factors are obtained; for multi-condition orthogonal experiment design, the MSW composition factors and operation factors are used as factors of the multi-condition orthogonal experiment, and the number of levels of the orthogonal experiment is calculated based on the feasible interval, and then the statistical analysis software is used to obtain the multi-factor and multi-level orthogonal experiment table to provide support for the full-process numerical simulation.

[0052] More specifically, determine the boundary conditions of numerical simulation, and perform numerical simulation on solid phase combustion, gas phase combustion and material energy balance according to the process design of actual MSWI process, and determine the input boundary conditions of orthogonal experiment based on the operation process; for MSW solid phase combustion simulation, use FLIC to simulate the solid phase combustion of MSW on the grate, obtain the flue gas data generated by solid phase combustion, the boundary conditions of gas phase combustion and the flue gas temperature and flue gas components required for material balance simulation; for MSW gas phase combustion simulation, use FLUENT to simulate the gas phase combustion of MSW in the furnace, obtain the furnace radiation data and material balance required for solid phase MSW combustion, and the gas temperature field, velocity field and concentration field required for simulation; for solid-gas coupled combustion simulation, interact with solid-gas phase combustion model until the boundary conditions and radiation data converge, and obtain the flue gas temperature, flue gas components, gas temperature field, velocity field and concentration field; for full process material balance simulation, use ASPEN based on the input conditions of solid-gas coupled combustion simulation. Plus simulates the material and energy balance of the entire process, obtains a numerical simulation model of the entire process, and generates furnace temperature, steam flow, oxygen content, and pollution emission data; conducts multi-condition orthogonal experimental research, and performs full-process numerical simulation according to the obtained multi-factor multi-level orthogonal experimental table, thereby generating virtual data under multiple conditions.

[0053] Furthermore, the initial flame combustion state global quantization value is input into the improved DFR combustion state quantization model in combination with the real process data to obtain the final flame combustion state global quantization value, including:

[0054] Acquire real process data and fuse multiple frames of flame video according to the time scale of the real process data to obtain a global quantized value of the initial flame combustion state;

[0055] The initial flame combustion state global quantization value is input into the improved DFR combustion state quantization model and aligned with the real process data to perform data combination to obtain the final flame combustion state global quantization value.

[0056] Specifically, the flame combustion image state quantization value calibration takes the time scale of the real process data as the benchmark, fuses multiple frames of the flame video, obtains the flame combustion image that needs to be quantized, inputs the improved DFR combustion state quantization model, and obtains the global quantization value of the flame combustion state; the flame combustion image data is aligned with the real process data, and the obtained global quantization value of the flame combustion state is combined with the current real process data to obtain the operating condition data set containing the control variables and the controlled variables at the current moment; the real data distribution boundary filling based on virtual sample generation (VSG) is used to store the obtained real operating condition data. In order to solve the problem of uneven distribution, the virtual (mechanism) data generated by numerical simulation under similar operating conditions are combined with VSG technology to expand the real operating data and obtain the complete distribution data under the real benchmark conditions; multi-condition virtual data preprocessing, anomaly detection and abnormal sample deletion processing are performed on the obtained multi-condition virtual data, and then the multi-condition data are segmented to obtain multiple single-condition data. Combined with the flame combustion image complete database constructed in the previous article, the same strategy as above is used to expand the local and global quantitative values ​​of flame combustion to obtain complete distribution data under multiple virtual single conditions; multi-condition data scenario Classification, in order to restore the real MSWI process operation state, based on the domain expert knowledge and MSWI process operation characteristics, in accordance with the multi-scenario requirements and data time series, the multi-condition virtual data and the benchmark real process data are mixed and recombined to obtain multi-condition time series data under multiple scenarios; the multi-input and multi-output controlled object model structure design, according to the MSWI process flow, the serial integrated output model structure is designed, the order is: flame combustion state → furnace temperature → steam flow → oxygen content; in the flame combustion state model part, the air distribution and material distribution control quantity is used as input to construct the first layer of LSTM, and the flame combustion state quantization value output is obtained. Output; In the furnace temperature model part, the air distribution and material distribution control value and the quantized value of the flame combustion state are used as input to construct the second layer of LSTM, and the furnace temperature output is obtained; in the steam flow model part, the air distribution and material distribution control value, the flame combustion state and the furnace temperature are used as input to construct the third layer of LSTM, and the steam flow output is obtained; in the oxygen content model part, the air distribution and material distribution control value, the flame combustion state, the furnace temperature and the steam flow are used as input to construct the fourth layer of LSTM, and the oxygen content output is obtained; in the multi-input and multi-output controlled object model, a controlled object model with timing characteristics is obtained by serially stacking four layers of LSTM.

[0057] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0058] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A multimodal data driven modeling method for a municipal solid waste incineration controlled object, characterized in that: include: Obtaining a global quantization value of an initial flame combustion state, specifically comprising: obtaining real process data and fusing multiple frames of a flame video according to a time scale of the real process data to obtain a global quantization value of the initial flame combustion state; Obtain virtual data under multiple working conditions; The obtaining of virtual data under multiple working conditions includes: Conduct MSW sampling and analyze MSW samples to obtain process input factors; Obtain a multi-factor and multi-level orthogonal experimental table based on the multi-operating condition orthogonal experimental design and the process input factors; Determine the boundary conditions for numerical simulation; According to the numerical simulation boundary conditions, FLIC was used to simulate the solid-phase combustion of MSW on the grate to obtain the solid-phase combustion products. Use FLUENT to simulate the gas phase combustion of MSW in the furnace and obtain the gas phase combustion products; Performing a solid-gas phase coupled combustion simulation based on the solid-phase combustion products and the gas-phase combustion products, and based on the input conditions of the solid-gas phase coupled combustion simulation, using ASPEN Plus to simulate the material and energy balance of the entire process and obtaining virtual data under multiple operating conditions based on a multi-factor and multi-level orthogonal experimental table; In combination with real process data, the initial flame combustion state global quantization value is input into the improved DFR combustion state quantization model to obtain the final flame combustion state global quantization value; The method of obtaining the final global quantization value of the flame combustion state specifically includes: constructing a data set using the quantization value of the flame combustion state as a dependent variable and the wind and material operation variable as an independent variable; determining the structure of a DFR model, including the type, number, and number of stacked layers of the forest algorithm for each layer; selecting an attention mechanism DFR model to construct a multi-input multi-output improved DFR model using three local quantization values ​​of flame height, brightness, and length as outputs, and using a selective attention mechanism between layers to improve the expressiveness of features; fusing the flame height, brightness, and length values, using a linear fusion strategy based on the three local quantization values ​​of height, brightness, and length, and combining the operation process state information to obtain the final global quantization value of the flame combustion state; Acquire real operating data and use VSG technology to expand real process data to obtain expanded data; Mixing the virtual data under the multiple working conditions and the expanded data to obtain multi-working condition time series data under multiple scenarios; Constructing a multi-input multi-output controlled plant model according to the multi-operating condition time series data under the multiple scenarios and the final global quantized value of the flame combustion state, wherein the multi-input multi-output controlled plant model includes a stacked four-layer LSTM; Constructing a multi-input multi-output controlled object model specifically includes: The structure design of the multi-input and multi-output controlled object model is based on the MSWI process flow to design a serial integrated output model structure, and the order is: flame combustion state → furnace temperature → steam flow → oxygen content; in the flame combustion state model part, the air distribution and material distribution control value is used as input to construct the first layer of LSTM, and the flame combustion state quantization value output is obtained; in the furnace temperature model part, the air distribution and material distribution control value and the flame combustion state quantization value are used as input to construct the second layer of LSTM, and the furnace temperature output is obtained; in the steam flow model part, the air distribution and material distribution control value, flame combustion state and furnace temperature are used as input to construct the third layer of LSTM, and the steam flow output is obtained; in the oxygen content model part, the air distribution and material distribution control value, flame combustion state, furnace temperature and steam flow are used as input to construct the fourth layer of LSTM, and the oxygen content output is obtained; the multi-input and multi-output controlled object model obtains a controlled object model with timing characteristics by serially stacking four layers of LSTM.

2. A multimodal data driven municipal solid waste incineration controlled object modeling method according to claim 1, characterized in that: The solid phase combustion products include: Flue gas temperature and flue gas composition.

3. The multimodal data driven municipal solid waste incineration controlled object modeling method according to claim 1 is characterized in that: The gas phase combustion products include: Gas temperature field, velocity field and concentration field.

4. The multimodal data driven municipal solid waste incineration controlled object modeling method according to claim 1 is characterized in that: The initial flame combustion state global quantization value is input into the improved DFR combustion state quantization model in combination with the real process data to obtain the final flame combustion state global quantization value, including: The initial flame combustion state global quantization value is input into the improved DFR combustion state quantization model and aligned with the real process data to perform data combination to obtain the final flame combustion state global quantization value.

Images