Intelligent Optimization Operation Methods for Urban Solid Waste Incineration Processes

Abstract

The application provides an intelligent optimization operation method for urban solid waste incineration process, and relates to the technical field of urban solid waste treatment and intelligent optimization. The method comprises the following steps: obtaining environmental data in the incinerator and concentration values of various incineration exhaust gases before a preset time length; iteratively optimizing the solution of the decision variable, so that the value of the target function obtained according to the solution of the decision variable is optimized in the direction of reduction; determining the solution of the target decision variable to obtain the set value of the oxygen content in flue gas and the hearth temperature in the current urban solid waste incineration process; the calculation step of the value of the target function comprises the following steps: inputting the solution of the decision variable, the environmental data in the current incinerator and the concentration values of various incineration exhaust gases before a preset time length into a pre-trained operation index model, outputting the predicted concentration values of various incineration exhaust gases, calculating the combustion efficiency and the concentration of flue gas pollutants, and obtaining the value of the target function. The combustion efficiency is improved while the emission of flue gas pollutants is reduced.

Description

Technical Field

[0001] This invention relates to the fields of urban solid waste treatment technology and intelligent optimization technology, and in particular to an intelligent optimization operation method for urban solid waste incineration process. Background Technology

[0002] Driven by rapid urbanization, urban solid waste is experiencing rapid growth, and incineration has become the mainstream method for solid waste treatment. During incineration, combustion efficiency reflects the adequacy of solid waste combustion, while pollutants emitted during the process cause significant harm to the environment and the lives of people in surrounding areas. Therefore, optimizing the urban solid waste incineration process to improve combustion efficiency while reducing flue gas pollutant emissions has important theoretical and practical value. Summary of the Invention

[0003] This invention provides an intelligent optimization operation method for urban solid waste incineration processes, which addresses the shortcomings of existing technologies that make it difficult to reduce flue gas pollutant emissions while improving combustion efficiency, thereby achieving the goal of reducing flue gas pollutant emissions while improving combustion efficiency.

[0004] This invention provides a method for intelligent optimization of the operation of urban solid waste incineration processes, comprising the following steps.

[0005] During the incineration of urban solid waste, environmental data inside the incinerator and the concentration values ​​of various incineration emission gases before a preset time are obtained.

[0006] The solution to the decision variables is iteratively optimized so that the value of the objective function obtained from the solution to the decision variables is optimized in the direction of decreasing; the decision variables include flue gas oxygen content and furnace temperature;

[0007] The solution of the target decision variable is determined from the solutions of each group of decision variables obtained in the last iteration, and the values ​​of flue gas oxygen content and furnace temperature contained in the solution of the target decision variable are determined as the set values ​​of flue gas oxygen content and furnace temperature in the current urban solid waste incineration process.

[0008] The calculation steps of the objective function include: inputting the solution of the decision variable, the current environmental data in the incinerator, and the concentration values ​​of various incineration emission gases before the preset time period into a pre-trained operational index model of various incineration emission gases; outputting the predicted concentration values ​​of various incineration emission gases; calculating the combustion efficiency and flue gas pollutant concentration based on the predicted concentration values ​​of various incineration emission gases; and calculating the value of the objective function based on the combustion efficiency and the flue gas pollutant concentration.

[0009] According to the intelligent optimization operation method for urban solid waste incineration processes provided by the present invention, the training steps of the pre-trained operation index models for various incineration emission gases include:

[0010] Based on historical data of urban solid waste incineration processes, a sample dataset is constructed. The sample dataset includes sample flue gas oxygen content, sample furnace temperature, sample environmental data inside the incinerator, and sample concentration values ​​of various incineration emission gases during the historical process of urban solid waste incineration.

[0011] Iteratively input the oxygen content of the sample flue gas, the sample furnace temperature, the sample environmental data in the incinerator, and the sample concentration values ​​of various incineration emission gases before the preset time at the time into the operation index model of various incineration emission gases to be trained, and output the sample predicted concentration values ​​of various incineration emission gases at the time.

[0012] Based on the difference between the predicted concentration values ​​of various incineration emission gases and the actual concentration values ​​of the samples at the specified time in the sample dataset, the parameters of the operational index model for various incineration emission gases to be trained are adjusted and the next iteration is initiated until the iteration stops, thus obtaining the trained operational index model for various incineration emission gases.

[0013] According to the present invention, an intelligent optimization operation method for urban solid waste incineration processes is provided, wherein the oxygen content of sample flue gas, sample furnace temperature, sample environmental data within the incinerator, and sample concentration values ​​of various incineration emission gases prior to a preset time at different times are iteratively input into an operational index model for various incineration emission gases to be trained, and the predicted sample concentration values ​​of various incineration emission gases at the specified time are output, including:

[0014] Iteratively, the oxygen content of the sample flue gas at different times, the sample furnace temperature, the sample environmental data inside the incinerator, and the sample concentration values ​​of various incineration emission gases before the preset time at the time are used as input variables and sequentially input into the input layer of each sub-module in the operational index model of various incineration emission gases to be trained; wherein, each sub-module is used to output the sample predicted concentration value of one incineration emission gas.

[0015] In each submodule, the input variables are normalized through the input layer, and the normalized input variables are input to the hidden layer. The hidden layer performs a spatial mapping transformation on the normalized input variables, and the spatially mapped input variables are input to the output layer. The output layer performs a weighted summation on the spatially mapped input variables and outputs the sample predicted concentration value of the incineration emission gas corresponding to the submodule.

[0016] According to the present invention, an intelligent optimization operation method for urban solid waste incineration process is provided, wherein the incineration emission gases include nitrogen oxides, carbon monoxide, carbon dioxide, sulfur dioxide and hydrogen chloride;

[0017] The calculation of combustion efficiency and flue gas pollutant concentration based on the predicted concentration values ​​of the various incineration emission gases includes:

[0018] The combustion efficiency is calculated based on the predicted concentration values ​​of the carbon dioxide and the carbon monoxide.

[0019] The concentrations of pollutants in the flue gas are calculated based on the predicted concentration values ​​of the nitrogen oxides, carbon monoxide, sulfur dioxide, and hydrogen chloride.

[0020] According to the present invention, an intelligent optimization operation method for urban solid waste incineration processes, wherein iteratively optimizing the solution of decision variables to optimize the value of the objective function obtained from the solution of the decision variables in a decreasing direction includes:

[0021] Initialize the decision variable population as the original population of decision variables in the first iteration;

[0022] In each iteration, a mutation operation is performed on each individual in the original population of the decision variables in this iteration to obtain the mutated population;

[0023] A crossover operation is performed on the original population and the mutated population of the decision variables in this iteration to obtain the experimental population;

[0024] The experimental population and the original population of the decision variables are merged to obtain a mixed population;

[0025] Based on the values ​​of the objective functions obtained from the solutions of each decision variable in the mixed population, the original population of decision variables in the next iteration is determined, and the next iteration begins.

[0026] According to the present invention, an intelligent optimization operation method for urban solid waste incineration processes is provided, wherein performing a mutation operation on each individual in the original population of decision variables in the current iteration to obtain a mutated population includes:

[0027] If the current iteration number is less than or equal to the target iteration number, a single mutation strategy is used to perform mutation operations on each individual in the original population of the decision variables in this round of iteration to obtain a mutated population; the target iteration number is obtained according to a preset proportion of the maximum iteration number;

[0028] If the current iteration number is greater than the target iteration number, then multiple mutation strategies are allocated to each individual in the original population of decision variables in the current iteration according to their respective target proportions, and the allocated mutation strategies are used to perform mutation operations on the corresponding individuals to obtain multiple mutated populations;

[0029] The target ratio is determined based on the performance of each mutation strategy in the previous iteration.

[0030] According to the present invention, a method for intelligent optimization of urban solid waste incineration process is provided, wherein the step of determining the target ratio includes:

[0031] Determine the first fitness of the solutions to the decision variables in the mutant population obtained by each mutation strategy in the previous iteration, and the second fitness of the solutions to the decision variables in the original population in the previous iteration; the fitness is determined based on the value of the objective function.

[0032] The performance of each mutation strategy is determined based on the distance between the first fitness and the second fitness corresponding to each mutation strategy.

[0033] The target ratio corresponding to each mutation strategy is determined based on the ratio between the performance of each mutation strategy and the total performance of all mutation strategies.

[0034] According to the present invention, an intelligent optimization operation method for urban solid waste incineration processes is provided, wherein determining the original population of decision variables for the next iteration based on the values ​​of the objective functions obtained from the solutions of each decision variable in the mixed population includes:

[0035] The mixed population was divided into multiple clusters;

[0036] Based on the values ​​of the objective function corresponding to the solutions of the decision variables in each cluster, the non-dominated solutions and dominated solutions in each cluster are determined, and the non-dominated solutions are assigned to the elite files and the dominated solutions are assigned to the non-elite files.

[0037] The original population of decision variables for the next iteration is determined based on the elite profile.

[0038] According to the present invention, a method for intelligent optimization of urban solid waste incineration process is provided, wherein determining the original population of decision variables for the next iteration based on the elite profile includes:

[0039] If the number of solutions to decision variables in the elite archive does not reach the preset scale, the distribution density of the solutions to decision variables in the non-elite archive in the decision space is quantified. Based on the distribution density, the solutions of decision variables in sparse regions are selected to supplement the elite archive. From the supplemented elite archive, the solutions of multiple decision variables are selected to form the original population of decision variables in the next iteration.

[0040] If the number of solutions to decision variables in the elite archives reaches a preset size, then the solutions to multiple decision variables in the elite archives are selected directly from the elite archives to form the original population of decision variables in the next iteration, based on the values ​​of the objective function corresponding to the solutions to each decision variable in the elite archives.

[0041] According to the intelligent optimization operation method for urban solid waste incineration process provided by the present invention, the step of determining the solution of the target decision variable from the solutions of each set of decision variables obtained in the last iteration includes:

[0042] For each set of decision variables obtained in the last iteration, the first and second optimization target values ​​corresponding to the solutions of the decision variables are weighted and summed according to the first and second preset weights to obtain the weighted summation result corresponding to the solutions of the decision variables; the first optimization target value is determined based on the combustion efficiency corresponding to the solutions of the decision variables; the second optimization target value is determined based on the flue gas pollutant concentration corresponding to the solutions of the decision variables.

[0043] The solution of the decision variable that minimizes the weighted summation result is determined as the solution of the target decision variable.

[0044] The intelligent optimization operation method for urban solid waste incineration provided by this invention determines the predicted concentration values ​​of various incineration emission gases through a pre-trained operational index model of various incineration emission gases. Based on the predicted concentration values ​​of various incineration emission gases, the combustion efficiency and flue gas pollutant concentration are calculated. Based on the combustion efficiency and flue gas pollutant concentration, the value of the objective function is calculated, and the solution of the decision variables is iteratively optimized so that the value of the objective function obtained from the solution of the decision variables is optimized in the direction of reduction. This enables the accurate determination of the set values ​​of flue gas oxygen content and furnace temperature, thereby reducing the emission of flue gas pollutants while improving combustion efficiency. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0046] Figure 1 This is one of the flowcharts illustrating the intelligent optimization operation method for urban solid waste incineration provided by this invention.

[0047] Figure 2 This is a schematic diagram of the operational index model for various incineration emission gases provided by the present invention.

[0048] Figure 3 This is a schematic diagram of the test results of the nitrogen oxide operating index model provided by the present invention.

[0049] Figure 4 This is a schematic diagram of the test results of the carbon monoxide operation index model provided by the present invention.

[0050] Figure 5 This is a schematic diagram of the test results of the carbon dioxide operation index model provided by the present invention.

[0051] Figure 6 This is a schematic diagram of the test results of the sulfur dioxide operation index model provided by the present invention.

[0052] Figure 7 This is a schematic diagram of the test results of the hydrogen chloride operation index model provided by the present invention.

[0053] Figure 8 This is a flowchart illustrating the solution of the iterative optimization decision variables provided by the present invention.

[0054] Figure 9 This is a schematic diagram of the optimized combustion efficiency provided by the present invention.

[0055] Figure 10 This is a schematic diagram illustrating the optimized concentration of flue gas pollutants provided by the present invention.

[0056] Figure 11 This is the second flowchart of the intelligent optimization operation method for urban solid waste incineration provided by the present invention. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0058] The following is combined Figures 1-11 This invention describes an intelligent optimization operation method for the urban solid waste incineration process.

[0059] Figure 1This is one of the flowcharts illustrating the intelligent optimization operation method for urban solid waste incineration provided by this invention, such as... Figure 1 As shown, the method includes the following:

[0060] Step 102: During the incineration of urban solid waste, acquire the environmental data inside the incinerator and the concentration values ​​of various incineration emission gases before a preset time.

[0061] In one embodiment, the environmental data within the incinerator includes at least one of primary air temperature, primary air pressure, secondary air temperature, and chimney flue gas temperature.

[0062] In one embodiment, the incineration emissions include nitrogen oxides, carbon monoxide, carbon dioxide, sulfur dioxide, and hydrogen chloride.

[0063] In one embodiment, the preset duration can be set according to actual conditions. For example, the preset duration can be two time periods, that is, acquiring the environmental data in the incinerator at the current time (time t) and the concentration values ​​of various incineration emission gases two time periods ago (time t-2).

[0064] In one embodiment, steps 102 to 106 can be executed at first preset time intervals during the municipal solid waste incineration process to obtain set values ​​for the current flue gas oxygen content and furnace temperature during the incineration process, and to set the current flue gas oxygen content and furnace temperature. For example, steps 102 to 106 can be executed every 5 minutes to obtain new set values ​​for the flue gas oxygen content and furnace temperature.

[0065] Step 104: Iteratively optimize the solution of the decision variables so that the value of the objective function obtained from the solution of the decision variables is optimized in the direction of decreasing; the decision variables include the oxygen content of the flue gas and the furnace temperature.

[0066] The calculation of the objective function in step 102 includes: inputting the solution of the decision variables, the current environmental data in the incinerator, and the concentration values ​​of various incineration emission gases before a preset time into the pre-trained operational index model of various incineration emission gases; outputting the predicted concentration values ​​of various incineration emission gases; calculating the combustion efficiency and flue gas pollutant concentration based on the predicted concentration values ​​of various incineration emission gases; and calculating the value of the objective function based on the combustion efficiency and flue gas pollutant concentration.

[0067] The pre-trained operational index model for various incineration emissions includes multiple sub-modules, each of which outputs a predicted concentration value for one type of incineration emission gas. Specifically, each sub-module outputs a predicted concentration value for the corresponding type of incineration emission gas based on the solution of the input decision variables, the current environmental data in the incinerator, and the concentration values ​​of various incineration emission gases before a preset time period.

[0068] In one embodiment, the operational index model for various incineration emissions can be composed of multiple independent RBF (Radial basis function) neural network sub-models.

[0069] It is understandable that, since it is necessary to maximize the combustion of solid waste while minimizing the concentration of pollutants in the emitted flue gas, the objective function is used to synergistically optimize combustion efficiency and flue gas pollutant concentration, which is a multi-objective optimization problem.

[0070] In one embodiment, the objective function may include the negative of the combustion efficiency and the concentration of flue gas pollutants. The objective function at time t can be expressed as: .in, This represents the combustion efficiency at time t. This represents the concentration of pollutants in the flue gas at time t.

[0071] Step 106: Determine the solution of the target decision variable from the solutions of each group of decision variables obtained in the last iteration, and determine the values ​​of flue gas oxygen content and furnace temperature contained in the solution of the target decision variable as the set values ​​of flue gas oxygen content and furnace temperature in the current urban solid waste incineration process.

[0072] In one embodiment, when the preset maximum number of iterations is reached, the set of solutions to the decision variables obtained in the last iteration is output, and the solution to the target decision variable is determined from the set obtained in the last iteration.

[0073] The aforementioned intelligent optimization operation method for urban solid waste incineration processes determines the predicted concentrations of various incineration emission gases through a pre-trained operational index model. This model effectively addresses the complex and dynamic nature of urban solid waste incineration processes. Then, based on the predicted concentrations of various incineration emission gases, combustion efficiency and flue gas pollutant concentrations are calculated. The objective function is then calculated based on these concentrations, and the solution for the decision variables is iteratively optimized to reduce the value of the objective function obtained from the solution. This allows for accurate determination of the setpoints for flue gas oxygen content and furnace temperature, thereby improving combustion efficiency while reducing flue gas pollutant emissions. The method exhibits high adaptability and effectively solves the problems of conflicting optimization objectives and slow computation.

[0074] In one embodiment, the training steps of the pre-trained operational index model for various incineration emissions include: constructing a sample dataset based on historical data of urban solid waste incineration processes; the sample dataset contains sample flue gas oxygen content, sample furnace temperature, sample environmental data inside the incinerator, and sample concentration values ​​of various incineration emissions during the historical process of urban solid waste incineration; iteratively inputting the sample flue gas oxygen content, sample furnace temperature, sample environmental data inside the incinerator, and sample concentration values ​​of various incineration emissions before a preset time at different times into the operational index model for various incineration emissions to be trained, and outputting the sample predicted concentration values ​​of various incineration emissions at the current time; adjusting the parameters of the operational index model for various incineration emissions to be trained according to the difference between the sample predicted concentration values ​​of various incineration emissions and the actual sample concentration values ​​at the current time in the sample dataset, and entering the next iteration until the iteration stops, thus obtaining the trained operational index model for various incineration emissions.

[0075] In one embodiment, data can be collected at a second preset time interval to obtain sample flue gas oxygen content, sample furnace temperature, sample environmental data inside the incinerator, and sample concentration values ​​of various incineration emission gases at multiple moments during the history of urban solid waste incineration. For example, data can be collected every 30 seconds.

[0076] In the above embodiments, a sample dataset is constructed based on historical data of urban solid waste incineration processes. Iteratively, the oxygen content of sample flue gas, sample furnace temperature, sample environmental data within the incinerator, and sample concentration values ​​of various incineration emission gases at different times, along with the concentration values ​​of various incineration emission gases before a preset time interval, are sequentially input into the operational index models for various incineration emission gases to be trained. The model is iteratively trained based on the difference between the output predicted sample concentration values ​​and the actual sample concentration values ​​in the sample dataset, resulting in accurate operational index models for various incineration emission gases. This model effectively solves the problems of complex mechanisms and dynamic time-varying processes in urban solid waste incineration.

[0077] In one embodiment, the oxygen content of sample flue gas, sample furnace temperature, sample environmental data within the incinerator, and sample concentration values ​​of various incineration emission gases prior to a preset time at different times are iteratively input into the operational index model of various incineration emission gases to be trained, and the predicted sample concentration values ​​of various incineration emission gases at each time are output. This includes iteratively inputting the oxygen content of sample flue gas, sample furnace temperature, sample environmental data within the incinerator, and sample concentration values ​​of various incineration emission gases prior to a preset time at different times as input variables into the operational index model of various incineration emission gases to be trained. The model for various operational indicators of incineration emissions includes the input layer of each sub-module. Each sub-module outputs a sample predicted concentration value for one type of incineration emission gas. Within each sub-module, the input variables are normalized through the input layer, and the normalized input variables are then input into the hidden layer. The hidden layer performs a spatial mapping transformation on the normalized input variables, and the spatially mapped input variables are then input into the output layer. Finally, the output layer performs a weighted sum of the spatially mapped input variables to output the sample predicted concentration value of the incineration emission gas corresponding to the sub-module.

[0078] In one embodiment, the operational index model for various incineration emissions can be composed of multiple independent RBF neural network sub-models. The topology of the RBF neural network model includes three layers: an input layer, a hidden layer, and an output layer. Each independent RBF neural network sub-model is a sub-module in the operational index model for various incineration emissions, used to output a sample predicted concentration value for one type of incineration emission gas. Figure 2 This is a schematic diagram of the operational indicator models for various incineration emissions.

[0079] In one embodiment, the sample environmental data within the incinerator includes at least one of the sample primary air temperature, sample primary air pressure, sample secondary air temperature, and sample chimney flue gas temperature.

[0080] In one embodiment, the input variables for each submodule include decision variables and auxiliary variables. The decision variables include the sample flue gas oxygen content and sample furnace temperature at time t. The auxiliary variables include the sample primary air temperature, sample primary air pressure, sample secondary air temperature, sample chimney flue gas temperature, and the sample concentration values ​​of various combustion emission gases at time t-τ. Here, τ represents the preset time duration. The input variables can be represented as follows: . This indicates the sequence number of the gases emitted during incineration.

[0081] In one embodiment, the normalization of input variables by the input layer can be expressed by the following formula:

[0082]

[0083] in, This represents the input variable. This represents the maximum value among the input variables. This represents the minimum value among the input variables. This represents the input variable after normalization.

[0084] In one embodiment, the number of neurons in the hidden layer is determined as needed, and the kernel function of the neurons in the hidden layer is a Gaussian function, which performs spatial mapping transformation on the input information. The spatial mapping transformation of the normalized input variables through the hidden layer can be expressed by the following formula:

[0085]

[0086] in, This represents the model's input vector (normalized input variables). This represents the center vector of the neuron in the i-th hidden layer. This represents the width value of the neuron in the i-th hidden layer. This represents the output of the i-th hidden layer.

[0087] In one embodiment, the input variables after spatial mapping transformation are weighted and summed through the output layer. The sample predicted concentration value of the incineration emission gas corresponding to the output submodule can be expressed by the following formula:

[0088]

[0089] in, This represents the output of the submodule (i.e., the RBF neural network). This represents the model's input vector (normalized input variables). Let represent the activation function of the neuron in the i-th hidden layer. This represents the output of the i-th hidden layer. This represents the connection weights of neurons in the i-th hidden layer. This indicates the total number of hidden layers.

[0090] In one embodiment, the data output from the output layer can be denormalized to obtain the corresponding sample predicted concentration value of the incineration emission gas.

[0091] In one embodiment, data from a solid waste incineration plant in Beijing is used to verify the effectiveness of the proposed method. 3000 sets of historical data are selected as the sample dataset, with the first 2000 sets used as training samples and the remaining 1000 sets as test samples. The model is trained based on the training samples to obtain operational index models for various incineration emissions. To verify the accuracy and reliability of the constructed operational index models for various incineration emissions, independent test sample datasets are used for model validation.

[0092] 1000 sets of test samples were input into sub-modules of the operational index models for various incineration emissions, and the concentration fitting values ​​for each incineration emission gas were output. Root mean square error (RMSE) and coefficient of determination (R²) were used. 2 The fitting accuracy of each submodule is quantitatively evaluated. RMSE and R0 are used. 2 The calculation is as follows:

[0093]

[0094]

[0095] in, This represents the concentration fit values ​​of various incineration emission gases output by the operational index model for the i-th test sample. This indicates the actual concentration values ​​of various incineration emission gases in the test sample. This represents the average output of each test sample. m represents the total number of test samples. A lower RMSE indicates a higher model fit accuracy. 2 The closer to 1, the better the model fit.

[0096] Figure 3 , Figure 4 , Figure 5 , Figure 6 and Figure 7 The test results of the operational index models for nitrogen oxides, carbon monoxide, carbon dioxide, sulfur dioxide, and hydrogen chloride are shown separately. The x-axis represents the number of test samples (samples / unit), and the y-axis represents the concentration values ​​of nitrogen oxides, carbon monoxide, carbon dioxide, sulfur dioxide, and hydrogen chloride (mg / m³). 3 The solid line represents the actual value in the test sample, and * represents the network's predicted output value in the test sample. The root mean square error (RMSE) and coefficient of determination (R²) are used. 2The fitting accuracy of the operational index models for various incineration emissions was quantitatively evaluated. For the operational index model of nitrogen oxides, the root mean square error (RMSE) obtained using test samples was 15.223, and the coefficient of determination (R²) was 0.9316; for carbon monoxide, the RMSE was 0.7460, and the R² was 0.9052; for carbon dioxide, the RMSE was 0.0975, and the R² was 0.9581; for sulfur dioxide, the RMSE was 0.9470, and the R² was 0.8900; and for hydrogen chloride, the RMSE was 0.7689, and the R² was 0.8665.

[0097] In the above embodiments, a multi-module, multi-task neural network model is constructed and trained to obtain operational index models for various incineration emissions. Each module can output the concentration value of one type of incineration emission gas, thereby enabling accurate prediction of the concentration values ​​of various incineration emissions through the trained operational index models. This model effectively solves the problems of complex mechanisms and dynamic time-varying processes in urban solid waste incineration.

[0098] In one embodiment, the incineration emissions include nitrogen oxides, carbon monoxide, carbon dioxide, sulfur dioxide, and hydrogen chloride. Calculating combustion efficiency and flue gas pollutant concentrations based on predicted concentration values ​​of the various incineration emissions includes: calculating combustion efficiency based on predicted concentration values ​​of carbon dioxide and carbon monoxide; and calculating flue gas pollutant concentrations based on predicted concentration values ​​of nitrogen oxides, carbon monoxide, sulfur dioxide, and hydrogen chloride.

[0099] In one embodiment, the predicted concentrations of carbon dioxide and carbon monoxide can be summed, and the combustion efficiency can be determined based on the ratio of the predicted carbon dioxide concentration to the summation result.

[0100] In one embodiment, the formula for calculating combustion efficiency is as follows:

[0101]

[0102] in, This represents the combustion efficiency at time t. This represents the predicted concentration of carbon dioxide at time t. This represents the predicted concentration of carbon monoxide at time t.

[0103] In one embodiment, the predicted concentration values ​​of various flue gas pollutants in the combustion emissions can be weighted and summed to obtain the flue gas pollutant concentration. Flue gas pollutants may include carbon monoxide, nitrogen oxides, sulfur dioxide, and hydrogen chloride. The weights of each flue gas pollutant are determined by the dimensions of the gas average value.

[0104] The formula for calculating the concentration of pollutants in flue gas is as follows:

[0105]

[0106] in, This represents the concentration of pollutants in the flue gas at time t. This represents the predicted concentration of carbon monoxide at time t. This represents the predicted concentration of nitrogen oxides at time t. This represents the predicted concentration of sulfur dioxide at time t. This represents the predicted concentration of hydrogen chloride at time t. , , , These represent the weights of carbon monoxide, nitrogen oxides, sulfur dioxide, and hydrogen chloride, respectively.

[0107] In the above embodiments, the combustion efficiency can be accurately calculated based on the predicted concentration values ​​of carbon dioxide and carbon monoxide, and the concentration of flue gas pollutants can be accurately calculated based on the predicted concentration values ​​of nitrogen oxides, carbon monoxide, sulfur dioxide and hydrogen chloride. Furthermore, the objective function can be accurately calculated based on the combustion efficiency and the concentration of flue gas pollutants, thereby optimizing the decision variables.

[0108] In one embodiment, iteratively optimizing the solutions to the decision variables to optimize the value of the objective function obtained from the solutions to the decision variables in a decreasing direction includes: initializing a population of decision variables as the original population of decision variables in the first iteration; in each iteration, performing a mutation operation on each individual in the original population of decision variables in the current iteration to obtain a mutated population; performing a crossover operation on the original population of decision variables and the mutated population in the current iteration to obtain a trial population; merging the trial population and the original population of decision variables to obtain a mixed population; determining the original population of decision variables for the next iteration based on the values ​​of the objective function obtained from the solutions to each decision variable in the mixed population, and entering the next iteration.

[0109] The decision variable population is a population of solutions containing multiple sets of decision variables. Each set of solutions includes values ​​for flue gas oxygen content and furnace temperature.

[0110] In one embodiment, the population size Np, mutation operator F, crossover rate CR, and maximum number of iterations G can be set. For example, the population size Np can be set to 100, the mutation operator F can be set to 0.5, the crossover rate CR can be set to 0.7, and the maximum number of iterations G can be set to 30.

[0111] In one embodiment, to maintain population diversity, a niche technique is used to initialize the decision variable population. Individuals in the decision variable population (i.e., solutions to the decision variables) are randomly numbered according to the population size Np and organized on a circular topology. For each individual, a neighborhood radius R is defined, and the R individuals within that neighborhood are grouped into a neighborhood group, thus assigning each individual an independent neighborhood. The index-based circular topology restricts direct interactions between individuals in different neighborhoods, thereby forming multiple stable search spaces during the search for solutions to the decision variables and effectively maintaining population diversity.

[0112] In one embodiment, Np real values ​​are used as individuals in the initial decision variable population. Each individual in the decision variable population is represented as:

[0113]

[0114] Where i represents the individual's index in the decision variable population, Np represents the population size, Gen represents the current iteration number, and G represents the maximum iteration number.

[0115] In one embodiment, the individuals (solutions to the decision variables) in the decision variable population are determined by constraints set based on actual industrial needs, with the limits of the parameter variables set as follows: Then the constraints on the individuals in the population for the decision variable are:

[0116]

[0117]

[0118] During initialization, Gen=0 is set. This represents a uniformly random number generated between [0,1]. Denotes the upper bound of the decision variable in the j-th dimension. This represents the lower bound of the decision variable in the j-th dimension. There are two defined decision variables (i.e., flue gas oxygen content and furnace temperature), so j = 1, 2. i represents the index of the individual in the population of decision variables. Np represents the population size. This represents the value of the j-th decision variable in the i-th individual of the decision variable population in the 0th iteration (i.e., initialization).

[0119] In one embodiment, in each iteration, a mutation operation is performed on each individual in the original population of decision variables for that iteration, resulting in a mutated population. A crossover operation is then performed on both the original and mutated populations to obtain a trial population. Boundary condition checks are then performed to ensure that the individuals in the trial population meet the constraints. The individuals in the trial population can be represented by the following formula:

[0120]

[0121] in, This represents the j-th estimate of a random number generator between [0,1]. CR represents the crossover operator. This represents the mutated value of the j-th decision variable in the solution of the i-th decision variable in the original population of the Gen-th iteration. Let represent the value of the j-th decision variable in the solution of the i-th decision variable in the original population during the Gen-th iteration. Let represent the value of the j-th decision variable in the solution of the i-th decision variable in the original population of the Gen+1-th iteration.

[0122] like Figure 8 The diagram illustrates the process of iteratively optimizing the solution of decision variables. First, the decision variable population is initialized to obtain the original population. Before reaching the maximum number of iterations, a mutation operation is performed on the original population in each iteration. If the current iteration number Gen is less than or equal to 20% of the maximum iteration number G, a single mutation strategy (mutation strategy) is used. If the current iteration number Gen is greater than or less than 20% of the maximum iteration number G, an adaptive mutation strategy is used, that is, multiple mutation strategies are assigned to each individual in the original population based on the performance of each mutation strategy in the previous iteration. Then, a crossover operation is performed between the mutated population and the original population to obtain a trial population. The trial population and the original population are merged to obtain a mixed population. An environment selection mechanism is used to select the original population for the next iteration from the mixed population, and then the next iteration begins. Iteration stops when the maximum number of iterations is reached.

[0123] In the above embodiments, in each iteration, a mutation operation is performed on individuals in the original population of decision variables. Then, a crossover operation is performed on the original population and the mutated population to obtain an experimental population. The experimental population and the original population of decision variables are merged to obtain a mixed population. Based on the objective function values ​​obtained from the solutions of each decision variable in the mixed population, the original population of decision variables for the next iteration is determined, and the next iteration begins. Mutation and crossover improve population diversity, thereby improving the accuracy of the solutions to the decision variables obtained through optimization.

[0124] In one embodiment, performing a mutation operation on each individual in the original population of decision variables in the current iteration to obtain a mutated population includes: if the current iteration number is less than or equal to the target iteration number, then using a single mutation strategy to perform a mutation operation on each individual in the original population of decision variables in the current iteration to obtain a mutated population; the target iteration number is obtained according to a preset proportion of the maximum iteration number; if the current iteration number is greater than the target iteration number, then allocating multiple mutation strategies to each individual in the original population of decision variables in the current iteration according to their respective target proportions, and performing a mutation operation on the corresponding individuals using the allocated mutation strategy to obtain multiple mutated populations; wherein, the target proportion is determined based on the performance of each mutation strategy in the previous iteration.

[0125] In one embodiment, a single mutation strategy can be a DE / rand / 1 mutation strategy. Multiple mutation strategies can include a DE / rand / 1 mutation strategy and a DE / best / 1 mutation strategy. That is, if the current iteration number is less than or equal to the target iteration number, the DE / rand / 1 mutation strategy is used to perform a mutation operation on each individual in the original population of decision variables in this iteration, resulting in a mutated population. If the current iteration number is greater than the target iteration number, the DE / rand / 1 mutation strategy and the DE / best / 1 mutation strategy are allocated to each individual in the original population of decision variables in this iteration according to their respective target proportions, and the allocated mutation strategy is used to perform a mutation operation on the corresponding individual, resulting in mutated populations corresponding to the DE / rand / 1 mutation strategy and the DE / best / 1 mutation strategy, respectively.

[0126] In one embodiment, the preset ratio can be set according to actual conditions. For example, the preset ratio can be 20%, that is, if the current iteration number is less than or equal to 20% of the maximum iteration number, a single mutation strategy is adopted; if the current iteration number is greater than 20% of the maximum iteration number, multiple mutation strategies are adopted.

[0127] In one embodiment, in the process of assigning multiple mutation strategies to each individual in the original population of decision variables in the current iteration according to their respective target proportions, a randomly arranged index array is first generated based on the target proportions of each mutation strategy. The mutation strategy for each individual is determined by the values ​​in the index array. Based on the index array, a mutation strategy is assigned to each individual in the population of decision variable elements and stored in a matrix.

[0128] In one embodiment, the individuals in the mutant population obtained using the DE / rand / 1 mutation strategy are represented as follows:

[0129]

[0130] Where r1, r2, and r3 are randomly selected distinct indices, and F represents the mutation operator.

[0131] In one embodiment, the individuals in the mutant population obtained using the DE / best / 1 mutation strategy are represented as follows:

[0132]

[0133] Where r1 and r2 are randomly selected distinct indices, x best It is the optimal solution found up to the current iteration. F represents the mutation operator.

[0134] In the above embodiments, a single mutation strategy is used when the number of iterations is small, while multiple mutation strategies are used when the number of iterations is large. The usage ratio of each mutation strategy is determined based on the performance of each mutation strategy in the previous iteration, thereby improving the effectiveness of the mutation operation and thus increasing population diversity.

[0135] In one embodiment, the step of determining the target proportion includes: determining the first fitness of the solutions to the decision variables in the mutant population obtained by each mutation strategy in the previous iteration, and the second fitness of the solutions to the decision variables in the original population of the decision variables in the previous iteration; the fitness is determined based on the value of the objective function; determining the performance corresponding to each mutation strategy based on the distance between the first fitness and the second fitness corresponding to each mutation strategy; and determining the target proportion corresponding to each mutation strategy based on the ratio between the performance corresponding to each mutation strategy and the total performance of all mutation strategies.

[0136] In one embodiment, the distance between fitness values ​​can be calculated using the following formula:

[0137]

[0138] in, The Pareto front generated by the algorithm is the set of non-dominated solutions in the current population. The actual Pareto front is difficult to obtain in practical applications, so an approximate Pareto front is used as a reference. For the true Pareto frontier The number of points in the array. v is... A point in the equation represents an ideal solution. u is the Pareto front generated by the algorithm. A point in the equation represents a solution found by the algorithm in the current iteration. d(v,u) is the Euclidean distance between the fitness of points v and u, which measures the distance between the solutions generated by the algorithm and the fitness of points on the true Pareto front.

[0139] In one embodiment, the performance of each mutation strategy can be calculated using the following formula:

[0140]

[0141]

[0142] in, This indicates the performance of the DE / rand / 1 mutation strategy. This indicates the performance of the DE / best / 1 mutation strategy. and represents the fitness (first fitness) of individuals in the mutant population obtained using the DE / rand / 1 mutation strategy and the fitness (first fitness) of individuals in the mutant population obtained using the DE / best / 1 mutation strategy, respectively. The DE / rand / 1 mutation strategy has good exploratory search capabilities, while the DE / best / 1 mutation strategy excels in local search and converges faster. This represents the fitness (second fitness) of an individual in the original population, which is a decision variable. This represents the distance between the fitness of individuals in the mutant population obtained using the DE / rand / 1 mutation strategy and the fitness of individuals in the original population, which is the decision variable. This represents the distance between the fitness of individuals in the mutated population obtained using the DE / best / 1 mutation strategy and the fitness of individuals in the original population (the decision variable). `rand` is a small random number added to increase numerical stability.

[0143] In one embodiment, the total performance of various mutation strategies can be determined by summing the performance of each strategy. The target proportion for each mutation strategy is obtained by comparing its performance to the total performance of all mutation strategies.

[0144] In one embodiment, the overall performance of various mutation strategies is calculated using the following formula:

[0145]

[0146] in, This represents a small constant, used to avoid the error of dividing by zero. This represents the overall performance of various mutation strategies. Indicates the sequence number of the mutation strategy. This represents the performance of the i-th mutation strategy.

[0147] In one embodiment, the number of individuals using each mutation strategy in this iteration can be calculated using the following formula:

[0148]

[0149]

[0150] in, This indicates the number of individuals that used the first mutation strategy (DE / rand / 1 mutation strategy). This indicates the number of individuals that used the second mutation strategy (DE / rand / 1 mutation strategy). This indicates rounding. This indicates the performance of the first mutation strategy. This indicates the performance of the second mutation strategy. This represents the overall performance of various mutation strategies. This represents the target proportion corresponding to the first mutation strategy. This represents the target proportion corresponding to the second mutation strategy. (Indicates the population size of the decision variable elements).

[0151] In the above embodiments, the performance of each mutation strategy is determined based on the distance between the fitness of individuals in the mutated population obtained by each mutation strategy in the previous iteration and the fitness of individuals in the original population, which is the decision variable. Based on the performance of each mutation strategy, the proportion of individuals using each mutation strategy in the current iteration is determined, which can accurately adjust the mutation strategy and thus better improve population diversity.

[0152] In one embodiment, determining the original population of decision variables for the next iteration based on the values ​​of the objective functions obtained from the solutions of each decision variable in the mixed population includes: dividing the mixed population into multiple clusters; determining the non-dominated and dominated solutions in each cluster based on the values ​​of the objective functions corresponding to the solutions of the decision variables in each cluster, assigning the non-dominated solutions to the elite archive, and assigning the dominated solutions to the non-elite archive; and determining the original population of decision variables for the next iteration based on the elite archive.

[0153] In one embodiment, an affinity propagation algorithm can be used to divide the mixed population into multiple clusters.

[0154] In the above embodiments, the mixed population is divided into multiple clusters. Based on the values ​​of the objective functions corresponding to the solutions of the decision variables in each cluster, the non-dominated and dominated solutions in each cluster are determined. The non-dominated solutions are assigned to the elite archive, and the dominated solutions are assigned to the non-elite archive. This can accurately determine the original population of decision variables in the next iteration.

[0155] In one embodiment, determining the original population of decision variables for the next iteration based on the elite archive includes: if the number of solutions to decision variables in the elite archive does not reach a preset scale, quantifying the distribution density of solutions to decision variables in the decision space in the non-elite archive, selecting solutions of decision variables in sparse regions based on the distribution density to supplement the elite archive, and selecting solutions of multiple decision variables from the supplemented elite archive to form the original population of decision variables for the next iteration; if the number of solutions to decision variables in the elite archive reaches a preset scale, directly selecting solutions of multiple decision variables from the elite archive based on the values ​​of the objective functions corresponding to the solutions of each decision variable in the elite archive to form the original population of decision variables for the next iteration.

[0156] In one embodiment, the harmonic mean distance criterion (HAD) can be used to quantify the distribution density of solutions to decision variables in the decision space for non-elite archives.

[0157] In one embodiment, the formula for calculating the distribution density is:

[0158]

[0159] Where S represents the set of all solutions in the population. |S| represents the total number of solutions. Y represents another solution in the population. This represents the Euclidean distance between solutions x and y.

[0160] In the above embodiments, if the number of solutions to decision variables in the elite archive does not reach a preset scale, the distribution density of solutions to decision variables in the non-elite archive in the decision space is quantified, and solutions to decision variables in sparse regions are selected to supplement the elite archive based on the distribution density. Solutions to multiple decision variables are then selected from the supplemented elite archive to form the original population of decision variables in the next iteration. If the number of solutions to decision variables in the elite archive reaches a preset scale, solutions to multiple decision variables are selected from the elite archive based on the values ​​of the objective functions corresponding to the solutions to each decision variable in the elite archive to form the original population of decision variables in the next iteration, thereby accurately determining the original population of decision variables in the next iteration.

[0161] In one embodiment, determining the solution of the target decision variable from the solutions of each group of decision variables obtained in the last iteration includes: for each group of decision variable solutions obtained in the last iteration, respectively, performing a weighted summation of the first optimization target value and the second optimization target value corresponding to the solution of the decision variable according to the first preset weight and the second preset weight, respectively, to obtain the weighted summation result corresponding to the solution of the decision variable; the first optimization target value is determined based on the combustion efficiency corresponding to the solution of the decision variable; the second optimization target value is determined based on the flue gas pollutant concentration corresponding to the solution of the decision variable; and the solution of the decision variable with the smallest weighted summation result is determined as the solution of the target decision variable.

[0162] In this context, the solution for each set of decision variables obtained in the final iteration refers to the set of solutions for the decision variables determined based on the elite files in the final round.

[0163] In one embodiment, the weighted summation of the first and second objective values ​​corresponding to the solutions of the decision variables can be performed using the following formula to obtain the weighted summation result corresponding to the solutions of the decision variables:

[0164]

[0165] in, These are the corresponding preset weights. ij Let be the j-th optimization objective value of the solution for the i-th group of decision variables. n is the number of optimization objectives (n=2 in this invention). m is the number of solutions in the Pareto optimal solution set (i.e., the set of solutions to the decision variables obtained in the last iteration).

[0166] It is understandable that when the weighted summation result... When the minimum value is reached, it indicates that the solution of the i-th decision variable is the optimal setpoint, and the optimal setpoint value of decision variable x can be obtained. Then, the optimal setpoint values ​​of furnace temperature and flue gas oxygen content can be determined, thereby improving combustion efficiency while reducing the emission concentration of flue gas pollutants.

[0167] In one embodiment, the settings are configured according to actual conditions and decision preferences. The weight of the first optimization objective (combustion efficiency) is set to 0.6, and the weight of the second optimization objective (flue gas pollutant concentration) is set to 0.4. During actual incineration, the decision variables need to be optimized once every first preset time interval. In each optimization process, the solution with the minimum objective function value (the solution of the objective decision variable) is calculated from the Pareto optimal solution set obtained in the last iteration using a linear weighted method, and this solution is determined as the optimal setpoint for furnace temperature and flue gas oxygen content.

[0168] In the above embodiments, for each set of decision variable solutions obtained in the last iteration, the first and second optimization target values ​​corresponding to the solutions of the decision variables are weighted and summed according to the first and second preset weights, respectively, to obtain the weighted summation result corresponding to the solutions of the decision variables. The solution of the decision variable with the smallest weighted summation result is determined as the solution of the target decision variable. This can balance the weights between the two targets of combustion efficiency and flue gas pollutant concentration, and accurately determine the solution of the target decision variable.

[0169] like Figure 9 and Figure 10The figures shown are the optimization results of combustion efficiency and flue gas pollutant concentration based on the above method. The x-axis represents the optimization time in minutes, and the y-axis represents the combustion efficiency and flue gas pollutant concentration in % and mg / m³, respectively. 3 The white bars represent the actual output values ​​of combustion efficiency and flue gas pollutant concentration before optimization, while the black bars represent the actual output values ​​of combustion efficiency and flue gas pollutant concentration after optimization. As can be seen from the graph, the optimized combustion efficiency is significantly improved compared to the unoptimized version, with an average improvement of 7.46%. The optimized flue gas pollutant concentration is significantly reduced compared to the unoptimized version, with an average reduction of 12.80%.

[0170] like Figure 11 The diagram shown is an overall flowchart of the intelligent optimization operation method for urban solid waste incineration in various embodiments of the present invention, including the following steps:

[0171] Step 1: Collect historical data on urban solid waste incineration processes to construct a sample dataset, and divide it into a training set and a test set.

[0172] Step 2: Use a multi-module, multi-task neural network to establish operational index models for nitrogen oxides, carbon monoxide, carbon dioxide, sulfur dioxide, and hydrogen chloride.

[0173] Step 3: Design and optimize the objective function based on the combustion efficiency and flue gas pollutant emission concentration of urban solid waste incineration.

[0174] Step 4: Using the designed optimization objective function as the objective function of the algorithm, the Pareto optimal solution set of the decision variables is obtained by using a multi-objective adaptive differential evolution algorithm based on the environment selection mechanism.

[0175] Step 5: Using the linear weighted sum method to balance the weights between the objectives, obtain the optimal settings of the decision variables from the Pareto optimal solution set.

[0176] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent optimization operation method for municipal solid waste incineration process, characterized in that, Comprise: In the process of municipal solid waste incineration, obtain the concentration values of various incineration exhaust gases before the preset time length and the environmental data in the current incinerator; Iteratively optimize the solution of the decision variables so that the value of the objective function obtained according to the solution of the decision variables is optimized in the direction of reduction; the decision variables include oxygen content in flue gas and hearth temperature; Determine the solution of the target decision variable from the solutions of the decision variables obtained in the last round of iteration, and determine the values of the oxygen content in flue gas and the hearth temperature contained in the solution of the target decision variable as the set values of the oxygen content in flue gas and the hearth temperature in the current municipal solid waste incineration process; Wherein, the calculation step of the value of the objective function comprises: inputting the solution of the decision variable, the environmental data in the current incinerator and the concentration values of various incineration exhaust gases before the preset time length into the pre-trained operation index model of various incineration exhaust gases, outputting the predicted concentration values of the current various incineration exhaust gases, calculating the combustion efficiency and the concentration of flue gas pollutants according to the predicted concentration values of the various incineration exhaust gases, and calculating the value of the objective function according to the combustion efficiency and the concentration of flue gas pollutants; The training step of the pre-trained operation index model of various incineration exhaust gases comprises: Based on the historical data of municipal solid waste incineration process, a sample data set is constructed; the sample data set contains sample oxygen content in flue gas, sample hearth temperature, sample environmental data in the incinerator and sample concentration values of various incineration exhaust gases in the historical municipal solid waste incineration process; Iteratively input the sample oxygen content in flue gas, the sample hearth temperature, the sample environmental data in the incinerator and the sample concentration values of various incineration exhaust gases before the preset time length at the time into the operation index model of various incineration exhaust gases to be trained in turn, and output the sample predicted concentration values of various incineration exhaust gases at the time; According to the difference between the sample predicted concentration values of various incineration exhaust gases and the real sample concentration values at the time in the sample data set, adjust the parameters of the operation index model of various incineration exhaust gases to be trained and enter the next round of iteration until the iteration is stopped, and the trained operation index model of various incineration exhaust gases is obtained.

2. The method for intelligent optimization and operation of municipal solid waste incineration process according to claim 1, characterized in that, The iterative input of the sample oxygen content in flue gas, the sample hearth temperature, the sample environmental data in the incinerator and the sample concentration values of various incineration exhaust gases before the preset time length at the time into the operation index model of various incineration exhaust gases to be trained in turn, and output the sample predicted concentration values of various incineration exhaust gases at the time, comprises: Iteratively input the sample oxygen content in flue gas, the sample hearth temperature, the sample environmental data in the incinerator and the sample concentration values of various incineration exhaust gases before the preset time length at the time into the input layer of each sub-module of the operation index model of various incineration exhaust gases to be trained as input variables; wherein each sub-module is used to output the sample predicted concentration value of one kind of incineration exhaust gas; In each of the sub-modules, the input variables are normalized by the input layer, the normalized input variables are input to the hidden layer, the normalized input variables are spatially mapped by the hidden layer, the spatially mapped input variables are input to the output layer, the spatially mapped input variables are weighted and summed by the output layer, and the sample predicted concentration values of the incineration exhaust gas corresponding to the sub-modules are output.

3. The method for intelligent optimization operation of municipal solid waste incineration process according to claim 1, characterized in that, The incineration exhaust gas includes nitrogen oxides, carbon monoxide, carbon dioxide, sulfur dioxide, and hydrogen chloride. The calculation of the combustion efficiency and the flue gas pollutant concentration according to the predicted concentration values of the various incineration exhaust gases includes: The combustion efficiency is calculated according to the predicted concentration values of the carbon dioxide and the carbon monoxide. The flue gas pollutant concentration is calculated according to the predicted concentration values of the nitrogen oxides, the carbon monoxide, the sulfur dioxide, and the hydrogen chloride.

4. The intelligent optimization operation method of municipal solid waste incineration process according to any one of claims 1 to 3, characterized in that, The iterative optimization of the solutions of the decision variables to optimize the values of the objective functions obtained according to the solutions of the decision variables in a decreasing direction includes: The decision variable population is initialized as the decision variable original population in the first iteration; In each iteration, a mutation operation is performed on each individual in the decision variable original population in the current iteration to obtain a mutation population; A crossover operation is performed on the decision variable original population in the current iteration and the mutation population to obtain a trial population; The trial population and the decision variable original population are merged to obtain a hybrid population; According to the values of the objective functions obtained by the solutions of the decision variables in the hybrid population, the decision variable original population in the next iteration is determined, and the next iteration is entered.

5. The intelligent optimization operation method of municipal solid waste incineration process according to claim 4, characterized in that, The mutation operation performed on each individual in the decision variable original population in the current iteration to obtain a mutation population includes: If the current iteration number is less than or equal to the target iteration number, a single mutation strategy is used to perform a mutation operation on each individual in the decision variable original population in the current iteration to obtain a mutation population; the target iteration number is obtained according to a preset proportion of the maximum iteration number; If the current iteration number is greater than the target iteration number, multiple mutation strategies are assigned to each individual in the decision variable original population in the current iteration according to their target proportions, and a mutation operation is performed on the corresponding individual using the assigned mutation strategy to obtain multiple mutation populations; The target proportion is determined according to the performance of each mutation strategy in the last iteration.

6. The intelligent optimization operation method of municipal solid waste incineration process according to claim 5, characterized in that, The determination of the target proportion includes: The first fitness of the solutions of the decision variables in the mutation population obtained by each mutation strategy in the last iteration, and the second fitness of the solutions of the decision variables in the decision variable original population in the last iteration are determined; the fitness is determined according to the value of the objective function; According to the distances between the first fitness and the second fitness corresponding to each mutation strategy, the performance corresponding to each mutation strategy is determined. The distances between the first fitness and the second fitness corresponding to each mutation strategy are determined according to the following formula: Determine the target proportion of each mutation strategy according to the proportion between the performance of each mutation strategy and the total performance of all mutation strategies.

7. The intelligent optimization operation method of municipal solid waste incineration process according to claim 4, characterized in that, Determine the decision variable original population in the next iteration according to the values of the objective function obtained from the solutions of each decision variable in the mixed population. Divide the mixed population into multiple clusters. Determine the non-dominated solution and the dominated solution in each cluster according to the values of the objective function corresponding to the solutions of the decision variables in each cluster, and divide the non-dominated solutions into the elite archive and the dominated solutions into the non-elite archive. Determine the decision variable original population in the next iteration according to the elite archive.

8. The intelligent optimization operation method of municipal solid waste incineration process according to claim 7, characterized in that, Determine the decision variable original population in the next iteration according to the elite archive. If the number of solutions of the decision variables in the elite archive does not reach the preset scale, quantify the distribution density of the solutions of the decision variables in the non-elite archive in the decision space, select the solutions of the decision variables in the sparse area according to the distribution density, and supplement them to the elite archive, and select multiple solutions of the decision variables from the supplemented elite archive to form the decision variable original population in the next iteration. If the number of solutions of the decision variables in the elite archive reaches the preset scale, directly select multiple solutions of the decision variables from the elite archive according to the values of the objective function corresponding to the solutions of each decision variable in the elite archive to form the decision variable original population in the next iteration.

9. The intelligent optimization operation method of municipal solid waste incineration process according to any one of claims 1 to 3, characterized in that, Determine the solution of the target decision variable from each group of solutions of the decision variables obtained in the last iteration, including: For each group of solutions of the decision variables obtained in the last iteration, respectively, according to the first preset weight and the second preset weight, the first optimization target value and the second optimization target value corresponding to the solution of the decision variable are weighted and summed to obtain the weighted sum result corresponding to the solution of the decision variable; the first optimization target value is determined according to the combustion efficiency corresponding to the solution of the decision variable; the second optimization target value is determined according to the flue gas pollutant concentration corresponding to the solution of the decision variable; Determine the solution of the target decision variable as the solution of the decision variable with the smallest weighted sum result.

Images