Incineration optimization method based on genetic algorithm improvement, electronic equipment and program product

Abstract

The invention relates to the technical field of combustion control optimization, in particular to an incineration optimization method based on genetic algorithm improvement, electronic equipment and a program product. The incineration optimization method comprises the steps that to-be-optimized incineration control parameters meeting preset conditions are determined; based on the to-be-optimized incineration control parameters, an initial population composed of a plurality of individuals is randomly generated; calculating the fitness of each individual in the initial population according to a fitness algorithm; based on the initial population and the fitness, performing iterative optimization calculation by using a genetic algorithm until a preset iteration termination condition is reached; and outputting the optimized incineration control parameters. The incineration process is accurately controlled, the problems existing in a traditional manual control optimization method can be solved, and the efficient and environment-friendly incineration process is achieved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of combustion control optimization, and in particular to an improved incineration optimization method, electronic equipment and program product based on a genetic algorithm. Background Art

[0002] Control optimization of body incineration is an increasingly important topic. With the increasing attention to environmental protection and energy consumption control, how to incinerate bodies efficiently and environmentally friendly has become an urgent problem to be solved. In traditional manual control optimization, operators often rely on experience to make adjustments. The limitation of this method is that it is easily affected by human factors and cannot respond to changes in the incineration process in a timely manner, resulting in inaccurate control of the incineration process, which in turn affects the incineration efficiency and the concentration of pollutant emissions. Summary of the invention

[0003] Based on the above technical problems, the present disclosure proposes an incineration optimization method, electronic device and program product improved based on genetic algorithm.

[0004] According to the first aspect of the present disclosure, an incineration optimization method improved based on a genetic algorithm is provided, comprising: determining incineration control parameters to be optimized that meet preset conditions; based on the incineration control parameters to be optimized, randomly generating an initial population consisting of a number of individuals; calculating the fitness of each individual in the initial population according to a fitness algorithm; based on the initial population and the fitness, performing iterative optimization calculations using a genetic algorithm until a preset iteration termination condition is reached; and outputting the optimized incineration control parameters.

[0005] In addition, according to the improved incineration optimization method based on genetic algorithm according to the first aspect of the present disclosure, the incineration control parameters to be optimized that meet predetermined conditions are determined, including: obtaining a first incineration control parameter set including incineration control parameters corresponding to multiple experimental samples; based on the first incineration control parameter set, obtaining multiple second incineration control parameter sets, each second incineration control parameter set including the first incineration control parameters corresponding to all experimental samples at the same time point; obtaining the first pollutant emission concentration and the first instantaneous fuel flow rate corresponding to each first incineration control parameter; determining the first incineration control parameter of the experimental sample in each second incineration control parameter set whose first pollutant emission concentration is lower than the pollutant emission concentration limit and whose first instantaneous fuel flow rate is lower than the median value of all first instantaneous fuel flow rates in the second incineration control parameter set as the incineration control parameter to be optimized.

[0006] In addition, according to the incineration optimization method based on genetic algorithm improvement according to the first aspect of the present disclosure, the incineration control parameters to be optimized include at least: main combustion chamber temperature, average temperature, secondary combustion chamber wind speed, top wind speed, left side wind speed, right side wind speed, left smoke elimination wind speed, right smoke elimination wind speed, burner status, blower status, main oil valve status and one or more of the number of minutes of incineration; the first pollutant emission concentration includes at least one or more of carbon monoxide CO emission concentration, nitrogen dioxide NO2 emission concentration and sulfur dioxide SO2 emission concentration.

[0007] In addition, according to the improved incineration optimization method based on genetic algorithm of the third aspect of the present disclosure, each individual represents a set of third incineration control parameters, and the fitness of each individual in the initial population is calculated according to the fitness algorithm, including: obtaining the pollutant emission concentration corresponding to each set of third incineration control parameters: inputting each set of third incineration control parameters into the pollutant emission concentration prediction model, and obtaining the third pollutant emission concentration output by the pollutant emission concentration prediction model; obtaining the third fuel instantaneous flow rate corresponding to each set of third incineration control parameters: inputting each set of third incineration control parameters into the fuel instantaneous flow rate prediction model, and obtaining the third fuel instantaneous flow rate output by the fuel instantaneous flow rate prediction model; determining the fitness based on the third pollutant emission concentration and the third fuel instantaneous flow rate.

[0008] In addition, according to the incineration optimization method based on genetic algorithm improvement of the first aspect of the present disclosure, the third pollutant emission concentration includes CO emission concentration, SO2 emission concentration and NO2 emission concentration; the fitness satisfies the following relationship: Among them, fitness is the fitness, represents the CO emission concentration, Indicates the SO2 emission concentration, Indicates NO2 emission concentration, Flow indicates the instantaneous flow rate of the third fuel, , and Respectively represent the weight values ​​of CO emission concentration, SO2 emission concentration and NO2 emission concentration, Indicates the weight value of the third fuel instantaneous flow rate.

[0009] In addition, according to the improved incineration optimization method based on genetic algorithm according to the first aspect of the present disclosure, the preset iterative termination conditions include: the number of iterations meets the preset threshold, based on the initial population and fitness, using the genetic algorithm to perform iterative optimization calculations until the preset iterative termination conditions are reached, including: using the initial population and fitness as initial parameters, inputting the genetic algorithm for calculation; iteratively optimizing the initial population, wherein each iteration generates a new population based on selection, crossover, and mutation; until the number of iterative optimization processing meets the preset iterative termination conditions.

[0010] In addition, according to the incineration optimization method based on genetic algorithm improvement according to the first aspect of the present disclosure, a first incineration control parameter set including incineration control parameters corresponding to a plurality of experimental samples is obtained, including: The sensor data and infrared temperature measurement data of the incinerator are collected on a daily basis. The sensor data include: main combustion chamber temperature, average temperature, secondary combustion chamber wind speed, top wind speed, left wind speed, right wind speed, left smoke elimination wind speed, right smoke elimination wind speed, burner status, blower status, and main oil valve status; the infrared temperature measurement data include: temperature values ​​of various points in the incineration furnace; the sensor data and infrared temperature measurement data are aligned and downsampled on a minute basis to obtain a first incineration control parameter set including incineration control parameters corresponding to multiple experimental samples.

[0011] In addition, according to the incineration optimization method based on genetic algorithm improvement of the first aspect of the present disclosure, the method also includes: outputting the pollutant emission concentration and the instantaneous fuel flow rate corresponding to the optimized incineration control parameters.

[0012] The present disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the improved incineration optimization method based on the genetic algorithm as described above is implemented.

[0013] The present disclosure also provides a computer program product, including a computer program, which implements any of the above-mentioned improved incineration optimization methods based on genetic algorithms when executed by a processor.

[0014] The present disclosure provides an improved incineration optimization method, electronic device and program product based on a genetic algorithm. The present disclosure determines the incineration control parameters to be optimized that meet preset conditions, randomly generates an initial population using the incineration control parameters to be optimized, calculates the fitness corresponding to a number of individuals in the initial population, and iterates using a genetic algorithm to output the optimized incineration control parameters. Compared with the traditional manual control optimization method, the incineration optimization method using a genetic algorithm can automatically and quickly find the optimal or near-optimal incineration control parameters. The method is not limited by human factors and can efficiently search for the optimized incineration control parameters, thereby achieving precise control of the incineration process, and can solve the problems existing in the traditional manual control optimization method to achieve an efficient and environmentally friendly incineration process. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the present disclosure or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 It is a schematic diagram illustrating a flow chart of an improved incineration optimization method based on a genetic algorithm provided in accordance with the present disclosure.

[0017] Figure 2 It is a schematic diagram illustrating the average temperature change of the No. 8 incinerator of the Dongjiao Funeral Home in one day according to an embodiment of the present disclosure.

[0018] Figure 3 It is a schematic diagram illustrating a line graph of average temperature-time of corpse combustion provided according to an embodiment of the present disclosure.

[0019] Figure 4 2 is a schematic diagram illustrating the actual value-predicted value of NO2, SO2 and CO provided according to an embodiment of the present disclosure.

[0020] Figure 5 It is a flowchart illustrating a genetic algorithm provided according to an embodiment of the present disclosure.

[0021] Figure 6 It is a schematic diagram illustrating the actual value-predicted value of the instantaneous fuel flow rate provided according to an embodiment of the present disclosure.

[0022] Figure 7 is a schematic diagram illustrating an electronic device provided according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the present disclosure more obvious, the exemplary embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments of the present disclosure, and it should be understood that the present disclosure is not limited to the exemplary embodiments described here.

[0024] like Figure 1 As shown, the incineration optimization method based on genetic algorithm improvement provided in the embodiment of the present disclosure specifically includes the following steps.

[0025] In step 101, the incineration control parameters to be optimized that meet preset conditions are determined.

[0026] In one embodiment of the present disclosure, the incineration control parameters to be optimized refer to those parameters that have a significant impact on key performance indicators such as incineration efficiency, pollutant emissions, and energy consumption during the incineration process. In the process of incineration optimization, it is necessary to first clarify the incineration control parameters to be optimized that need to meet preset conditions during the incineration process. Among them, the preset conditions refer to pre-set standards, which can be determined based on historical data, simulation results, expert experience, energy consumption requirements, pollutant emissions or regulatory requirements. By determining the incineration control parameters to be optimized that meet the predetermined conditions, it is helpful to better optimize the incineration control parameters during the incineration optimization process.

[0027] In one embodiment of the present disclosure, the incineration control parameters to be optimized include at least but are not limited to: one or more of the main combustion chamber temperature, average temperature, secondary combustion chamber wind speed, top wind speed, left wind speed, right wind speed, left smoke elimination wind speed, right smoke elimination wind speed, burner state, blower state, main oil valve state, and the number of minutes of incineration. Among them, the main combustion chamber temperature refers to the temperature in the main combustion chamber of the incinerator, which is a key parameter to ensure the full combustion of waste. The average temperature refers to the overall average air temperature in the incinerator, which helps to monitor the uniformity of heat distribution in the furnace. The secondary combustion chamber wind speed refers to the speed of air flow used to assist combustion in the secondary combustion chamber of the incinerator. The top wind speed refers to the air flow rate controlled from the top of the furnace, which affects the sufficiency of combustion and temperature distribution. The left wind speed and the right wind speed refer to the air flow rate controlled from the left and right sides of the incinerator, which helps to maintain the stability and uniform combustion of the flame in the furnace. The left smoke elimination wind speed and the right smoke elimination wind speed refer to the air flow rate controlled from the left and right sides of the incinerator, which helps to reduce the emission of harmful gases and particulate matter. The burner status indicates the on / off status of the burner. Turning on the burner can assist in starting and maintaining the combustion process. The blower status indicates the on / off status of the blower. The blower is used to provide the necessary oxygen to support combustion. The main oil valve status indicates the on / off status of the main oil valve, which controls the fuel supply and affects the intensity and stability of combustion. The burning minute refers to a specific time point in the burning process, which is used to track and adjust control parameters to ensure that specific operating conditions are met throughout the burning process. The above-mentioned various burning control parameters to be optimized can jointly affect the parameters of the burning condition and efficiency at the same time point. Among them, the burner status includes burner on and burner off, the blower status includes blower on and blower off, and the main oil valve status includes main oil valve on and main oil valve off.

[0028] In step 102, an initial population consisting of a number of individuals is randomly generated based on the incineration control parameters to be optimized.

[0029] Specifically, the incineration control parameters to be optimized have a predefined value range or fluctuation range. A series of possible incineration control parameter combinations are randomly generated according to the value range or fluctuation range of the incineration control parameters to be optimized, and the incineration control parameter combinations are individuals in the initial population.

[0030] In step 103, the fitness of each individual in the initial population is calculated according to the fitness algorithm.

[0031] In one embodiment of the present disclosure, the fitness algorithm is an algorithm used to evaluate the performance or quality of each individual. For each individual in the initial population, the fitness algorithm is used to combine the incineration control parameters to calculate a fitness value. The fitness value is a quantitative representation of the individual performance, usually a numerical value, which is used to guide the retention and elimination of individuals in the selection process of the genetic algorithm.

[0032] In step 104, based on the initial population and fitness, an iterative optimization calculation is performed using a genetic algorithm until a preset iteration termination condition is reached.

[0033] In one embodiment of the present disclosure, the initial population and fitness are used as the initial input variables of the genetic algorithm, and the genetic algorithm is used to perform iterative optimization calculations. The genetic algorithm gradually selects individuals with higher fitness by simulating natural selection and genetic mechanisms, and generates a new population. This process continues until the preset iteration termination condition is reached, and the algorithm stops iterating.

[0034] In step 105, the optimized incineration control parameters are output.

[0035] In one embodiment of the present disclosure, when the iteration termination condition is met, the algorithm stops iterating and outputs the individual with the highest fitness as the optimal solution. This optimal solution represents the optimized incineration control parameters and can be used to guide the actual incineration process.

[0036] In summary, according to the technical solution provided by the embodiment of the present disclosure, by determining the incineration control parameters to be optimized that meet the preset conditions, the initial population is randomly generated using the incineration control parameters to be optimized, and the fitness corresponding to several individuals in the initial population is calculated, and the genetic algorithm is used for iteration to output the optimized incineration control parameters. Compared with the traditional manual control optimization method, the incineration optimization method using the genetic algorithm can automatically and quickly find the optimal or near-optimal incineration control parameters. This method is not limited by human factors, and can efficiently search for the optimized incineration control parameters, thereby achieving precise control of the incineration process, and can solve the problems existing in the traditional manual control optimization method, and achieve an efficient and environmentally friendly incineration process.

[0037] In one embodiment of the present disclosure, determining the incineration control parameters to be optimized that meet predetermined conditions includes: obtaining a first incineration control parameter set including incineration control parameters corresponding to multiple experimental samples; based on the first incineration control parameter set, obtaining multiple second incineration control parameter sets, each second incineration control parameter set including the incineration control parameters corresponding to all experimental samples at the same time point; obtaining the first pollutant emission concentration and the first fuel instantaneous flow rate corresponding to each first incineration control parameter; determining the incineration control parameters of the experimental samples in each second incineration control parameter set whose first pollutant emission concentration is lower than the pollutant emission concentration limit and whose first fuel instantaneous flow rate is lower than the median of all first fuel instantaneous flow rates in the second incineration control parameter set as the incineration control parameters to be optimized.

[0038] Specifically, first, the combustion control parameters of multiple experimental samples are collected, and these parameters constitute the first combustion control parameter set. Among them, the combustion control parameters of each experimental sample include one or more of the main combustion chamber temperature, average temperature, secondary combustion chamber wind speed, top wind speed, left wind speed, right wind speed, left smoke elimination wind speed, right smoke elimination wind speed, burner state, blower state, main oil valve state, and the number of minutes of combustion. Then, the first combustion control parameter set is divided according to the time point and each experimental sample to obtain multiple second combustion control parameter sets, and each second combustion control parameter set includes the first combustion control parameters corresponding to all experimental samples at the same time point. Pollutant emission concentration refers to the concentration of pollutants released during the combustion process, and fuel instantaneous flow rate refers to the amount of fuel entering the incinerator at a certain moment. According to the first combustion control parameter corresponding to each experimental sample in each second combustion control parameter set, the first pollutant emission concentration and the first fuel instantaneous flow rate corresponding to the first combustion control parameter are determined. By comparing the first pollutant emission concentration and the pollutant emission concentration limit in each second incineration control parameter set, as well as the first fuel instantaneous flow rate in each second incineration control parameter set, the first incineration control parameter of the experimental sample in each second incineration control parameter set that simultaneously satisfies the conditions that the first pollutant emission concentration is lower than the pollutant emission concentration limit and the first fuel instantaneous flow rate is lower than the median value of all first fuel instantaneous flow rates in the same second incineration control parameter set is determined as the incineration control parameter to be optimized. Here, the incineration control parameter to be optimized that satisfies the above conditions can ensure incineration efficiency and reduce pollutant emissions while not over-consuming fuel. Through the above method, the optimal incineration control parameters can be found to achieve a more efficient and environmentally friendly incineration process.

[0039] In one embodiment of the present disclosure, the first pollutant emission concentration includes at least one or more of carbon monoxide CO emission concentration, nitrogen dioxide NO2 emission concentration and sulfur dioxide SO2 emission concentration.

[0040] By setting the above preset conditions, strict requirements are imposed on the operation of the incineration control parameters, aiming to ensure that the incineration process reaches the best state in terms of environmental protection and fuel economy.

[0041] Furthermore, the incineration control parameters to be optimized that meet the predetermined conditions can also be determined by the following method: Obtaining a first incineration control parameter set including incineration control parameters corresponding to a plurality of experimental samples, including: The sensor data and infrared temperature measurement data of the incinerator are collected on a daily basis. The sensor data include: main combustion chamber temperature, average temperature, secondary combustion chamber wind speed, top wind speed, left wind speed, right wind speed, left smoke elimination wind speed, right smoke elimination wind speed, burner status, blower status, main oil valve status; infrared temperature measurement data include: temperature values ​​of various points in the incinerator chamber; The sensor data and the infrared temperature measurement data are aligned and downsampled in units of minutes to obtain a first incineration control parameter set including incineration control parameters corresponding to a plurality of experimental samples.

[0042] Specifically, the sensor data of the incinerator is collected, and the incinerator refers to an incinerator used for cremation of remains. The sensor data includes the temperature of the main combustion chamber, the average temperature, the wind speed of the secondary combustion chamber, the top wind speed, the left and right side wind speeds, the smoke elimination wind speed, and the status of the burner, the blower, and the main oil valve. These data are key factors affecting the incineration efficiency and pollutant emissions. In addition, it is necessary to collect the temperature values ​​of each point in the incineration furnace through an infrared thermometer. Infrared temperature measurement data can more accurately reflect the changes in combustion temperature. Since the sensor data and the infrared temperature measurement data come from different storage devices and there is a time inconsistency, it is necessary to align these two types of data in minutes. This means that the data within each minute is sorted to ensure the synchronization of the data. After alignment, in order to improve the robustness of the data, especially the resistance to outliers, the median downsampling method is adopted. This method can reduce the impact of outliers more than the average value because it is not affected by extreme values. Thus, a first incineration control parameter set including incineration control parameters corresponding to multiple experimental samples is obtained. Each experimental sample corresponds to a body.

[0043] In one embodiment of the present invention, the minimum value points of the incineration temperature curve can be extracted by analyzing the average temperature of the incinerator. These minimum value points can help distinguish the incineration stages of different bodies (i.e., the combustion process corresponding to each experimental sample). Through the above steps, the incineration stages corresponding to different bodies can be divided. The incineration control parameters corresponding to the incineration stage of each body are divided according to the number of minutes of incineration of each body (i.e., the number of minutes of incineration mentioned above), and multiple groups of incineration control parameters can be obtained. A group of incineration control parameters corresponding to the same time point (such as the number of minutes of incineration) in each incineration stage (i.e., the first incineration control parameters corresponding to each experimental sample mentioned above) are extracted from the multiple groups of incineration control parameters to form a second incineration control parameter set, and then the first incineration control parameter set includes multiple second incineration control parameter sets. Each second incineration control parameter set is a set composed of the incineration control parameters (i.e., the first incineration control parameters) corresponding to different bodies at the same time point. Each first incineration control parameter in the first incineration control parameter set is used as input data and input into the XGBoost classification model. The XGBoost classification model will predict the energy consumption and pollutant emissions of the incineration process based on the input control parameters, and classify them into four categories: "low energy consumption-low pollution", "low energy consumption-high pollution", "high energy consumption-low pollution" or "high energy consumption-high pollution". Through this classification model, the incineration control parameters to be optimized that meet the preset conditions (i.e. "low energy consumption-low pollution") can be screened out.

[0044] Exemplarily, an XGBoost classification model is constructed in the following way: The first combustion control parameters include the main combustion chamber temperature, average temperature, secondary combustion chamber wind speed, top wind speed, left wind speed, right wind speed, left smoke elimination wind speed, right smoke elimination wind speed, burner status, blower status, main oil valve status, and the number of minutes of combustion. Each first combustion control parameter in the first combustion control parameter set is taken as a sample. There are multiple samples corresponding to the combustion control parameters in different time periods corresponding to multiple combustion stages. The multiple samples are taken as input data and input into the CO emission concentration prediction model, NO2 emission concentration prediction model, SO2 emission concentration prediction model, and fuel instantaneous flow prediction model to calculate the CO emission concentration, NO2 emission concentration, SO2 emission concentration, and fuel instantaneous flow of each sample. For each pollutant emission concentration (CO emission concentration, NO2 emission concentration, SO2 emission concentration), refer to the national standard GB13801-2015, and convert the pollutant emission concentration of each sample into the atmospheric pollutant baseline oxygen content emission concentration of 11% according to the following formula:

[0045] Where: c represents the baseline oxygen content of atmospheric pollutants, in mg / m³; O sIndicates the measured oxygen concentration in dry flue gas, in %; c s It indicates the measured concentration of air pollutant emissions in mg / m³.

[0046] Each sample is compared with the emission limit to determine whether the emission meets the standard. If the pollutant concentration is lower than the emission limit (i.e. the pollutant emission limit), the sample is considered to be "low pollution"; as long as one pollutant concentration exceeds the standard, the sample is considered to be "high pollution". The emission limits corresponding to the pollutants NO2, SO2 and CO considered are 150 mg / m³, 30 mg / m³ and 200 mg / m³ respectively.

[0047] Considering that the normal energy consumption at different stages of the incineration of remains is obviously different, to determine whether the current sample has high energy consumption, multiple samples are grouped according to the number of minutes of incineration, the median of the instantaneous flow of each group (i.e., the second incineration control parameter set) is calculated, and the instantaneous flow of each sample is compared with the median of the group to which it belongs. If it is higher than the median, the sample is marked as "high energy consumption", otherwise it is "low energy consumption".

[0048] Finally, all samples were divided into four categories: "low energy consumption-low pollution", "low energy consumption-high pollution", "high energy consumption-low pollution", and "high energy consumption-high pollution". The XGBoost classifier was used to build an XGBoost classification model, which judged which of the four categories the sample belonged to based on the input controllable variables. Thus, the incineration control parameters corresponding to the "low energy consumption-low pollution" samples were used as the incineration control parameters to be optimized.

[0049] In summary, according to the technical solution provided by the embodiment of the present disclosure, by analyzing the temperature curve of the main combustion chamber, accurately dividing the incineration stage, and eliminating abnormal data, the accuracy and representativeness of the data are ensured, providing a reliable basis for subsequent parameter optimization. And using the XGBoost classification model, the energy consumption and pollutant emissions of the incineration process are predicted according to the incineration control parameters, and the incineration control parameters to be optimized that meet the "low energy consumption-low pollution" conditions are screened out, thereby realizing the intelligent optimization of the incineration process. The optimized incineration control parameters significantly reduce energy consumption and pollutant emissions, improve the combustion efficiency and environmental protection performance of the incinerator, and meet the requirements of sustainable development. And the selection of "low energy consumption-low pollution" data as the incineration control parameters to be optimized can effectively optimize the control parameters in the incineration process to improve the incineration efficiency, reduce pollutant emissions, and ensure the stable operation of the incineration equipment.

[0050] Furthermore, each individual represents a set of third burning control parameters, and the fitness of each individual in the initial population is calculated according to the fitness algorithm, including: Obtain the third pollutant emission concentration corresponding to each set of third incineration control parameters: input each set of third incineration control parameters into the pollutant emission concentration prediction model, and obtain the pollutant emission concentration output by the third pollutant emission concentration prediction model; obtain the third fuel instantaneous flow rate corresponding to each set of third incineration control parameters: input each set of third incineration control parameters into the fuel instantaneous flow rate prediction model, and obtain the third fuel instantaneous flow rate output by the third fuel instantaneous flow rate prediction model; determine the fitness based on the third pollutant emission concentration and the third fuel instantaneous flow rate.

[0051] Specifically, after determining the fluctuation range of the incineration control parameter to be optimized, a series of possible incineration control parameter combinations are randomly generated within the fluctuation range of the incineration control parameter to be optimized. Each incineration control parameter combination forms an individual, namely the third incineration control parameter, and each individual represents a possible control strategy in the incineration process. Each group of third incineration control parameters is used as the initial input data of the XGBoost algorithm, and a pollutant emission concentration prediction model is established with the pollutant emission concentration as the target, and the third pollutant concentration corresponding to each group of third incineration control parameters is obtained. Each group of third incineration control parameters is used as the initial input data of the XGBoost algorithm, and a fuel instantaneous flow rate prediction model is established with the fuel instantaneous flow rate as the target, and the third fuel instantaneous flow rate corresponding to each group of third incineration control parameters is obtained. The fitness of each individual in the initial population is calculated using the third pollutant emission concentration and the third fuel instantaneous flow rate corresponding to each group of third incineration control parameters.

[0052] Exemplarily, the main combustion chamber temperature, average temperature, secondary combustion chamber wind speed, top wind speed, left wind speed, right wind speed, left smoke elimination wind speed, right smoke elimination wind speed, burner state, blower state, main oil valve state and burning minutes are used as the third burning control parameter. The third pollutant emission concentration corresponding to the third burning control parameter includes CO emission concentration, NO2 emission concentration, and SO2 emission concentration. With the goal of obtaining the CO emission concentration, NO2 emission concentration, SO2 emission concentration and the third fuel instantaneous flow rate corresponding to the third burning control parameter, the CO emission concentration prediction model, the NO2 emission concentration prediction model, the SO2 emission concentration prediction model and the fuel instantaneous flow rate prediction model are respectively constructed through the XGBoost algorithm. The CO emission concentration, NO2 emission concentration, SO2 emission concentration and the third fuel instantaneous flow rate are predicted. The fitness corresponding to the third burning control parameter is calculated according to the CO emission concentration, NO2 emission concentration, SO2 emission concentration and the third fuel instantaneous flow rate.

[0053] In summary, according to the technical solution provided by the embodiment of the present disclosure, by using the XGBoost algorithm to construct a pollutant emission concentration prediction model and a fuel instantaneous flow prediction model, it is possible to quickly predict key parameters in the incineration process. This avoids a large number of experiments and adjustments during the actual incineration process, thereby significantly improving the optimization efficiency. And the prediction model can accurately reflect the relationship between the incineration control parameters and the pollutant emission concentration. By optimizing the incineration control parameters, the emission concentration of pollutants such as CO, NO2, SO2, etc. can be effectively reduced, meeting environmental protection requirements and reducing pollution to the environment.

[0054] At the same time, the prediction model can also accurately predict the instantaneous flow rate of fuel. By optimizing the incineration control parameters, fuel consumption can be reduced and energy utilization efficiency can be improved while ensuring the incineration effect. By establishing a prediction model, a large amount of incineration process data can be accumulated. These data can provide strong data support for subsequent incineration process optimization, equipment maintenance and management. It can achieve precise control and optimization of the incineration process, improve incineration efficiency, reduce pollutant emissions and fuel consumption, and has important technical effects and application value.

[0055] Furthermore, the fitness satisfies the following relationship:

[0056] Among them, fitness is the fitness, represents the CO emission concentration, Indicates the SO2 emission concentration, Indicates NO2 emission concentration, Flow indicates the instantaneous flow rate of the third fuel, , and Respectively represent the weight values ​​of CO emission concentration, SO2 emission concentration and NO2 emission concentration, Indicates the weight value of the third fuel instantaneous flow rate.

[0057] Specifically, the CO emission concentration, SO2 emission concentration, NO2 emission concentration, and the third fuel instantaneous flow rate corresponding to the third incineration control parameter are respectively substituted into the above fitness calculation formula, and the setting , , and are all 1, and the fitness corresponding to the third incineration control parameter can be calculated.

[0058] In summary, according to the technical solution provided by the embodiment of the present disclosure, the present disclosure uses CO emission concentration, SO2 emission concentration, NO2 emission concentration and instantaneous fuel flow rate to construct a fitness calculation formula, which can comprehensively evaluate the environmental performance and economic benefits under different control parameter combinations during the incineration process. The fitness calculation formula can be used to guide the optimization of the incineration process. By continuously adjusting the incineration control parameters and calculating the fitness value under each incineration control parameter combination, the optimal or approximately optimal parameter combination can be found to achieve the best balance between environmental performance and economic benefits.

[0059] Furthermore, based on the initial population and fitness, a genetic algorithm is used to perform iterative optimization calculations until a preset iterative termination condition is reached, including: The initial population and fitness are used as initial parameters and input into the genetic algorithm for calculation; The initial population is iteratively optimized, where each iteration generates a new population based on selection, crossover, and mutation.

[0060] In one embodiment of the present disclosure, the initial population and the fitness of each individual are input into the genetic algorithm as initial parameters for calculation. The initial population is iteratively optimized. In this stage, each iteration sequentially performs the three core operations of selection, crossover and mutation to generate a new population.

[0061] Specifically, according to the fitness value of the individual, a certain number of excellent individuals are selected from the current population as the parents of the new population. The selected parent individuals are paired and new offspring individuals are generated by exchanging some of their genes. The newly generated offspring individuals are subjected to random small genetic changes to introduce new genetic information. By repeating the above three operations, the genetic algorithm will generate a series of new populations and gradually approach the optimal solution to the problem. When the number of iterative optimization processes reaches the preset iterative termination condition, the genetic algorithm stops iterating.

[0062] Exemplarily, in the selection stage, a roulette selection strategy is adopted, which allocates selection probabilities according to the fitness values ​​of individuals, so that individuals with higher fitness values ​​have a greater chance of being selected as parents. In each selection process, two individuals are randomly selected as a pair of parents for subsequent crossover operations. In the crossover stage, a single-point crossover operation is adopted. For each pair of selected parents, crossover is performed at a random position of a continuous control variable, that is, part of the control variables of the two parents before and after the position are exchanged to generate two new individuals. This step aims to generate offspring with potentially better performance by combining excellent genes from different individuals. In order to increase the diversity of the population and prevent the algorithm from falling into a local optimal solution, the generated new individuals are mutated in the mutation stage. At a certain mutation rate, each control variable has the opportunity to be randomly disturbed. For continuous variables, a random fluctuation value is added to its current value to introduce new genetic information. By continuously repeating the above selection, crossover and mutation operations, the genetic algorithm will generate a series of new populations and gradually approach the optimal solution of the problem. When the number of iterations reaches the preset termination condition, the genetic algorithm stops iterating.

[0063] In summary, according to the technical solution provided by the embodiment of the present disclosure, the present disclosure can efficiently search for the optimal or approximately optimal combination of incineration control parameters through iterative optimization of the genetic algorithm. This optimization process avoids the tediousness and time-consuming nature of the traditional trial and error method, and significantly improves the optimization efficiency.

[0064] Furthermore, the preset iteration termination condition includes: the number of iterations meets a preset threshold.

[0065] Specifically, after each iteration, the genetic algorithm will check whether the iteration counter has reached the preset threshold. If it has, the algorithm will stop iterating and output the currently found optimal solution (i.e., the optimized incineration control parameters). By setting an upper limit on the number of iterations, it is possible to avoid the algorithm from performing a large number of invalid calculations when it is close to the optimal solution, thereby saving computing resources and time. For example, the preset threshold is set to 100 times, that is, the maximum number of iterations is 100, ensuring that the iteration is completed within 1-2 seconds.

[0066] In summary, according to the technical solution provided by the embodiments of the present disclosure, by setting an upper limit on the number of iterations, the genetic algorithm can stop iterating when a preset threshold is reached, thereby avoiding a large number of invalid calculations, which helps to save computing resources.

[0067] Furthermore, each of the above-mentioned incineration optimization methods improved based on genetic algorithms also includes: outputting pollutant emission concentration and instantaneous fuel flow rate corresponding to the optimized incineration control parameters.

[0068] Specifically, after optimizing the incineration control parameters using the genetic algorithm, the algorithm not only finds the optimal or approximately optimal combination of incineration control parameters, but also calculates the corresponding pollutant emission concentration and instantaneous fuel flow rate based on these optimized parameters. Specifically, since the design of the fitness function takes into account the pollutant emission concentration and instantaneous fuel flow rate, when determining the optimized incineration control parameters, the pollutant emission concentration and instantaneous fuel flow rate corresponding to the optimized incineration control parameters can also be determined. The optimized incineration control parameters can not only improve the incineration efficiency, but also ensure that the pollutant emissions are below the limit, while reducing fuel consumption.

[0069] In summary, according to the technical solution provided by the embodiment of the present disclosure, while outputting the optimized incineration control parameters, the pollutant concentration and instantaneous fuel flow rate corresponding to the optimized incineration control parameters can also be output, which helps operators understand the impact of parameter changes on the incineration process and make more informed decisions, with good flexibility and adaptability. In addition, outputting the optimized pollutant emission concentration and instantaneous fuel flow rate can also be used as one of the important indicators for evaluating algorithm performance. By comparing with experimental data or historical data, the optimization effect and accuracy of the algorithm can be evaluated, providing feedback for further improvement and optimization of the algorithm.

[0070] In order to illustrate the technical solution provided by the present disclosure in detail, the incineration optimization method improved based on genetic algorithm of the present disclosure is described in detail with reference to the following embodiments.

[0071] 1. Data collection and preprocessing: (1) Data source

[0072] The experimental data used mainly come from the incineration equipment of Dongjiao Funeral Home in Beijing, China. Dongjiao Funeral Home has multiple incinerators, and the experimental data is mainly collected based on Furnace No. 8. The cremation combustion related data collected from Furnace No. 8 is mainly divided into two parts: sensor data and infrared temperature measurement data.

[0073] (2) Sensor data

[0074] Sensor data is the data measured by the sensors installed in the incinerator, such as the main combustion chamber temperature, secondary combustion chamber wind speed, top wind speed, left wind speed, right wind speed, left smoke elimination wind speed, right smoke elimination wind speed, main oil valve status, blower status, burner status, etc. These data are obtained from the equipment sensors and stored in seconds.

[0075] (3) Infrared temperature measurement data

[0076] Infrared temperature measurement data is the temperature value of each point in the incinerator measured by an infrared thermometer. Compared with traditional sensors, infrared thermometers can more accurately reflect the changes in combustion temperature, especially in the main combustion chamber and the secondary combustion chamber, with higher measurement accuracy.

[0077] (4) Data alignment and downsampling

[0078] Since the sensor data and infrared temperature measurement data are measured by two instruments respectively, they are located in different storages. Although both are stored in seconds as the minimum unit, due to the complexity of the sampling process, some data are missing, that is, the time interval of the measurement data is not strictly 1 second. In addition, due to accuracy issues, the sensor measurement data sometimes fluctuates greatly. Through observation, it is found that in general, the pressure changes continuously in the measurement with the granularity of seconds, and the number of abnormal values ​​is small and discontinuous. Based on this consideration, the alignment method adopted is to align the two files in minutes, and downsample the data within minutes by taking the median. Compared with taking the average, this method has better robustness to abnormal values, and the data changes with the granularity of minutes are more obvious than the data changes with the granularity of seconds.

[0079] (5) Division of the body incineration process

[0080] During the incineration process, the burning time and temperature changes of each body have regularity. By extracting the minimum value point of the incineration temperature curve, the incineration data within a day is divided into multiple incineration stages. To ensure the accuracy of the division, the intervals with a burning time of less than 30 minutes and a temperature below 300°C are eliminated to ensure that the data can accurately reflect the actual incineration process. Figure 2 This corresponds to the average temperature change of the No. 8 incinerator in Dongjiao Funeral Home in one day. From the figure, we can see that the temperature of each body first increases, then maintains stability, and finally decreases during the combustion process. The red inverted triangle in the figure represents the minimum point of the curve found. It can be seen that the bodies can be well divided from each other, but due to the fluctuation of the average temperature, there are some minimum points that are dense in time. Figure 3 This is a line graph of the average temperature-time of the first body burning on March 16, 2024. This data is obtained by the rule method of dividing the body into parts. It can be seen that the burning time of the body is about 1 hour, and the temperature at the beginning and end of the burning is above 300, which is consistent with the general burning characteristics of the body.

[0081] 2. Smoke concentration prediction and fuel instantaneous flow prediction: During the incineration process, the control of flue gas concentration (i.e. the above-mentioned pollutant concentration) is an important part of optimizing the incineration process. Here, the three main pollutants, NO2, SO2 and CO, and the instantaneous flow rate of fuel are taken as the targets, and the flue gas concentration and the instantaneous flow rate of fuel are predicted through the XGBoost model (i.e., the CO emission concentration prediction model, NO2 emission concentration prediction model, SO2 emission concentration prediction model and fuel instantaneous flow rate prediction model are constructed).

[0082] (1) Input characteristics (i.e. incineration control parameters)

[0083] Input features include main combustion chamber temperature, average temperature, top wind speed, secondary combustion chamber wind speed, left wind speed, right wind speed, left smoke elimination wind speed, right smoke elimination wind speed, burner on, blower on, main oil valve on, and the current number of minutes of cremation (i.e., the number of minutes of cremation for each body corresponding to the incineration control parameters). These features can reflect the current combustion status, are key factors affecting the flue gas emission concentration, and are controllable in actual industrial production.

[0084] (2) Model performance

[0085] Select the XGBoost model and use the data from furnace 8 for model training. After optimization, the model performs well on the test set. Figure 4 As shown. For SO2, the model's prediction R² is 0.43; for NO2, R² is 0.76; for CO, R² is 0.61, and for instantaneous flow, R² is as high as 0.98. The model prediction results show that the XGBoost model is effective in predicting pollutant concentration and energy consumption flow.

[0086] In one embodiment of the present disclosure, in a genetic algorithm, fitness calculation is used to evaluate the pros and cons of each individual. For the incineration optimization problem, pollutant emission concentrations (such as CO emission concentrations, SO2 emission concentrations, and NO2 emission concentrations) and instantaneous fuel flow rates are considered in the fitness calculation process. Therefore, when the genetic algorithm is optimizing, it is not only looking for the best incineration control parameters, but also looking for the corresponding pollutant emission concentrations and instantaneous fuel flow rates under these parameters. Therefore, when the algorithm finds the optimal incineration control parameters, it can also simultaneously obtain the corresponding pollutant emission concentrations under these parameters.

[0087] 3. Incineration optimization control based on genetic algorithm, refer to Figure 5 As shown: (1) XGBoost model for classifying energy consumption (i.e., instantaneous fuel flow) and pollutant concentration (i.e., the aforementioned XGBoost classification model)

[0088] The data for training the classification model also comes from the cremation monitoring log of Dongjiao No. 8 furnace. The outliers of the instantaneous flow rate and the data related to the first body cremated every day were removed because the boiler needs to be preheated when the first body is cremated every day. After removing useless and practically uncontrollable fields and conventional standardization operations, these controllable variables are finally left as the input of the model: main combustion chamber temperature, top wind speed, secondary combustion chamber wind speed, left wind speed, right wind speed, left smoke elimination wind speed, right smoke elimination wind speed, burner on, blower on, main oil valve on, average temperature, and the number of minutes of incineration (that is, the incineration control parameters corresponding to the number of minutes of incineration of a body).

[0089] Since it is impossible to obtain more specific personal information such as the weight, age, cause of death, etc. of each body, it is temporarily assumed that this information will not have a significant impact on subsequent experiments. Considering that the normal energy consumption in different stages of cremation is obviously not the same, in order to determine whether the current sample has high energy consumption, the data is first grouped according to the number of minutes of incineration, the median of the instantaneous flow rate of each group is calculated, and the instantaneous flow rate of each sample is compared with the median of the group to which it belongs. If it is higher than the median, the sample is marked as "high energy consumption", otherwise it is "low energy consumption". For the emission of pollutant concentrations, refer to the national standard GB13801-2015, and convert the measured emission concentrations of various atmospheric pollutants of each sample into the baseline oxygen content emission concentration of atmospheric pollutants with a baseline oxygen content of 11% according to the following formula:

[0090] Where: c represents the baseline oxygen content of atmospheric pollutants, in mg / m³; O S Indicates the measured oxygen concentration in dry flue gas, in %; c s It indicates the measured concentration of air pollutant emissions in mg / m³.

[0091] Each sample is then compared with the emission limit to determine whether the emission meets the standard. As long as one pollutant exceeds the standard, the sample is considered "high pollution". If the pollutant emission concentration is lower than the emission limit (i.e., the pollutant emission limit), the sample is considered "low pollution". The emission limits corresponding to the pollutants NO2, SO2, and CO considered are 150mg / m³, 30mg / m³, and 200mg / m³, respectively. In the end, all samples were divided into four categories: "low energy consumption-low pollution", "low energy consumption-high pollution", "high energy consumption-low pollution", and "high energy consumption-high pollution". Then the XGBoost classifier was used to build the XGBoost classification model, which determines which of the four categories the sample belongs to based on the input controllable variables. Model training uses 80% of the data as the training set and 20% of the data as the test set. The initialization parameters of the XGBoost classifier are all default configurations.

[0092] The results of the classification model on the test set are shown in Table 1 below. The performance is relatively good, with an overall accuracy of 88%, so it can be used for subsequent analysis.

[0093] Table 1: Precision, recall, and F1-score of the classification model

[0094] Among them, accuracy represents the accuracy of the classification model.

[0095] In addition, the results of the feature importance analysis of the model are shown in Table 2 below, which shows the impact of different control variables on energy consumption and pollutant classification. Among them, the importance of features such as burner status, left wind speed, and blower status is relatively high, indicating that these parameters have a significant impact on the incineration process.

[0096] Table 2: Feature importance ranking

[0097] (2) Genetic algorithm input

[0098] The input control variables of the genetic algorithm (i.e., the incineration control parameters) are the same as those of the four-classification model. In addition, the numerical fluctuation range of each control variable in the existing training data is calculated in order to determine the range of random changes of each control variable during the initialization population and mutation stage of the genetic algorithm. Taking into account that it is impossible to make large adjustments to various continuous control variables such as the main combustion chamber temperature in actual production, the fluctuation ranges of the "main combustion chamber temperature" and "average temperature" are ultimately limited to 10 and 30 respectively, and the fluctuation range of other continuous control variables such as "headwind wind speed" is set to 5. The input of the final algorithm is based on the various control variables (i.e. Figure 5 Controllable variables: main combustion chamber temperature, top wind speed, secondary combustion chamber wind speed, left wind speed, right wind speed, left smoke elimination wind speed, right smoke elimination wind speed, burner on, blower on, main oil valve on, average temperature, burning minutes) and variables (i.e. Figure 5 The measured variables, that is, the pollutant emission concentration and instantaneous fuel flow rate corresponding to the control variables) and the allowable fluctuation range corresponding to each control variable and variable. In the initialization process, an initial population consisting of several individuals is generated according to the initial input value (that is, the above control variables) within the allowable fluctuation range, and each individual represents a combination of control variables.

[0099] (3) Genetic algorithm hyperparameter setting

[0100] In actual production, the time for a single genetic algorithm run should be controlled within 1-2 seconds, so that real-time control of the combustion process can be practical. Through multiple experiments, the maximum number of iterations of the genetic algorithm was finally set to 100, the population size was 10, the mutation rate was 0.1, and the crossover rate was 0.8, so that the time for a single genetic algorithm run can be maintained at about 1.5 seconds.

[0101] (4) Fitness function

[0102] In order to evaluate the quality of the control variable (i.e. the controllable variables mentioned above), a fitness function is designed. The fitness function combines the XGBoost classification model and the regression model (i.e. the pollutant emission concentration prediction model and the fuel instantaneous flow prediction model) to determine the energy consumption and pollutant emissions of each control variable combination (i.e. the controllable variables mentioned above, i.e. each individual): 1. Use the XGBoost classification model to predict whether the current combination (i.e. the controllable variables mentioned above) belongs to “low energy consumption-low pollution”.

[0103] 2. Use the XGBoost regression model to predict the emission concentration of pollutants (NO2, SO2, CO) and the instantaneous fuel flow rate.

[0104] 3. The calculation formula of fitness is:

[0105] Among them, fitness is the fitness, represents the CO emission concentration, Indicates the SO2 emission concentration, Indicates NO2 emission concentration, Flow indicates instantaneous fuel flow rate, , and Respectively represent the weight values ​​of CO emission concentration, SO2 emission concentration and NO2 emission concentration, Indicates the weight value of the instantaneous fuel flow rate.

[0106] All weights are set to 1, indicating the same importance.

[0107] After initialization, the iteration is performed. During the iteration process, it is determined whether the maximum number of iterations has been reached. If so, the global best fitness individual (i.e., the individual with the highest fitness) is output. Otherwise, the pollutant concentration prediction model and the fuel instantaneous flow prediction model, fitness calculation, and XGBoost classification model are used to compare the fitness and continue to iterate.

[0108] (5) Iteration process of genetic algorithm

[0109] In each iteration of the genetic algorithm, the population undergoes selection, crossover, and mutation operations. The description of each operation is as follows: Selection: Roulette selection strategy is adopted. Specifically, the higher the fitness value of an individual, the greater the probability of being selected. Two individuals are selected as parents each time to perform crossover operation.

[0110] Crossover: In order to generate a new generation of individuals, a single-point crossover operation is used. Each pair of parents crosses over at a random position of a continuous control variable, exchanging some of the control variables of the two parents to generate two new individuals.

[0111] Mutation: In order to increase diversity and prevent the algorithm from falling into a local optimal solution, mutation operations are performed on the generated new individuals. At a certain mutation rate, each control variable will be randomly perturbed. For continuous variables, random fluctuations are performed within a range based on their current values.

[0112] (6) Population update and iteration termination conditions

[0113] Each iteration of the genetic algorithm generates a new population based on selection, crossover, and mutation. The size of the new population remains constant, and after each generation, the worst performing individuals of the previous generation are eliminated, and individuals with higher fitness are retained. In order to complete within 1-2 seconds, the maximum number of iterations is set to 100.

[0114] (7) Output the optimal solution

[0115] After the maximum number of iterations is reached, the algorithm outputs the individual with the highest fitness found in the iteration process as the optimal solution. The optimal solution includes the optimized control variable value (i.e., incineration control parameter) combination, and outputs the corresponding pollutant and flow prediction values ​​for reference. Through this optimal control variable combination, the optimization of energy consumption and pollutant emissions during the combustion process can be achieved.

[0116] (8) Comparison with random sampling

[0117] In order to illustrate the effect of this embodiment, a simple comparative experiment is designed to compare the optimization strategies of genetic algorithm (GA) and random sampling. Under the same control parameter settings and number of iterations, the genetic algorithm and random sampling are run 100 times respectively, and the fitness value of the best individual found in each iteration is recorded. The individual can be classified as "low energy consumption-low pollution" when input into the four-classification model.

[0118] like Figure 6As shown in the figure, the genetic algorithm can continuously find individuals with higher fitness through multiple rounds of iterations, thereby achieving better energy consumption and pollutant emission control, while random sampling cannot achieve this progressive optimization effect. This shows that the genetic algorithm has significant advantages over random sampling and can effectively optimize the control parameter combination in the incineration process. Through the control parameter combination optimized by the genetic algorithm, the emission concentrations of NO2, SO2 and CO can also be reduced to a certain extent, and fuel consumption is effectively controlled.

[0119] Figure 7 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 7 As shown, the electronic device may include: a processor 710, a communications interface 720, a memory 730 and a communication bus 740, wherein the processor 710, the communications interface 720 and the memory 730 communicate with each other through the communication bus 740. The processor 710 may call the logic instructions in the memory 730 to execute the above-mentioned improved incineration optimization method based on the genetic algorithm.

[0120] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0121] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the improved incineration optimization method based on genetic algorithm provided by the above methods.

[0122] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0123] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present disclosure.

Claims

1. An improved incineration optimization method based on genetic algorithm, characterized in that: include: Determine the incineration control parameters to be optimized that meet the preset conditions; Based on the incineration control parameters to be optimized, randomly generating an initial population consisting of a number of individuals; Calculating the fitness of each individual in the initial population according to a fitness algorithm; Based on the initial population and the fitness, an iterative optimization calculation is performed using a genetic algorithm until a preset iterative termination condition is reached; Output optimized incineration control parameters.

2. The incineration optimization method based on genetic algorithm improvement according to claim 1 is characterized in that: The step of determining the incineration control parameters to be optimized that meet the predetermined conditions includes: Acquire a first incineration control parameter set including incineration control parameters corresponding to a plurality of experimental samples; Based on the first incineration control parameter set, a plurality of second incineration control parameter sets are obtained, each of the second incineration control parameter sets including the first incineration control parameters corresponding to all the experimental samples at the same time point; Obtaining a first pollutant emission concentration and a first fuel instantaneous flow rate corresponding to each of the first incineration control parameters; Determine the first incineration control parameter of the experimental sample in each of the second incineration control parameter sets in which the first pollutant emission concentration is lower than the pollutant emission concentration limit and the first fuel instantaneous flow rate is lower than the median value of all the first fuel instantaneous flow rates in the second incineration control parameter set as the incineration control parameter to be optimized.

3. The incineration optimization method based on genetic algorithm improvement according to claim 2 is characterized in that: The incineration control parameters to be optimized include at least: main combustion chamber temperature, average temperature, secondary combustion chamber wind speed, top wind speed, left wind speed, right wind speed, left smoke elimination wind speed, right smoke elimination wind speed, burner status, blower status, main oil valve status and one or more of the incineration minutes; the first pollutant emission concentration includes at least one or more of carbon monoxide CO emission concentration, nitrogen dioxide NO2 emission concentration and sulfur dioxide SO2 emission concentration.

4. The incineration optimization method based on genetic algorithm improvement according to claim 2 is characterized in that: Each of said individuals represents a set of third incineration control parameters; The step of calculating the fitness of each individual in the initial population according to the fitness algorithm includes: Obtaining the third pollutant emission concentration corresponding to each set of the third incineration control parameters: inputting each set of the third incineration control parameters into a pollutant emission concentration prediction model, and obtaining the third pollutant emission concentration output by the third pollutant emission concentration prediction model; Obtaining the third fuel instantaneous flow rate corresponding to each set of the third incineration control parameters: inputting each set of the third incineration control parameters into a fuel instantaneous flow rate prediction model, and obtaining the third fuel instantaneous flow rate output by the third fuel instantaneous flow rate prediction model; The adaptability is determined based on the third pollutant emission concentration and the third instantaneous fuel flow rate.

5. The incineration optimization method based on genetic algorithm improvement according to claim 4 is characterized in that: The third pollutant emission concentration includes CO emission concentration, SO2 emission concentration and NO2 emission concentration; The fitness satisfies the following relationship: Among them, fitness is the fitness, represents the CO emission concentration, Indicates the SO2 emission concentration, Indicates NO2 emission concentration, Flow indicates the instantaneous flow rate of the third fuel, , and Respectively represent the weight values ​​of CO emission concentration, SO2 emission concentration and NO2 emission concentration, Indicates the weight value of the third fuel instantaneous flow rate.

6. The incineration optimization method based on genetic algorithm improvement according to claim 1 is characterized in that: The preset iteration termination condition includes: the number of iterations meets a preset threshold; The iterative optimization calculation is performed using a genetic algorithm based on the initial population and the fitness until a preset iterative termination condition is reached, including: Using the initial population and the fitness as initial parameters, inputting into a genetic algorithm for calculation; Iteratively optimizing the initial population, wherein each iteration generates a new population based on selection, crossover, and mutation; Until the number of iterative optimization processes meets the preset iteration termination condition.

7. The incineration optimization method based on genetic algorithm improvement according to claim 4 is characterized in that: The step of obtaining a first incineration control parameter set including incineration control parameters corresponding to a plurality of experimental samples comprises: The sensor data and infrared temperature measurement data of the incinerator are collected on a daily basis. The sensor data include: main combustion chamber temperature, average temperature, secondary combustion chamber wind speed, top wind speed, left wind speed, right wind speed, left smoke elimination wind speed, right smoke elimination wind speed, burner status, blower status, main oil valve status; the infrared temperature measurement data include: temperature values ​​of various points in the incinerator chamber; The sensor data and the infrared temperature measurement data are aligned and downsampled in units of minutes to obtain the first incineration control parameter set including incineration control parameters corresponding to a plurality of experimental samples.

8. The incineration optimization method based on genetic algorithm improvement according to claim 1 is characterized in that: The method further comprises: The pollutant emission concentration and the instantaneous fuel flow rate corresponding to the optimized incineration control parameters are output.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the improved incineration optimization method based on genetic algorithm as claimed in any one of claims 1 to 8 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the improved incineration optimization method based on genetic algorithm as described in any one of claims 1 to 8 is implemented.

Images