A cost prediction method based on a solid waste heat treatment neural network model

Abstract

The application discloses a cost prediction method based on a solid waste heat treatment neural network model and belongs to the technical field of solid waste heat treatment.The application is used for solving the problem of accurate prediction of the cost of solid waste heat treatment.A first hidden layer of a solid waste heat treatment prediction model based on deep learning is constructed; a second hidden layer of the solid waste heat treatment prediction model based on deep learning is constructed; an output layer of the solid waste heat treatment prediction model based on deep learning is constructed; a loss function of the solid waste heat treatment prediction model based on deep learning is designed; the input layer, the first hidden layer, the second hidden layer and the output layer are sequentially connected to obtain the solid waste heat treatment prediction model based on deep learning; based on the prediction result after reverse normalization, the prediction result of the solid waste heat treatment prediction model based on deep learning with actual physical quantity is obtained, and the cost prediction method based on the solid waste heat treatment neural network model is constructed.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of solid waste heat treatment, and particularly relates to a cost prediction method based on a solid waste heat treatment neural network model. BACKGROUND

[0002] The solid waste heat treatment process involves complex conversion of matter and energy. First, the components of solid waste have high heterogeneity and uncertainty, including combustibles, moisture, inorganics, and other ingredients, the proportions of which change over time and source. Second, drying, pyrolysis, combustion, and other physical and chemical processes occur simultaneously during the heat treatment process, and these processes have significant coupling effects. The drying process affects the subsequent pyrolysis and combustion effects, the composition of pyrolysis products affects the combustion characteristics, and the combustion heat release in turn affects drying and pyrolysis. At the same time, the distribution of temperature field, flow field, and concentration field also interact with each other, forming a complex multi-field coupling. In addition, there are multi-scale processes such as heat and mass transfer, chemical reaction, and phase change in the furnace, and the nonlinear characteristics and dynamic changes of these processes bring great challenges to process control.

[0003] In the solid waste heat treatment process, conversion efficiency directly reflects the degree of material conversion and is related to the completeness of incineration and the generation of pollutants. For example, carbon conversion rate affects the completeness of combustion and carbon dioxide emissions, and sulfur and chlorine conversion rates are closely related to the generation of acid gases. Product distribution determines the pollution control strategy and resource utilization direction, and accurate prediction helps to optimize the subsequent treatment process. Energy conversion reflects the level of energy utilization, which is of great significance to improve system efficiency and reduce operating costs. Accurate prediction of these indicators can provide decision-making basis for process optimization and intelligent control, and is of great significance to improve system operation efficiency, reduce pollutant emissions, and ensure safe and stable operation.

[0004] Currently, the prediction of key indicators in the solid waste heat treatment process mainly has the following shortcomings: First, the prediction accuracy is generally not high, especially when the working conditions fluctuate greatly, the predicted results deviate significantly from the actual values. Second, the existing models consider only a single influencing factor, often focusing on only some major process parameters, while ignoring the influence of important factors such as material properties and environmental conditions, resulting in poor adaptability of the model to complex working conditions. In addition, most models only predict a single or a few indicators, and lack comprehensive prediction capability for conversion efficiency, product distribution, and energy conversion of multiple key indicators. It is urgent to establish a comprehensive model that considers more comprehensive factors and has better prediction effect to meet the actual engineering needs.

[0005] On the other hand, the solid waste heat treatment cost prediction method is the inevitable demand for scientific management of modern solid waste treatment facilities. The lack of cost prediction method will make enterprises unable to accurately evaluate the treatment cost of solid waste with different characteristics, resulting in lack of scientific basis for treatment strategy making; it is difficult to predict the influence of process parameter adjustment on cost, which increases the risk of operation decision; it is difficult to respond to the cost change caused by the fluctuation of solid waste characteristics in time, which affects the economic benefit of enterprises. SUMMARY

[0006] The problem to be solved by the present application is the accurate prediction of the cost of solid waste heat treatment, and a cost prediction method based on a solid waste heat treatment neural network model is proposed.

[0007] To achieve the above-mentioned purpose, the present application realizes by the following technical scheme:

[0008] A cost prediction method based on a solid waste heat treatment neural network model, comprising the following steps:

[0009] S1. Determine the input parameters of the input layer of the solid waste heat treatment prediction model based on deep learning, including material characteristic parameters, process operation parameters and process parameters;

[0010] S2. Construct the first hidden layer in the main layer of the solid waste heat treatment prediction model based on deep learning;

[0011] S3. Construct the second hidden layer in the main layer of the solid waste heat treatment prediction model based on deep learning, and the first hidden layer is connected to the second hidden layer;

[0012] S4. Construct the output layer of the solid waste heat treatment prediction model based on deep learning, divide the output layer into conversion efficiency group, product distribution group and energy conversion group, and determine the output parameters of the output layer of the solid waste heat treatment prediction model based on deep learning;

[0013] S5. Design the loss function of the solid waste heat treatment prediction model based on deep learning;

[0014] S6. Connect the input layer, main layer and output layer of the solid waste heat treatment prediction model based on deep learning in sequence to obtain the solid waste heat treatment prediction model based on deep learning;

[0015] S7. Collect data through the solid waste treatment plant and the automatic control system, and normalize the data; divide the data into training set, validation set and test set according to the ratio of 7:2:1, train, validate and test the solid waste heat treatment prediction model based on deep learning obtained in step S6 to obtain the trained solid waste heat treatment prediction model based on deep learning;

[0016] S8. Based on the prediction results of the trained deep learning-based solid waste thermal treatment prediction model, the prediction results of the deep learning-based solid waste thermal treatment prediction model with actual physical quantities are obtained after inverse normalization, and a cost prediction method based on the solid waste thermal treatment neural network model is constructed.

[0017] Further, the number of nodes of the input layer in step S1 is determined as 35, respectively denoted as X1-X 35 ;

[0018] X1~X 14 is a material characteristic parameter, and the specific parameters are as follows: X1 is moisture content, X2 is ash content, X3 is combustible content, X4 is carbon content, X5 is hydrogen content, X6 is oxygen content, X7 is nitrogen content, X8 is sulfur content, X9 is chlorine content, X 10 is calorific value, X 11 is bulk density, X 12 is particle size, X 13 is sand content, X 14 is metal content;

[0019] X 15 ~X 25 is a process operating parameter, and the specific parameters are as follows: X 15 is grate speed, X 16 is feed rate, X 17 is primary air volume, X 18 is secondary air volume, X 19 is furnace temperature, X 20 is furnace negative pressure, X 21 is flue gas temperature, X 22 is steam temperature, X 23 is steam pressure, X 24 is steam flow, X 25 is ambient temperature;

[0020] X 26 ~X 35 is a process parameter, and the specific parameters are as follows: X 26 is feed moisture content, X 27 is discharge moisture content, X 28 is feed volatile matter, X 29 is discharge volatile matter, X 30 is feed mass flow rate, X 31 is discharge mass flow rate, X 32 is feed calorific value, X 33 is discharge calorific value, X 34 is smoke oxygen concentration, X 35 is ash content.

[0021] Further, step S2 divides the nodes of the first hidden layer into a material property group, a reaction property group, and an energy property group.

[0022] The specific implementation method of step S2 includes the following steps:

[0023] S2.1. Temperature correction and material moisture correction are introduced in the construction of the material property group. The temperature correction considers the driving effect of temperature on the conversion rate of the material and reflects the differences in the conversion law of the material in different temperature intervals. The material moisture correction considers the influence of the water evaporation process on the subsequent conversion;

[0024] S2.2. The mass conversion term is introduced in the construction of the reaction property group to directly reflect the reaction degree of the material and embody the depth of the reaction. The air excess term is added to represent the sufficiency of the combustion environment and reflect the influence of the oxidation reaction condition on the conversion effect;

[0025] S2.3. The heat value conversion term is introduced in the construction of the energy property group to directly represent the energy conversion effect. The temperature gradient term is added to reflect the driving force and intensity of heat transfer and describe the spatial distribution law of energy.

[0026] Further, step S3 divides the nodes of the second hidden layer into a drying process group, a pyrolysis process group, and a combustion process group. The drying process group describes the influence of the drying process on the subsequent pyrolysis and combustion, contains 15 neuron nodes, and the ith4 neuron node takes a value of 1-15. The pyrolysis process group describes the composition change characteristics of the pyrolysis products, contains 20 neuron nodes, and the ith5 neuron node takes a value of 16-35. The combustion process group describes the process characteristics of the final oxidation stage, contains 15 neuron nodes, and the ith6 neuron node takes a value of 36-50.

[0027] The specific implementation method of step S3 includes the following steps:

[0028] S3.1. The drying process group is constructed to represent the process characteristics of water removal in solid waste, reflect the driving effect of temperature on water evaporation, and describe the influence of the drying process on the subsequent pyrolysis and combustion;

[0029] S3.2. The pyrolysis process group is constructed to reflect the influence law of temperature on the volatilization, describe the composition change characteristics of the pyrolysis products, and realize the extraction of the key characteristics of the pyrolysis stage through the kinetic characteristics of organic matter pyrolysis decomposition;

[0030] S3.3. The combustion process group is constructed to represent the oxidation process of fixed carbon and volatile matter, reflect the influence of oxygen concentration on the combustion effect, and describe the process characteristics of the final oxidation stage

[0031] Further, step S4 divides the output layer into conversion efficiency group, product distribution group and energy conversion group based on the evaluation needs of the solid waste thermal treatment process; the conversion efficiency group indexes reflect the integrity and sufficiency of material conversion, the product distribution group indexes reveal the migration law of material and environmental impact, and the energy conversion group indexes embody the energy utilization level;

[0032] The conversion efficiency group indexes include Y1-Y6, Y1 is water conversion rate, Y2 is volatile conversion rate, Y3 is fixed carbon conversion rate, Y4 is carbon conversion rate, Y5 is sulfur conversion rate, and Y6 is chlorine conversion rate;

[0033] The product distribution group indexes include material distribution proportion indexes Y 71 ~Y 75 and residue distribution indexes Y8-Y9, Y 71 is the mass distribution proportion of water after drying into flue gas, Y 72 is the mass distribution proportion of volatile after pyrolysis into flue gas, Y 73 is the distribution proportion of carbon converted into carbon dioxide into flue gas, Y 74 is the distribution proportion of sulfur converted into sulfur dioxide into flue gas, Y 75 is the distribution proportion of chlorine converted into hydrogen chloride into flue gas; Y8 is the mass distribution proportion of fixed carbon remaining in the slag after combustion, and Y9 is the mass distribution proportion of fly ash component into fly ash after combustion of fixed carbon;

[0034] The energy conversion group indexes include Y 10 ~Y 12 , Y 10 is thermal efficiency, Y 11 is combustion efficiency, and Y 12 is heat loss rate.

[0035] Further, the loss function of step S5 considers the prediction loss and the thermal treatment loss, and the specific implementation method includes the following steps:

[0036] S5.1. Constructing a prediction loss based on conversion efficiency prediction loss, product distribution prediction loss and energy conversion prediction loss, the expression is:

[0037]

[0038] Among them, is the conversion efficiency prediction loss; is the product distribution prediction loss; is the energy conversion prediction loss;

[0039]

[0040] Among them, is the mean square error calculation function; is the model predicted conversion efficiency group different index output value; is the actual measured conversion efficiency group different index output value;

[0041]

[0042] wherein, is the model predicted product distribution group different index output value; is the actual measured product distribution group different index output value;

[0043]

[0044] wherein, is the model predicted energy conversion group different index output value; is the actual measured energy conversion group different index output value;

[0045] S5.2. Constructing heat treatment loss based on material conversion constraint, component conversion constraint and heat conversion constraint , the expression is:

[0046]

[0047] wherein, is the material conversion constraint, is the weight corresponding to the material conversion constraint; is the component conversion constraint, is the weight corresponding to the component conversion constraint; is the heat conversion constraint, is the weight corresponding to the heat conversion constraint;

[0048] The material conversion constraint is established as:

[0049]

[0050] wherein, is the volatile matter conversion term; is the fixed carbon distribution term;

[0051] The component conversion constraint is established as:

[0052]

[0053] wherein, is the sulfur element conversion constraint, is the chlorine element conversion constraint;

[0054] The heat conversion constraint is established as:

[0055] ;

[0056] S5.3. Constructing total loss based on predicted loss and heat treatment loss ;

[0057]

[0058] wherein, is the weight corresponding to the predicted loss, is the weight corresponding to the heat treatment loss, which is determined by expert experience.

[0059] Further, the model training in step S7 adopts a back propagation algorithm; in the training process, the model calculates a predicted value through forward propagation, substitutes the predicted value and a true value into a loss function to calculate a loss, and then updates network parameters through back propagation; to prevent overfitting, an early stopping strategy is adopted, and when the loss of a verification set does not decrease for 10 consecutive epochs, the training is stopped; and the output result is restored to an actual physical quantity through inverse normalization.

[0060] Further, step S8 considers the close correlation among material conversion, energy utilization and environmental influence in the system operation process, and comprehensively evaluates the influence of material conversion efficiency on power generation cost, the reagent consumption caused by different material conversion paths, the cascade utilization benefits of primary and secondary energy recovery, the equipment full life cycle cost and the environmental treatment cost.

[0061] Advantages of the present application:

[0062] The cost prediction method based on the solid waste heat treatment neural network model has been designed and optimized for the solid waste heat treatment characteristics in the input parameters, the first hidden layer, the second hidden layer and the output layer, and the scientificity and rationality of the prediction of the conversion efficiency, the product distribution and the energy conversion in the solid waste heat treatment process are realized.

[0063] The cost prediction method based on the solid waste heat treatment neural network model establishes a comprehensive cost evaluation system, which includes five sub-models of the power generation cost under the influence of the conversion efficiency, the reagent cost of the material conversion path, the energy cascade utilization benefit, the equipment full life cycle cost and the environmental cost. The method provides a reliable calculation method for the solid waste heat treatment cost prediction, and can help enterprises to evaluate the treatment cost under different solid waste characteristics and operation conditions in advance, and provide a scientific basis for operation optimization and decision-making. BRIEF DESCRIPTION OF DRAWINGS

[0064] Figure 1 The flowchart of the cost prediction method based on the solid waste heat treatment neural network model is shown in the drawings.

[0065] Figure 2 Loss curve diagram for the solid waste heat treatment prediction model training process of the present application;

[0066] Figure 3 Comparison chart of predicted value and true value of the solid waste heat treatment prediction model of the present application;

[0067] Figure 4 Total cost calculation result chart under different working conditions of the present application. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical scheme and advantages of the present application clearer and more understandable, the present application will be further described in detail below in combination with the drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application, that is, the described specific embodiments are only a part of the embodiments of the present application, but not all the specific embodiments. The components of the specific embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations, and the present application can also have other embodiments.

[0069] Therefore, the detailed description of the specific embodiments of the present application provided in the drawings below is not intended to limit the scope of the claimed present application, but only represents selected specific embodiments of the present application. Based on the specific embodiments of the present application, all other specific embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0070] In order to further understand the invention content, characteristics and effects of the present application, the following specific embodiments are exemplified, and the drawings are Figure 1 - the drawings Figure 4 The detailed description is as follows:

[0071] Example 1:

[0072] A cost prediction method based on a solid waste heat treatment neural network model, comprising the following steps:

[0073] S1. Determine the input parameters of the input layer of the solid waste heat treatment prediction model based on deep learning, including material characteristic parameters, process operation parameters and process parameters;

[0074] In step S1, the number of nodes of the input layer is determined to be 35, respectively denoted as X1-X 35 ;

[0075] X1~X 14For material characteristic parameters, the corresponding specific parameters are as follows: X1 is moisture content, X2 is ash content, X3 is combustible content, X4 is carbon content, X5 is hydrogen content, X6 is oxygen content, X7 is nitrogen content, X8 is sulfur content, X9 is chlorine content, X 10 is heat value, X 11 is bulk density, X 12 is particle size, X 13 is sand content, X 14 is metal content;

[0076] X 15 ~X 25 For process operation parameters, the corresponding specific parameters are as follows: X 15 is grate speed, X 16 is feed rate, X 17 is primary air volume, X 18 is secondary air volume, X 19 is hearth temperature, X 20 is hearth negative pressure, X 21 is flue gas temperature, X 22 is steam temperature, X 23 is steam pressure, X 24 is steam flow, X 25 is ambient temperature;

[0077] X 26 ~X 35 For process parameters, the corresponding specific parameters are as follows: X 26 is feed moisture content, X 27 is discharge moisture content, X 28 is feed volatile matter, X 29 is discharge volatile matter, X 30 is feed mass flow rate, X 31 is discharge mass flow rate, X 32 is feed heat value, X 33 is discharge heat value, X 34 is fume oxygen concentration, X 35 is ash content.

[0078] Further, among the above parameters, all can be directly collected by an automatic control system. The data is derived from historical data of the solid waste treatment plant. In order to ensure the training effect and prediction accuracy of the deep learning model, it is necessary to normalize the input parameters and output parameters. Normalization processing can eliminate the influence of different dimensions on the model, so that all features are in the same scale range, which helps to speed up the model convergence speed and improve the model performance. At the same time, the normalization processing can also avoid the problem of numerical overflow or underflow in numerical calculation. For the element content indicators X1~X9 in the material characteristic parameters, since they are all mass percentages, they are directly divided by 100 to map to the 0-1 interval; X 10 12 Then linearly map based on the maximum and minimum value range of historical data statistics; X 13 and X 14 are directly divided by 100 as percentages. In the process operation parameters, X 15 and X 16 are mapped based on the rated range of the equipment, X 17 and X 18 are mapped based on the rated range of the blower, X 19 ~ X 25 are all linearly mapped based on the statistical range of historical operation data. In the process parameters, X 26 ~ X 29 are directly divided by 100 as percentages, X 30 , X 31 are mapped based on the design operating condition range, X 32 , X 33 are mapped based on the historical data range, X 34 is mapped based on the theoretical air excess coefficient range, and X 35 is directly divided by 100 as a percentage. Through such normalization processing, all parameters are uniformly mapped to the 0-1 interval, laying a foundation for subsequent training and prediction of the deep learning model.

[0079] S2. Constructing a first hidden layer in the main body layer of the solid waste thermal treatment prediction model based on deep learning;

[0080] Further, step S2 divides the nodes of the first hidden layer into a material characteristic group, a reaction characteristic group and an energy characteristic group. The material characteristic group focuses on the physical and chemical property changes of the material in the thermal treatment process, contains 30 neuron nodes, and the ith neuron node takes a value of 1~30. The reaction characteristic group focuses on describing the degree of progress of various chemical reactions, contains 25 neuron nodes, and the ith neuron node takes a value of 31~55. The energy characteristic group focuses on representing energy conversion and transmission characteristics, contains 25 neuron nodes, and the ith neuron node takes a value of 56~80. ​

[0081] The specific implementation method of step S2 includes the following steps:

[0082] S2.1. Temperature correction and material moisture correction are introduced in the material property group. The temperature correction considers the driving effect of temperature on material conversion rate and reflects the difference in material conversion law in different temperature intervals. The material moisture correction considers the influence of water evaporation process on subsequent conversion. The calculation method of 30 neuron nodes in the material property group is as follows:

[0083]

[0084] wherein, represents the output value of the i1th neuron node in the first hidden layer, i1 takes a value of 1~30; is an activation function, used to introduce nonlinear characteristics; is the weight coefficient of the i1th neuron node in the first hidden layer corresponding to the jth input parameter; is the input parameter, ; is the bias term of the i1th neuron node in the first hidden layer corresponding to the jth input parameter; is a temperature correction coefficient, determined by expert experience, design documents, experimental research, etc. is a moisture correction coefficient, determined by expert experience, design documents, experimental research, etc. is a temperature correction term; is a material moisture correction term;

[0085] S2.2. The mass conversion term is introduced in the reaction property group, which directly reflects the reaction degree of the material and embodies the depth of the reaction. The air excess term is added to represent the sufficiency of the combustion environment and reflect the influence of oxidation reaction conditions on the conversion effect. The calculation method of 25 neuron nodes in the reaction property group is as follows:

[0086]

[0087] wherein, is the output value of the i2th neuron node in the first hidden layer, i2 takes a value of 31~55; is the weight coefficient of the i2th neuron node in the first hidden layer corresponding to the jth input parameter; is the bias term of the i2th neuron node in the first hidden layer corresponding to the jth input parameter; is a mass conversion coefficient, determined by expert experience, design documents, experimental research, etc. is an air excess coefficient, determined by expert experience, design documents, experimental research, etc. is air density, determined by actual measurement; for mass conversion term; for air excess term;

[0088] S2.3. Constructing energy characteristic group to introduce heat value conversion term to directly represent energy conversion effect; adding temperature gradient term to reflect driving force and transfer strength of heat transfer and describe energy distribution in space; 25 neuron nodes in energy characteristic group are calculated as follows:

[0089]

[0090] wherein, is output value of the i3th neuron node in the first hidden layer, i3 takes value of 56-80; is weight coefficient of the jth input parameter and the i3th neuron node in the first hidden layer corresponding to the jth input parameter; is bias term of the i3th neuron node in the first hidden layer corresponding to the jth input parameter; is heat value conversion term; is heat value conversion coefficient, which is determined by expert experience, design file, test research and the like; is temperature gradient term; is temperature gradient coefficient, which can be determined by expert experience, design file, test research and the like.

[0091] S3. Constructing second hidden layer in main body layer of solid waste heat treatment prediction model based on deep learning, the first hidden layer being connected to the second hidden layer;

[0092] Further, step S3 divides nodes of the second hidden layer into drying process group, pyrolysis process group and combustion process group, wherein the drying process group describes influence of drying process on subsequent pyrolysis and combustion, contains 15 neuron nodes, the i4th neuron node taking value of 1-15; the pyrolysis process group describes composition change characteristics of pyrolysis products, contains 20 neuron nodes, the i5th neuron node taking value of 16-35; the combustion process group describes process characteristics of final oxidation stage, contains 15 neuron nodes, the i6th neuron node taking value of 36-50;

[0093] The specific implementation method of step S3 includes the following steps:

[0094] S3.1. Constructing drying process group to represent process characteristics of water removal in solid waste, reflect driving action of temperature on water evaporation and describe influence of drying process on subsequent pyrolysis and combustion; 15 neuron nodes in the drying process group are calculated as follows:

[0095]

[0096] wherein, is the output value of the i4th neuron node of the second hidden layer, i4 takes values from 1 to 15; is the weight coefficient of the i4th neuron node of the second hidden layer corresponding to the kth neuron of the first hidden layer, =1~80; is the bias term of the i4th neuron node of the second hidden layer corresponding to the kth neuron of the first hidden layer; is the output value of the kth neuron of the first hidden layer; is the drying process term, is the drying process coefficient, determined by expert experience, design documents, experimental research, etc;

[0097] S3.2. The pyrolysis process group is constructed to reflect the influence of temperature on the volatilization, describe the composition change characteristics of the pyrolysis products, and realize the extraction of the key characteristics of the pyrolysis stage. The calculation method of the 20 neuron nodes in the pyrolysis process group is as follows:

[0098]

[0099] wherein, is the output value of the i5th neuron node of the second hidden layer, i5 takes values from 16 to 35; is the weight coefficient of the i5th neuron node of the second hidden layer corresponding to the kth neuron of the first hidden layer; is the bias term of the i5th neuron node of the second hidden layer corresponding to the kth neuron of the first hidden layer; is the pyrolysis process term, is the pyrolysis process coefficient, determined by expert experience, design documents, experimental research, etc;

[0100] S3.3. The combustion process group is constructed to reflect the influence of oxygen concentration on the combustion effect, describe the process characteristics of the final oxidation stage; the calculation method of the 15 neuron nodes in the combustion process group is as follows:

[0101]

[0102] wherein, is the output value of the i6th neuron node of the second hidden layer, i6 takes values from 36 to 50; is the weight coefficient of the i6th neuron node of the second hidden layer corresponding to the kth neuron of the first hidden layer; is the bias term of the i6th neuron node of the second hidden layer corresponding to the kth neuron of the first hidden layer; is the combustion process term, For the combustion process coefficient, For the reference oxygen concentration, determined by expert experience, design documents, experimental research methods.

[0103] S4. The output layer of the deep learning-based solid waste thermal treatment prediction model is constructed, and the output layer is divided into a conversion efficiency group, a product distribution group and an energy conversion group. The output parameters of the output layer of the deep learning-based solid waste thermal treatment prediction model are determined.

[0104] Further, step S4 divides the output layer into a conversion efficiency group, a product distribution group and an energy conversion group based on the evaluation requirements of the solid waste thermal treatment process. The conversion efficiency group index reflects the integrity and sufficiency of material conversion, the product distribution group index reveals the migration rule of material and environmental impact, and the energy conversion group index embodies the energy utilization level.

[0105] The conversion efficiency group index includes Y1-Y6, Y1 is the moisture conversion rate, Y2 is the volatile conversion rate, Y3 is the fixed carbon conversion rate, Y4 is the carbon conversion rate, Y5 is the sulfur conversion rate, and Y6 is the chlorine conversion rate.

[0106] The product distribution group index includes the material distribution proportion index Y 71 ~Y 75 and the residue distribution index Y8-Y9, Y 71 is the mass distribution proportion of moisture after drying into flue gas, Y 72 is the mass distribution proportion of volatile after pyrolysis into flue gas, Y 73 is the distribution proportion of carbon converted into carbon dioxide into flue gas, Y 74 is the distribution proportion of sulfur converted into sulfur dioxide into flue gas, Y 75 is the distribution proportion of chlorine converted into hydrogen chloride into flue gas; Y8 is the mass distribution proportion of fixed carbon remaining in the slag after combustion, and Y9 is the mass distribution proportion of fly ash component into fly ash after combustion of fixed carbon.

[0107] The energy conversion group index includes Y 10 ~Y 12 , Y 10 is the thermal efficiency, Y 11 is the combustion efficiency, and Y 12 is the heat loss rate.

[0108] The specific implementation method of step S4 includes the following steps:

[0109] S4.1. The calculation method of the output value of different indexes in the conversion efficiency group is as follows:

[0110]

[0111] wherein, s1 is the output value of different indicators in the conversion efficiency group; is the weight coefficient of the output value of different indicators in the conversion efficiency group corresponding to the vth neuron of the second hidden layer; is the bias term of the output value of different indicators in the conversion efficiency group corresponding to the vth neuron of the second hidden layer; is the output value of the vth neuron of the second hidden layer;

[0112] s1=1, v∈{1,2,…,15}; s1=2, v∈{16,17,…,35}; s1=3~6, v∈{36,37,…,50};

[0113] S4.2. The calculation method of the output value of different indicators in the product distribution group is as follows:

[0114]

[0115] wherein, s2 is the output value of different indicators in the product distribution group; is the weight coefficient of the output value of different indicators in the product distribution group corresponding to the vth neuron of the second hidden layer; is the bias term of the output value of different indicators in the product distribution group corresponding to the vth neuron of the second hidden layer;

[0116] s2=71, v∈{1,2,…,15}; s2=72, v∈{16,17,…,35}; s2=73,74,75,8,9, v∈{36,37,…,50};

[0117] S4.3. The calculation method of the output value of different indicators in the energy conversion group is as follows:

[0118]

[0119] wherein, s3 is the output value of different indicators in the energy conversion group; is the weight coefficient of the output value of different indicators in the energy conversion group corresponding to the vth neuron of the second hidden layer; is the bias term of the output value of different indicators in the energy conversion group corresponding to the vth neuron of the second hidden layer;

[0120] s3=10,12, v∈{1,2,…,50}; s3=11, v∈{36,37,…,50}.

[0121] Further, the above data is derived from historical data of a solid waste treatment plant during model training.

[0122] S5. Design the loss function of the solid waste heat treatment prediction model based on deep learning;

[0123] Further, the loss function in step S5 considers the prediction loss and the heat treatment loss, and the specific implementation method includes the following steps:

[0124] S5.1. Construct the prediction loss based on the conversion efficiency prediction loss, the product distribution prediction loss and the energy conversion prediction loss, and the expression is:

[0125]

[0126] wherein, is the conversion efficiency prediction loss; is the product distribution prediction loss; is the energy conversion prediction loss;

[0127]

[0128] wherein, is the mean square error calculation function; is the output value of different indicators in the conversion efficiency group predicted by the model; is the output value of different indicators in the conversion efficiency group actually measured;

[0129]

[0130] wherein, is the output value of different indicators in the product distribution group predicted by the model; is the output value of different indicators in the product distribution group actually measured;

[0131]

[0132] wherein, is the output value of different indicators in the energy conversion group predicted by the model; is the output value of different indicators in the energy conversion group actually measured;

[0133] S5.2. Construct the heat treatment loss based on the material conversion constraint, the component conversion constraint and the heat conversion constraint , and the expression is:

[0134]

[0135] wherein, is the material conversion constraint, is the weight corresponding to the material conversion constraint; is the component conversion constraint, is the weight corresponding to the component conversion constraint; a heat conversion constraint, a weight corresponding to the heat conversion constraint;

[0136] The material conversion constraint is established as:

[0137]

[0138] wherein, a volatile matter conversion term; a fixed carbon allocation term;

[0139] The component conversion constraint is established as:

[0140]

[0141] wherein, a sulfur element conversion constraint, a chlorine element conversion constraint;

[0142] The heat conversion constraint is established as:

[0143] ;

[0144] S5.3. Constructing a total loss based on a predicted loss and a heat treatment loss ;

[0145]

[0146] wherein, a weight corresponding to the predicted loss, a weight corresponding to the heat treatment loss, determined by expert experience.

[0147] S6. Connecting an input layer, a main body layer, and an output layer of the deep learning-based solid waste heat treatment prediction model in sequence to obtain the deep learning-based solid waste heat treatment prediction model;

[0148] S7. Collecting data through a solid waste treatment plant and an automatic control system, normalizing the data, dividing the data into a training set, a validation set, and a test set according to a ratio of 7:2:1, training, validating, and testing the deep learning-based solid waste heat treatment prediction model obtained in step S6 to obtain a trained deep learning-based solid waste heat treatment prediction model;

[0149] Further, the model parameter settings include: network structure parameters, the number of neurons in each layer, which are hyperparameters that need to be set by humans based on expert experience; learning rate, batch size, training rounds, and weight coefficients in the loss function, which are hyperparameters that need to be set by humans during the model training process; and weight coefficients and bias terms in the model, which are parameters determined by the model during the training process.

[0150] Further, the model training in step S7 adopts a back propagation algorithm; during the training process, the model calculates a predicted value through forward propagation, substitutes the predicted value and a true value into a loss function to calculate a loss, and then updates network parameters through back propagation; to prevent overfitting, an early stopping strategy is adopted, and when the loss of a verification set does not decrease for 10 consecutive epochs, the training is stopped; the output result is recovered into an actual physical quantity through inverse normalization.

[0151] S8. After the prediction result of the trained deep learning-based solid waste heat treatment prediction model is inverse normalized, the prediction result of the deep learning-based solid waste heat treatment prediction model with an actual physical quantity is obtained, and a cost prediction method based on the solid waste heat treatment neural network model is constructed.

[0152] Further, step S8 starts from the overall operation characteristics of the solid waste heat treatment system, considers the close correlation among material conversion, energy utilization and environmental impact during system operation, and comprehensively evaluates the influence of material conversion efficiency on power generation cost, the consumption of reagents caused by different material conversion paths, the step utilization benefits of primary and secondary energy recovery, the equipment life cycle cost and the environmental governance cost.

[0153] Further, the specific implementation method of step S8 includes the following steps:

[0154] S8.1. The cost under the influence of conversion efficiency is predicted by establishing a multi-level efficiency correction system, and the expression of the actual power generation cost FD is obtained as follows:

[0155]

[0156] wherein, , , are efficiency correction coefficients related to element conversion, turbine efficiency and combustion temperature deviation, respectively; is the price per kilowatt-hour; is the total amount of electricity required;

[0157]

[0158] wherein, is a carbon conversion influence coefficient, is a sulfur conversion influence coefficient, is a chlorine conversion influence coefficient, which is determined by expert experience, design documents and experimental research methods; , , are standard conversion rates of carbon conversion rate, sulfur conversion rate and Y chlorine conversion rate, respectively, which are determined by expert experience, design documents and experimental research methods.

[0159]

[0160] wherein, is the temperature influence coefficient, is the pressure influence coefficient, determined by expert experience, design documents, experimental research methods; is the design steam temperature, is the design steam pressure, which can be determined by expert experience, design documents, experimental research, etc;

[0161]

[0162] wherein, is the temperature deviation influence coefficient, determined by expert experience, design documents, experimental research, etc; is the optimal combustion temperature related to the process, determined by expert experience, design documents, experimental research, etc;

[0163] The calculation of power generation cost needs to consider the comprehensive influence of material conversion efficiency, steam parameters and combustion temperature. The conversion efficiency of elements such as carbon, sulfur and chlorine directly affects the combustion quality, and then affects the power generation efficiency; the fluctuation of steam temperature and pressure will cause the change of turbine efficiency; the deviation of combustion temperature from the optimal value will reduce the overall thermal efficiency. By establishing a multi-level efficiency correction system, the actual power generation cost can be accurately evaluated;

[0164] S8.2. In the solid waste incineration process, the conversion path of different substances will affect the consumption of reagents, so a reagent consumption model considering multiple factors is established to accurately evaluate the actual operation cost. The calculation of deacidifying agent dosage TS is as follows:

[0165]

[0166] wherein, is the reference deacidifying agent dosage; is the model prediction value, =73、74 and 75; is the corresponding standard value; is the corresponding standard value; is the stoichiometric coefficient, CO2 takes 0.1, SO2 takes 1.0, and HCl takes 1.0; is the reaction activity coefficient, determined by expert experience, design documents, experimental research, etc; is the adjustment coefficient related to the combustion efficiency, is the adjustment coefficient related to the furnace negative pressure, determined by expert experience, design documents, experimental research methods;

[0167] Then the cost PTS of the material conversion path is calculated as follows:

[0168]

[0169] wherein, is the price of deacidification agent, obtained from the market;

[0170] S8.3. Energy utilization in solid waste incineration process includes primary energy recovery and secondary energy recovery. Primary energy is mainly achieved through steam power generation, and its efficiency is affected by thermal efficiency and heat loss. Secondary energy includes flue gas waste heat and cooling system waste heat. By establishing a complete energy recovery system, the overall energy utilization efficiency of the system can be improved.

[0171] The expression of the primary energy recovery amount NY is:

[0172]

[0173] wherein, is the primary energy recovery coefficient, determined by expert experience, design documents, and experimental research methods;

[0174] The expression of the secondary energy recovery amount NE is:

[0175]

[0176] wherein, m3=71, 72, 73, 74, and 75; D m3 is the specific heat capacity of each component, determined from the chemical handbook or experimental research; is the initial temperature, directly obtained from the automatic control system; is the secondary energy recovery coefficient, determined by expert experience, design documents, and experimental research methods;

[0177] The total energy cascade utilization benefit ZN is:

[0178]

[0179] wherein, is the unit price of energy, determined by querying the local market price;

[0180] S8.4. Equipment life is affected by material conversion characteristics, corrosion environment, and thermal stress. Different material conversions will cause different degrees of equipment wear, and acid gas will accelerate equipment corrosion. By establishing a life prediction model considering multiple influences, the full life cycle cost of equipment can be accurately evaluated;

[0181] The equipment predicted life SM is calculated as follows:

[0182]

[0183] wherein, The design life of the equipment is determined by the design specification; The coefficient of wear contribution, The value is 1~6; The basic corrosion coefficient is determined by expert experience; The corrosion contribution coefficient is determined by expert experience, The value is 74 and 75; The standard value corresponding to The standard value corresponding to The standard value corresponding to The standard value corresponding to is determined by expert experience, design documents, experimental research, etc;

[0184] The cost NH of the equipment is calculated as follows:

[0185]

[0186] Wherein, The initial investment amount; The annual maintenance cost is determined according to market conditions; The discount rate is determined according to market conditions; The discounting period;

[0187] S8.5. The environmental cost includes direct treatment cost and indirect environmental impact cost. The direct treatment cost is affected by the amount of pollutants generated and the treatment process, and the treatment efficiency change brought by the working condition fluctuation is considered. The indirect environmental cost is related to the regional environmental carrying capacity and sensitivity. At the same time, the carbon emission cost is considered as a special environmental cost, which considers the influence of carbon trading market and emission reduction policy;

[0188] The direct treatment cost CLz is calculated as follows:

[0189]

[0190] Wherein, The value is 71~75, The treatment unit price is determined by market research; The working condition correction coefficient is determined by expert experience, design documents, experimental research;

[0191] The indirect environmental cost CLj is calculated as follows:

[0192]

[0193] Wherein, The indirect treatment unit price is determined by market research; The regional sensitivity coefficient is related to the regional environmental quality, which is determined by expert experience, design documents, experimental research;

[0194] The carbon emission cost CBc is calculated as follows:

[0195]

[0196] wherein, is the molecular weight ratio of carbon dioxide to carbon; is the price of carbon, obtained from the market; is the carbon emission reduction contribution rate, determined by investigation;

[0197] The total environmental cost CBh is calculated as follows:

[0198]

[0199] S8.6. Calculate the total operating cost .

[0200] The embodiment proposes a solid waste thermal treatment process characteristic analysis deep learning model construction method. The method is designed and optimized for the characteristics of solid waste thermal treatment in the input parameters, the first hidden layer, the second hidden layer, and the output layer, and realizes the scientificity and rationality of the prediction of the conversion efficiency, product distribution, and energy conversion in the solid waste thermal treatment process.

[0201] The embodiment proposes a solid waste thermal treatment cost prediction method. This solid waste thermal treatment cost prediction method establishes a comprehensive cost evaluation system, including five sub-models of power generation cost under the influence of conversion efficiency, reagent cost of material conversion path, energy cascade utilization benefit, equipment full life cycle cost, and environmental cost. The method provides a reliable calculation method for solid waste thermal treatment cost prediction, which can help enterprises to evaluate the treatment cost under different solid waste characteristics and operating conditions in advance, and provide scientific basis for operation optimization and decision-making.

[0202] It should be noted that the relational terms, such as "first" and "second", and the like, are used solely to distinguish one from another entity or action, without necessarily requiring or implying any such actual relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without more limitations, an element defined by the statement "comprising a" does not exclude the existence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0203] ​Although the present application has been described with reference to the specific embodiments thereof, it should be understood by those skilled in the art that various changes can be made and equivalents can be substituted for elements thereof without departing from the scope of the present application. In particular, various features and aspects of the present application can be used individually or in any combination depending on the specific application and implementation. Therefore, it is expressly intended that the specific embodiments of the present application both as set forth and including any equivalents thereof should not limit the present application or scope of the claims herein, but rather the overall scope of pertaining solely to the methods and the articles of manufacture specifically recited in the following claims.

Claims

1. A method for cost prediction based on a solid waste thermal treatment neural network model, characterized by, Comprising the following steps: S1. Determine the input parameters of the input layer of the deep learning-based solid waste thermal treatment prediction model, including material property parameters, process operation parameters and process parameters; S2. Construct the first hidden layer in the main layer of the deep learning-based solid waste thermal treatment prediction model; Step S2 divides the nodes of the first hidden layer into a material property group, a reaction property group and an energy property group, wherein the material property group focuses on the physical and chemical property changes of the material in the thermal treatment process, contains 30 neuron nodes, and the i1th neuron node takes a value of 1-30; the reaction property group focuses on describing the degree of progress of various chemical reactions, contains 25 neuron nodes, and the i2th neuron node takes a value of 31-55; the energy property group focuses on representing the energy conversion and transmission characteristics, contains 25 neuron nodes, and the i3th neuron node takes a value of 56-80; The specific implementation method of step S2 comprises the following steps: S2.

1. Construct the material property group to introduce temperature correction and material moisture correction, the temperature correction considers the driving effect of temperature on the conversion rate of the material, and reflects the difference in the conversion law of the material in different temperature intervals; the material moisture correction considers the influence of the evaporation process of the material on the subsequent conversion; the calculation method of the 30 neuron nodes in the material property group is as follows: ； wherein, represents the output value of the i1th neuron node in the first hidden layer, i1 takes values from 1 to 30; is an activation function, used to introduce nonlinear characteristics; is a weight coefficient of the i1th neuron node in the first hidden layer corresponding to the jth input parameter; is an input parameter, ; is a bias term of the i1th neuron node in the first hidden layer corresponding to the jth input parameter; is a temperature correction coefficient, determined by expert experience, design documents, and experimental research methods; is a moisture correction coefficient, determined by expert experience, design documents, and experimental research methods; is a temperature correction term; is a material moisture correction term; S2.

2. Construct the reaction property group to introduce a mass conversion term, which directly reflects the reaction degree of the material and embodies the depth of the reaction; add an air excess term to represent the sufficiency of the combustion environment and reflect the influence of the oxidation reaction condition on the conversion effect; the calculation method of the 25 neuron nodes in the reaction property group is as follows: ； wherein, is an output value of the i2th neuron node in the first hidden layer, i2 is 31~55; is a weight coefficient of the i2th neuron node in the first hidden layer corresponding to the jth input parameter; is a bias term of the i2th neuron node in the first hidden layer corresponding to the jth input parameter; is a mass conversion coefficient, determined by expert experience, design documents, and experimental research methods; is an air excess coefficient, determined by expert experience, design documents, and experimental research methods; is an air density, determined by actual measurement; is a mass conversion term; is an air excess term; S2.

3. Construct the energy property group to introduce a heat value conversion term to directly represent the energy conversion effect; add a temperature gradient term to reflect the driving force and transmission strength of heat transfer and describe the spatial distribution law of energy; the calculation method of the 25 neuron nodes in the energy property group is as follows: ； wherein, is an output value of the i3th neuron node in the first hidden layer, i3 is 56-80; is a weight coefficient of the jth input parameter and the i3th neuron node in the corresponding first hidden layer; is a bias term of the i3th neuron node in the first hidden layer corresponding to the jth input parameter; is a heat value conversion term; is a heat value conversion coefficient, determined by expert experience, design documents, and experimental research methods; is a temperature gradient term; is a temperature gradient coefficient, determined by expert experience, design documents, and experimental research methods; S3. Construct the second hidden layer in the main layer of the deep learning-based solid waste thermal treatment prediction model, and the first hidden layer is connected to the second hidden layer; Step S3 divides the nodes of the second hidden layer into a drying process group, a pyrolysis process group and a combustion process group, wherein the drying process group describes the influence of the drying process on the subsequent pyrolysis and combustion, contains 15 neuron nodes, and the i4th neuron node takes a value of 1-15; the pyrolysis process group describes the composition change characteristics of the pyrolysis product, contains 20 neuron nodes, and the i5th neuron node takes a value of 16-35; the combustion process group describes the process characteristics of the final oxidation stage, contains 15 neuron nodes, and the i6th neuron node takes a value of 36-50; The specific implementation method of step S3 comprises the following steps: S3.

1. Construct the drying process group to represent the process characteristics of the water removal in the solid waste, reflect the driving effect of temperature on the evaporation of water, and describe the influence of the drying process on the subsequent pyrolysis and combustion; the calculation method of the 15 neuron nodes in the drying process group is as follows: ； wherein, is the output value of the i4th neuron node of the second hidden layer, i4 is 1~15; is the weight coefficient of the i4th neuron node in the second hidden layer corresponding to the kth neuron of the first hidden layer, =1~80; is the bias term of the i4th neuron node in the second hidden layer corresponding to the kth neuron of the first hidden layer; is the output value of the kth neuron of the first hidden layer; is the drying process term, is the drying process coefficient, which is determined by expert experience, design documents, and experimental research methods; S3.

2. Constructing the pyrolysis process group to reflect the influence of temperature on the volatilization of the pyrolysis products, describe the composition change characteristics of the pyrolysis products, and extract the key characteristics of the pyrolysis stage; the calculation method of the 20 neuron nodes in the pyrolysis process group is as follows: ； wherein, is the output value of the i5th neuron node of the second hidden layer, i5 is 16-35; is the weight coefficient of the i5th neuron node in the second hidden layer corresponding to the kth neuron of the first hidden layer; is the bias term of the i5th neuron node in the second hidden layer corresponding to the kth neuron of the first hidden layer; is a pyrolysis process term, is a pyrolysis process coefficient, determined by expert experience, design documents, and experimental research methods; S3.

3. Constructing the combustion process group to reflect the influence of oxygen concentration on the combustion effect, describe the process characteristics of the final oxidation stage; the calculation method of the 15 neuron nodes in the combustion process group is as follows: ； wherein, is the output value of the i6th neuron node of the second hidden layer, i6 is 36-50; is the weight coefficient of the i6th neuron node in the second hidden layer corresponding to the kth neuron of the first hidden layer; is the bias term of the i6th neuron node in the second hidden layer corresponding to the kth neuron of the first hidden layer; is the combustion process term, is the combustion process coefficient, is the reference oxygen concentration, determined by expert experience, design documents, and experimental research methods; S4. Constructing the output layer of the deep learning-based solid waste thermal treatment prediction model, dividing the output layer into a conversion efficiency group, a product distribution group, and an energy conversion group, and determining the output parameters of the output layer of the deep learning-based solid waste thermal treatment prediction model; S5. Designing the loss function of the deep learning-based solid waste thermal treatment prediction model; S6. Connecting the input layer, the main layer, and the output layer of the deep learning-based solid waste thermal treatment prediction model in sequence to obtain the deep learning-based solid waste thermal treatment prediction model; S7. Collecting data through a solid waste treatment plant and an automatic control system, normalizing the data, dividing the data into a training set, a validation set, and a test set according to a ratio of 7:2:1, training, validating, and testing the deep learning-based solid waste thermal treatment prediction model obtained in step S6 to obtain a trained deep learning-based solid waste thermal treatment prediction model; S8. Based on the prediction results of the trained deep learning-based solid waste thermal treatment prediction model, obtaining the prediction results of the deep learning-based solid waste thermal treatment prediction model with actual physical quantities after reverse normalization, and constructing a cost prediction method based on the solid waste thermal treatment neural network model. 2.The method of claim 1, wherein, The number of nodes of the input layer is determined as 35 in step S1, and are respectively denoted as X1-X35. 35 ; X1~X 14 For material characteristic parameters, the corresponding specific parameters are as follows: X1 is moisture content, X2 is ash content, X3 is combustible content, X4 is carbon content, X5 is hydrogen content, X6 is oxygen content, X7 is nitrogen content, X8 is sulfur content, X9 is chlorine content, X 10 For heat value, X 11 For bulk density, X 12 For particle size, X 13 For sand content, X 14 For metal content; X 15 ~X 25 For process operating parameters, the corresponding parameters are as follows:X 15 For grate speed, X 16 For feed rate, X 17 For primary air volume, X 18 For secondary air volume, X 19 For furnace temperature, X 20 For furnace negative pressure, X 21 For flue temperature, X 22 For steam temperature, X 23 For steam pressure, X 24 For steam flow, X 25 For ambient temperature; X 26 ~X 35 For process parameters, the specific parameters are as follows: X 26 For feed moisture content, X 27 For discharge moisture content, X 28 For feed volatile matter, X 29 For discharge volatile matter, X 30 For feed mass flow rate, X 31 For discharge mass flow rate, X 32 For feed heat value, X 33 For discharge heat value, X 34 For flue gas oxygen concentration, X 35 For ash content.

3. The method of claim 2, wherein the method is characterized by, Step S4 divides the output layer into a conversion efficiency group, a product distribution group, and an energy conversion group based on the evaluation requirements of the solid waste thermal treatment process; the conversion efficiency group indicators reflect the completeness and sufficiency of material conversion, the product distribution group indicators reveal the migration law of materials and environmental impact, and the energy conversion group indicators represent the energy utilization level; The conversion efficiency group indicators include Y1-Y6, Y1 is the moisture conversion rate, Y2 is the volatile conversion rate, Y3 is the fixed carbon conversion rate, Y4 is the carbon conversion rate, Y5 is the sulfur conversion rate, and Y6 is the chlorine conversion rate; The product distribution group index includes a substance distribution proportion index Y 71 Y 75 and a residue distribution index Y8-Y9, Y 71 is the mass distribution proportion of water after drying into flue gas, Y 72 is the mass distribution proportion of volatile matter after pyrolysis into flue gas, Y 73 is the distribution proportion of carbon converted into carbon dioxide into flue gas, Y 74 is the distribution proportion of sulfur converted into sulfur dioxide into flue gas, Y 75 is the distribution proportion of chlorine converted into hydrogen chloride into flue gas; Y8 is the mass distribution proportion of fixed carbon remaining in the slag after combustion, and Y9 is the mass distribution proportion of the fly ash component distribution into fly ash after combustion of fixed carbon; Energy conversion group indicators include Y 10 ~Y 12 , Y 10 is thermal efficiency, Y 11 is combustion efficiency, Y 12 is thermal loss rate; The specific implementation method of step S4 includes the following steps: S4.

1. The calculation method of the output values of different indicators in the conversion efficiency group is as follows: ； wherein, is the output value of the different indicators in the conversion efficiency group, and s1 is the different indicators in the conversion efficiency group corresponding to the second hidden layer vth neuron; is the weight coefficient of the output value of the different indicators in the conversion efficiency group corresponding to the second hidden layer vth neuron; is the bias term of the output value of the different indicators in the conversion efficiency group corresponding to the second hidden layer vth neuron; is the output value of the vth neuron of the second hidden layer. s1=1, v∈{1,2,…,15}; s1=2, v∈{16,17,…,35}; s1=3~6, v∈{36,37,…,50}; S4.

2. The calculation method of the output values of different indicators in the product distribution group is as follows: ； wherein, is the output value of the different index in the product distribution group, s2 is the output value of the different index in the product distribution group corresponding to the vth neuron of the second hidden layer; is the weight coefficient of the output value of the different index in the product distribution group corresponding to the vth neuron of the second hidden layer; is the bias term of the output value of the different index in the product distribution group corresponding to the vth neuron of the second hidden layer. s2=71, v∈{1,2,…,15}; s2=72, v∈{16,17,…,35}; s2=73,74,75,8,9, v∈{36,37,…,50}; S4.

3. The calculation method of the output values of different indicators in the energy conversion group is as follows: ； wherein, s3 is the output value of the different indicators in the energy conversion group; is the weight coefficient of the output value of the different indicators in the energy conversion group corresponding to the vth neuron of the second hidden layer; is the bias term of the output value of the different indicators in the energy conversion group corresponding to the vth neuron of the second hidden layer. s3=10,12 when v∈{1,2, …, 50}; s3=11 when v∈{36,37, …, 50}.

4. The method of claim 3, wherein the method further comprises: The loss function in step S5 considers the prediction loss and the heat treatment loss, and the specific implementation method includes the following steps: S5.

1. The prediction loss is constructed based on the conversion efficiency prediction loss, the product distribution prediction loss, and the energy conversion prediction loss, and the expression is: ； wherein, conversion efficiency prediction loss; product distribution prediction loss; energy conversion prediction loss; ； wherein, is a mean squared error calculation function; is a model predicted conversion efficiency set of different index output values; is an actually measured conversion efficiency set of different index output values; ； wherein, are the different indicator output values in the product distribution set predicted by the model; are the different indicator output values in the product distribution set actually measured; ； wherein, are the model predicted different indicator output values in the energy conversion group; are the actually measured different indicator output values in the energy conversion group; S5.

2. Constructing heat treatment losses based on material conversion constraints, composition conversion constraints, and heat conversion constraints , the expression is: ； wherein, is a material conversion constraint, is a weight corresponding to the material conversion constraint; is a component conversion constraint, is a weight corresponding to the component conversion constraint; is a heat conversion constraint, is a weight corresponding to the heat conversion constraint; The material conversion constraint is established as: ； wherein, is a volatile conversion term; is a fixed carbon allocation term; The component conversion constraint is established as: ； wherein, is a sulfur element conversion constraint, is a chlorine element conversion constraint; The heat conversion constraint is established as: ； S5.

3. Constructing the total loss based on the predicted loss and the heat treatment loss ; ； wherein, is the weight corresponding to the prediction loss, is the weight corresponding to the heat treatment loss, determined by expert experience.

5. The method of claim 4, wherein the method further comprises: The model training in step S7 adopts the back propagation algorithm; during the training process, the model calculates the predicted value through forward propagation, substitutes the predicted value and the true value into the loss function to calculate the loss, and then updates the network parameters through back propagation; to prevent overfitting, the early stopping strategy is adopted, and the training is stopped when the validation set loss does not decrease for 10 consecutive epochs; the output result is restored to the actual physical quantity through inverse normalization.

6. The method of claim 5, wherein the method further comprises: Step S8 considers the close relationship between material conversion, energy utilization and environmental impact during the operation of the solid waste heat treatment system, and comprehensively evaluates the influence of material conversion efficiency on power generation cost, the consumption of different substances caused by the conversion path, the cascade utilization benefit of primary and secondary energy recovery, the life cycle cost of equipment, and the environmental governance cost.

7. The method of claim 6, wherein the method further comprises: The specific implementation method of step S8 includes the following steps: S8.

1. The actual power generation cost FD is calculated by establishing a multi-level efficiency correction system to predict the cost under the influence of conversion efficiency, and the expression is: ； wherein, , , are efficiency correction factors related to element conversion, turbine efficiency, combustion temperature deviation, respectively; is the price per kilowatt-hour; is the total amount of electricity required; ； Wherein, is the carbon conversion influence coefficient, is the sulfur conversion influence coefficient, is the chlorine conversion influence coefficient, determined by expert experience, design documents, and experimental research methods; , , are the standard conversion rates of carbon conversion rate, sulfur conversion rate, and Y chlorine conversion rate, respectively, determined by expert experience, design documents, and experimental research methods. ； Wherein, is the temperature influence coefficient, is the pressure influence coefficient, determined by expert experience, design documents, and experimental research methods; is the design steam temperature, is the design steam pressure, determined by expert experience, design documents, and experimental research methods; ； Wherein, is the temperature deviation influence coefficient, determined by expert experience, design documents, and experimental research methods; is the optimal combustion temperature related to the process, determined by expert experience, design documents, and experimental research methods; S8.

2. In the solid waste incineration process, the conversion path of different substances will affect the consumption of reagents, so an accurate evaluation of the actual operation cost is made by establishing a reagent consumption model considering multiple factors, and the calculation of the amount of deacidification agent TS is as follows: ； Wherein, is the reference deacidification agent dosage; is the model prediction value, =73, 74 and 75; is the standard value corresponding to ; is the stoichiometric coefficient, CO2 is 0.1, SO2 is 1.0, HCl is 1.0; is the reaction activity coefficient, determined by expert experience, design documents, and experimental research methods; is the adjustment coefficient related to the combustion efficiency, is the adjustment coefficient related to the furnace negative pressure, determined by expert experience, design documents, and experimental research methods; Then the cost PTS of the material conversion path is calculated as follows: ； wherein, The price of the deacidifier is obtained from the market. S8.

3. The energy utilization in the solid waste incineration process includes primary energy recovery and secondary energy recovery, the primary energy is mainly recovered through steam power generation, and its efficiency is affected by thermal efficiency and heat loss; the secondary energy includes flue gas waste heat and cooling system waste heat, and a complete energy recovery system is established to improve the overall energy utilization efficiency of the system; The expression of the primary energy recovery amount NY is as follows: ； wherein, is the coefficient of primary energy recovery, determined by expert experience, design documentation, and experimental research methods; The expression of the secondary energy recovery amount NE is as follows: ； wherein m3 = 71, 72, 73, 74 and 75; D m3 Cp is the specific heat capacity of each component, determined from the Handbook of Chemistry or experimental research; T0 is the initial temperature, directly obtained from the automation control system; C2 is the secondary energy recovery coefficient, determined by expert experience, design documents, and experimental research methods; Then the total energy cascade utilization benefit ZN is as follows: ； wherein, The unit price of energy is determined by inquiring the local market price. S8.

4. The equipment life is affected by the comprehensive influence of material conversion characteristics, corrosion environment and thermal stress; the conversion of different substances will cause different degrees of equipment wear, and acid gas will accelerate equipment corrosion; an accurate evaluation of the life cycle cost of equipment is made by establishing a life prediction model considering multiple influences; The predicted life SM of the equipment is calculated as follows: ； Wherein, is the design life of the equipment, determined by the design specification; is the wear contribution coefficient, is 1-6; is the base corrosion coefficient, determined by expert experience; is the corrosion contribution coefficient, determined by expert experience, is 74 and 75; is the corresponding standard value, is the corresponding standard value, is the corresponding standard value, is the corresponding standard value, determined by expert experience, design documents, and experimental research methods; The cost NH of the equipment is calculated as follows: ； Wherein, is the initial investment amount; is the annual maintenance cost, determined according to market conditions; is the discount rate, determined according to market conditions; is the discounting period; S8.

5. The environmental cost includes direct treatment cost and indirect environmental impact cost, the direct treatment cost is affected by the amount of pollutants generated and the treatment process, and the treatment efficiency change caused by working condition fluctuation is considered; the indirect environmental cost is related to the regional environmental carrying capacity and sensitivity; at the same time, the carbon emission cost as a special environmental cost considers the influence of carbon trading market and emission reduction policy; The direct treatment cost CLz is calculated as follows: ； Wherein, The value is 71-75, The processing unit price is determined by market research; The working condition correction coefficient is determined by expert experience, design documents, and experimental research methods. The indirect environmental cost CLj is calculated as follows: ； Wherein, The indirect processing unit price is determined by market research; The regional sensitivity coefficient is related to the regional environmental quality and is determined by expert experience, design documents, and experimental research methods. The carbon emission cost CBc is calculated as follows: ； Where, The price of carbon is obtained from the market; The carbon emission reduction contribution rate is determined by investigation The total environmental cost CBh is calculated as follows: ； S8.

6. Calculate total operating cost .

Images