Carbon emission regulation method and low-carbon concrete product production system adopting the same

Abstract

The application relates to a carbon emission regulation method and a low-carbon concrete product production system adopting the method, and solves the problem that in the whole concrete production system, each link parameter lacks effective regulation means, the production process is difficult to be optimized as a whole, and efficient carbon emission control is difficult to be realized. The method comprises the following steps: combining the parameters optimized by a genetic algorithm into actual production equipment to realize real-time regulation of the production process; after the equipment is regulated, key parameter data of each link of the production system is continuously collected; according to the collected key parameter data, carbon emission equivalents are calculated according to a set carbon emission calculation formula, the carbon emission equivalents comprise direct carbon emission and indirect carbon emission; and according to the calculation result, a carbon emission accounting report is automatically generated. The application has the following effects: carbon emission intelligent regulation of a low-carbon concrete production process is realized, and the total carbon emission is effectively reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of concrete production emission reduction, and in particular to a carbon emission regulation method and a low-carbon concrete product production system adopting the same. BACKGROUND

[0002] With the increasing global concern about climate change, reducing carbon emissions has become an important task for all industries. The construction industry, as a major carbon emitter, faces a particularly severe emission reduction situation.

[0003] In the field of concrete product production, traditional cement production is one of the main sources of carbon emissions. In order to reduce carbon emissions during concrete production, the industry has begun to explore the use of by-products from coal-fired power generation. Coal-fired power generation produces solid by-products such as fly ash and gaseous by-products such as water vapor and carbon dioxide. Using gaseous by-products from coal-fired power generation to carbonize fly ash generates carbonized fly ash, which is then used to replace cement for producing concrete products, and reduces carbon emissions through carbonization curing. This technology has reduced carbon emissions in concrete production to some extent.

[0004] However, the current technology using coal-fired power generation by-products has obvious drawbacks. Coal-fired power plants are usually far away from low-carbon concrete production plants, which results in additional carbon emissions during transportation of the coal-fired power generation by-products, greatly reducing the emission reduction effect of this technology. Moreover, in the entire low-carbon concrete production process, there is a lack of effective regulation means for the parameters of each link, making it difficult to optimize the production process as a whole and achieve efficient carbon emission control. Therefore, a new carbon emission regulation method is urgently needed to solve these problems. SUMMARY

[0005] In order to realize intelligent regulation of carbon emissions in the production process of low-carbon concrete and effectively reduce the total amount of carbon emissions, the present application provides a carbon emission regulation method and a low-carbon concrete product production system adopting the same.

[0006] In a first aspect, the present application provides a carbon emission regulation method, which adopts the following technical solution:

[0007] A carbon emission regulation method, comprising:

[0008] Obtaining key parameter data of each link in the entire production system of low-carbon concrete products and performing data preprocessing;

[0009] Taking parameters related to carbon emissions and adjustable as decision variables of a genetic algorithm;

[0010] Randomly generating a preset number of individuals within the value range of the decision variables to form an initial population, each individual representing a possible combination of parameters;

[0011] According to the preset comprehensive evaluation index and the preset target, a preset fitness function construction method is used to construct a suitable fitness function;

[0012] For each individual in the initial population, according to its parameter combination, combined with the relevant data and calculation formula of the production system, its fitness value is calculated;

[0013] The tournament selection method is used to select individuals from the current population into the next generation population according to the fitness value of the individual;

[0014] Two individuals are randomly selected from the selected population as parents, and crossover operation is performed according to the preset crossover probability, and the individuals in the population are mutated according to the preset mutation probability;

[0015] The fitness function calculation, selection, crossover and mutation operations are repeatedly performed to form a new population until the set iteration termination condition is reached, then the iteration is stopped, and the optimal individual in the current population is output as the optimized parameter combination;

[0016] The parameter combination optimized by the genetic algorithm is applied to the actual production equipment to real-time control the production process;

[0017] After the equipment is controlled, the key parameter data of each link of the production system is continuously collected;

[0018] According to the collected key parameter data, the carbon emission equivalent is calculated according to the set carbon emission calculation formula, and the carbon emission equivalent includes direct carbon emission and indirect carbon emission;

[0019] According to the calculation result, a carbon emission accounting report is automatically generated.

[0020] By using the above technical scheme, the carbon emission of the low-carbon concrete product production system can be effectively reduced. By obtaining and processing key parameters, the parameter combination is optimized by the genetic algorithm, and the production process is real-time controlled. Continuous collection of parameters and calculation of carbon emission equivalent can accurately grasp the carbon emission situation, and automatically generate an accounting report, providing data support for enterprise energy saving and emission reduction, and helping enterprises to realize green production.

[0021] In the second aspect, the application provides a low-carbon concrete product production system, which adopts the following technical scheme:

[0022] A low-carbon concrete product production system comprises:

[0023] A coal-fired power generation system located in the concrete production area is used to produce electricity, fly ash, and carbon dioxide water vapor mixture.

[0024] A solid waste treatment system located in the concrete production area is used to generate carbonized fly ash;

[0025] A low-carbon concrete production system located in a concrete production area uses carbonized fly ash, alkali activators, and raw materials to produce low-carbon concrete products.

[0026] A gas delivery system located in the concrete production area is used to deliver carbon dioxide water vapor mixed gas to the solid waste treatment system and the carbonization curing chamber of the low-carbon concrete production system.

[0027] A belt conveyor system located in the concrete production area is used to deliver fly ash to the solid waste treatment system and carbonized fly ash to the low-carbon concrete production system.

[0028] An electricity delivery system located in the concrete production area provides electricity for the low-carbon concrete product production system and the carbon emission intelligent accounting system. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 is a schematic diagram of the overall process of a carbon emission regulation method according to an embodiment of the present application.

[0030] Figure 2 is a schematic diagram of the process of constructing an adaptive fitness function according to a predetermined comprehensive evaluation index and a predetermined target using a predetermined fitness function construction method according to another embodiment of the present application. DETAILED DESCRIPTION

[0031] The present application will be described in further detail below with reference to the accompanying drawings.

[0032] REFERENCE Figure 1 A carbon emission regulation method disclosed by the present application includes the following steps, which are as follows:

[0033] Step S100: Obtain key parameter data of each link of the entire production system of low-carbon concrete products and perform data preprocessing.

[0034] Among them, the links of the production system include a coal-fired power generation system, a solid waste treatment system, a low-carbon concrete production system, a gas delivery system, a belt conveyor system, and an electricity delivery system; the key parameter data refers to data that has an important influence on the low-carbon concrete production process and carbon emissions, including but not limited to: fly ash particle size data, reflecting the size of fly ash particles, affecting their participation in carbonation reaction activity and concrete performance; fly ash quality data, related to material proportioning in solid waste treatment and concrete production; carbonation reaction temperature data, affecting the carbonation reaction rate and degree of fly ash; carbon dioxide concentration data, directly reflecting the carbon emission level of each link; exhaust gas flow rate data, related to the calculation of total carbon dioxide emissions; raw material consumption data, determining the concrete mix proportion and indirect carbon emissions; carbonation curing temperature and humidity data, affecting concrete strength growth and carbonation effect.

[0035] The acquisition method is as follows: corresponding sensors and equipment are deployed at each key position of the production system to acquire data. For example, a particle size analyzer is installed at the fly ash screening equipment of the solid waste treatment system to monitor the fly ash particle size in real time; an electronic belt scale of the belt conveying system is used to measure the fly ash mass; a temperature sensor is arranged in the carbonization reaction device to collect the carbonization reaction temperature; a carbon dioxide concentration sensor and a flow rate sensor are deployed in the gas conveying system to acquire the carbon dioxide concentration and the exhaust gas flow rate in the gas; a multi-parameter sensor is installed in the carbonation curing chamber of the low-carbon concrete production system to collect the temperature and humidity data of the carbonation curing; and the raw material dosage data are acquired through the batching record and alkali activator adding device of the low-carbon concrete production system.

[0036] Data preprocessing is a process of cleaning, converting and arranging the collected raw data to improve the data quality and make it more suitable for subsequent analysis and processing. It includes operations such as removing outliers, filling missing values, standardizing processing, etc.

[0037] Step S200, the parameters related to carbon emissions and adjustable are taken as decision variables of the genetic algorithm.

[0038] Among them, the decision variable is a parameter that can be adjusted and optimized in the genetic algorithm, which affects carbon emissions, such as fly ash particle size, carbonization reaction temperature, carbonation curing temperature and humidity, etc. Changing them can adjust the carbon dioxide concentration in the exhaust gas and affect the carbon emissions of the production system.

[0039] Acquisition method: from the solid waste treatment system and the low-carbon concrete production system, relevant parameter data are collected through sensors, and adjustable parameters are selected as decision variables.

[0040] The necessary process is as follows: first, select the parameters related to carbon emissions and adjustable, and then determine the value range according to the production process and experience. For example, a certain plant determines that the fly ash particle size value range is 0.01-0.1 mm, the carbonization reaction temperature is 50-80℃, the carbonation curing temperature is 20-40℃, and the humidity is 60%-90%.

[0041] Step S300, randomly generate a preset number of individuals within the value range of the decision variable to form an initial population, and each individual represents a possible combination of parameters.

[0042] Among them, the initial population is composed of a preset number of individuals, and each individual is a combination of parameters of decision variables. In the low-carbon concrete production carbon emission regulation, it represents different production parameter collocations, which is used for subsequent genetic algorithm optimization.

[0043] The acquisition method is as follows: within the determined value range of the decision variable, use a random number generator to randomly generate parameter values to form individuals, and then form an initial population.

[0044] The general process is as follows: 1. Determine the decision variable value range, such as fly ash particle size 0.01-0.1mm, carbonization reaction temperature 50-80℃, etc. 2. Set the number of individuals in the population, such as 50. 3. Randomly generate parameter combinations to form individuals to form the initial population. For example, randomly generate a set of individuals (fly ash particle size 0.05mm, carbonization reaction temperature 65℃) as a member of the population.

[0045] Step S400, according to the preset comprehensive evaluation index and the preset target, using the preset fitness function construction method to construct the adaptive fitness function.

[0046] The fitness function is used to evaluate the pros and cons of each individual (parameter combination) in the population, and the adaptive fitness function can accurately reflect the degree of fit between the individual and the comprehensive evaluation index and the preset target.

[0047] The acquisition method is as follows: according to the preset comprehensive evaluation index (such as carbon emission, cost, production efficiency, etc.) and the preset target (such as carbon emission standard value, cost budget, production efficiency lower limit, etc.), using a specific construction method to obtain.

[0048] The general process is as follows: 1. Clearly define the comprehensive evaluation index and target of each system, such as carbonization reaction efficiency and carbon emission for solid waste treatment system, product quality and cost for low-carbon concrete production system. 2. Build the initial fitness function for each system. For example, for the solid waste treatment system, the function can be constructed according to the difference in carbon dioxide concentration before and after carbonization reaction and energy consumption. 3. Combine the functions of each system to construct the overall fitness function using weighted summation or other methods. For example, set the weight of the solid waste treatment system to 0.4 and the weight of the low-carbon concrete production system to 0.6.

[0049] Step S500, for each individual in the initial population, according to its parameter combination, according to the fitness function that has been constructed, combined with the relevant data and calculation formula of the production system, calculate its fitness value.

[0050] The fitness value is a quantitative index for evaluating the pros and cons of individuals (parameter combinations) in the initial population, which is obtained by specific calculation and is used to measure the degree of fit between the individual and the goal of reducing carbon emissions and improving production efficiency.

[0051] The acquisition method is as follows: according to the fitness function that has been constructed, combined with the data (such as carbon dioxide concentration, exhaust gas flow rate, fly ash quality, etc.) collected in real time by the production system and the relevant calculation formula, the parameter combination of each individual is calculated.

[0052] The data is calculated as follows: taking an individual as an example, assuming that its parameter combination is a specific fly ash particle size, carbonization reaction temperature, carbonization curing temperature and humidity.

[0053] According to formula (1) Calculate direct CO2 emissions. Indicates direct carbon dioxide emissions (in kg), is the concentration of carbon dioxide in the gas that changes with time t (in units of ), is the gas flow rate that changes with time t (in units of ).

[0054] According to formula (2) Calculate the amount of base activator required. is the amount of alkali activator used (in kg), is the conversion factor, is the mass of fly ash (in kg).

[0055] Using formula (3) Calculate the amount of desulfurization and denitrification reagents. Represents the amount of desulfurization and denitrification reagents used (in kg), is the conversion factor (assumed to be 0.03), Same meaning as above.

[0056] Using formula (4) Calculate the amount of other raw materials i, is the amount of raw material i (in kg), is the conversion coefficient of raw material i (assuming that the is 0.2), It is still the quality of fly ash.

[0057] According to formula (5) Calculate indirect CO2 equivalent emissions, Indirect carbon dioxide emissions equivalent (unit: ), are the amounts of desulfurization and denitrification reagents, alkali activators, and raw material i (in kg); are their carbon emission factors respectively.

[0058] Through formula (6) Calculate the total CO2 equivalent emissions, where are direct carbon dioxide emissions and indirect carbon dioxide emissions equivalent, That is, total carbon dioxide emissions equivalent.

[0059] Substitute these carbon emission data into the fitness function to obtain the fitness value of the individual.

[0060] Step S600: Using the tournament selection method, individuals are selected from the current population according to their fitness values ​​to enter the next generation population.

[0061] Tournament Selection: A selection strategy in genetic algorithms, where a certain number of individuals (this number is called tournament size) are randomly selected from the population, and the individual with the highest fitness value among these is chosen to pass to the next generation. This process is repeated until the number of individuals in the new population reaches a predetermined value. In this way, individuals with higher fitness values have a greater chance of being selected and passed to the next generation, guiding the population towards better solutions.

[0062] Current Population: Refers to the collection of all individuals at a certain iteration step in a genetic algorithm. These individuals represent different combinations of decision variables (such as different fly ash particle sizes, carbonation reaction temperatures, etc.). Their fitness values have been calculated in previous calculation steps (such as S500).

[0063] Next Generation Population: A new population formed by selecting individuals from the current population using the tournament selection method. It is the starting point for the next iteration of the genetic algorithm optimization process, and its composition will affect subsequent genetic operations and the direction of algorithm convergence.

[0064] The general process is as follows: 1. Set the tournament size: Determine the tournament size based on experience or experiment. For example, set the tournament size to 5, i.e. randomly select 5 individuals from the current population for comparison each time. 2. Selection operation: Randomly select 5 individuals from the current population, compare their fitness values, and select the individual with the highest fitness value to join the next generation population. Repeat the process of random selection and comparison until the number of individuals in the next generation population reaches the predetermined number. For example, if the population is set to 50, 50 such operations are needed to select 50 individuals to form the next generation population. In actual operation, a computer program can be used to traverse the current population, randomly select individuals each time, record and compare their fitness values, and complete the selection.

[0065] Step S700, randomly select two individuals from the selected population as parents, perform crossover operation according to the preset crossover probability, and perform mutation operation on the individuals in the population with the preset mutation probability.

[0066] Where, Parent: Two individuals in genetic algorithms selected to produce offspring, carrying production parameter combinations. Crossover operation: Simulate gene exchange, exchange part of the production parameter combinations of two parent individuals to produce new individuals. Mutation operation: Change some production parameters of individuals randomly according to a certain probability to avoid the algorithm falling into local optimum. Preset crossover probability: Set the possibility of parent crossover operation, value between 0 and 1. Preset mutation probability: Set the possibility of individual mutation, value between 0 and 1.

[0067] The necessary process is as follows:

[0068] Parent selection: randomly select two individuals from the selected population as parents. For example, in a population of 50 individuals, generate two different random numbers by random number generator, and select the individuals with these two numbers as parents.

[0069] Crossover operation: determine whether to perform crossover operation. Based on the preset crossover probability, for example, if the crossover probability is 0.8, generate a random number between 0 and 1 by random number generator, and if the random number is less than 0.8, perform crossover operation on the selected parents. Suppose the parameter combination of parent individual A is (fly ash particle size 0.05 mm, carbonation reaction temperature 60℃), and the parameter combination of parent individual B is (fly ash particle size 0.07 mm, carbonation reaction temperature 70℃), after crossover, offspring individual C (fly ash particle size 0.05 mm, carbonation reaction temperature 70℃) and offspring individual D (fly ash particle size 0.07 mm, carbonation reaction temperature 60℃) may be generated.

[0070] Mutation operation: for each individual in the population, determine whether to mutate based on the preset mutation probability. For example, if the mutation probability is 0.05, generate a random number between 0 and 1 for each individual, and if the random number is less than 0.05, perform mutation operation on the individual. For example, individual E (fly ash particle size 0.06 mm, carbonation reaction temperature 65℃) mutates to (fly ash particle size 0.08 mm, carbonation reaction temperature 65℃), only changing the fly ash particle size parameter.

[0071] Step S800, repeatedly perform fitness function calculation, selection, crossover and mutation operation to form a new population, until the set iteration termination condition is reached, then stop iteration and output the optimal individual in the current population as the optimized parameter combination.

[0072] Iteration: in genetic algorithm, repeatedly perform fitness function calculation, selection, crossover and mutation operation to make the population evolve continuously and gradually approach the optimal solution. Each such cycle is an iteration.

[0073] Setting iteration termination condition: in order to control the running process of genetic algorithm, set the condition for stopping iteration in advance. Usually includes reaching the maximum number of iterations (such as setting the maximum number of iterations to 100), the fitness value of the population converges (such as the fitness value of the optimal individual in the population changes less than 0.01 for 5 consecutive iterations), etc.

[0074] Optimal individual: in the current population, the individual with the highest fitness value calculated by the fitness function, whose production parameter combination performs best in reducing carbon emissions and improving production efficiency, etc.

[0075] The general process is as follows:

[0076] Iteration calculation: Repeat the fitness function calculation (step S500), selection (step S600), crossover (step S700), and mutation (step S700) operations to generate a new population.

[0077] Termination condition judgment: After each iteration, check if the set iteration termination condition is met. For example, set the maximum number of iterations to 50, when the number of iterations reaches 50, or the population fitness value has converged, stop iteration.

[0078] Output optimal individual: When the termination condition is met, find the individual with the highest fitness value from the current population, and output the corresponding production parameter combination. For example, after several iterations, the individual with the highest fitness value in the current population is the optimal result, which can be used to regulate the actual production.

[0079] Step S900, apply the parameter combination optimized by genetic algorithm to the actual production equipment to real-time control the production process.

[0080] Parameter combination optimized by genetic algorithm: refers to a set of parameter values that can achieve optimal results in reducing carbon emissions, improving production efficiency, etc. through genetic algorithm optimization, such as specific fly ash particle size, carbonation reaction temperature, carbonation curing temperature and humidity, etc.

[0081] Actual production equipment: in the low-carbon concrete product production system, various devices used in coal-fired power generation, solid waste treatment, concrete production, etc. such as combustion equipment of coal-fired power generation system, carbonation reaction device of solid waste treatment system, mixing equipment of low-carbon concrete production system, etc.

[0082] Real-time control: according to the parameters optimized by genetic algorithm, adjust the operating parameters of the production equipment in real time, so that the production process always remains in the optimized state, to achieve energy saving and emission reduction and efficient production.

[0083] The general process is as follows: 1. Parameter docking: clearly define the correspondence between the parameters optimized by the genetic algorithm and the adjustable parameters of the actual production equipment. For example, the optimized carbonization reaction temperature parameter corresponds to the temperature adjustment parameter of the carbonization reaction device in the solid waste treatment system. 2. Equipment adjustment: according to the correspondence, input the optimized parameters into the control system of the actual production equipment. For example, input the optimized carbonation curing temperature and humidity parameters into the temperature and humidity control system of the carbonation curing chamber of the low-carbon concrete production system. 3. Real-time monitoring and fine-tuning: after adjusting the equipment, use sensors to monitor the equipment operating state and production process indicators in real time. If it is found that the actual situation deviates from the expected value, the parameters can be fine-tuned according to the feedback. For example, if it is found that the actual temperature of the carbonation curing chamber is 2°C lower than the set temperature, fine-tune the temperature and humidity control system to achieve the optimized set value.

[0084] Step SA00, after equipment regulation, continuously collect key parameter data of each link of the production system.

[0085] The general process is as follows: 1. Determine the monitoring point: determine the monitoring point of the key parameter data in each production link. For example, in the solid waste treatment system, the monitoring points include the fly ash screening equipment (to obtain the fly ash particle size), the carbonization reaction device (to obtain the reaction temperature and carbon dioxide concentration); in the low-carbon concrete production system, including the batching area equipment (to obtain the raw material dosage), the carbonation curing chamber (to obtain the temperature and humidity), etc. 2. Data collection: sensors collect data in real time or periodically from recording equipment. For example, every 5 minutes, each sensor transmits the collected data to the data collection unit; the batching system uploads the raw material dosage data after completing each batching. 3. Data transmission and storage: the collected data is transmitted through wired or wireless transmission methods to the data collection unit of the carbon emission intelligent accounting system and stored in the system database for subsequent analysis and processing. For example, the sensors of the gas delivery system send carbon dioxide concentration and flow rate data to the data collection unit through a wireless transmission module.

[0086] Step SB00, according to the collected key parameter data, calculate the carbon emission equivalent according to the set carbon emission calculation formula, which includes direct carbon emission and indirect carbon emission.

[0087] Carbon emission calculation formula: a set of mathematical expressions for calculating carbon emissions during production, including formulas for calculating direct carbon emissions and indirect carbon emissions, which are the basis for quantifying carbon emissions. Reference can be made to the formulas (1) to (6) provided in step S500.

[0088] Carbon emission equivalent: the total carbon emission index obtained by considering direct and indirect carbon emissions, used to comprehensively measure the carbon impact of production activities on the environment, usually in kgCO2e.

[0089] Direct carbon dioxide emissions: Carbon emissions generated by exhaust gas from carbonation reaction device in solid waste treatment system and exhaust gas from carbonation curing chamber in low-carbon concrete production system.

[0090] Indirect carbon dioxide emissions: Carbon emission equivalent indirectly generated due to other activities (such as the use of alkali activator, desulfurization and denitrification reagent, and other raw materials) in addition to direct carbon dioxide emissions.

[0091] Step SC00, automatically generate carbon emission accounting report according to the calculation results.

[0092] Among them, carbon emission analysis report: a file that summarizes the carbon emission related data, analysis results and conclusions in the production process, including carbon emission equivalent, direct and indirect carbon emission equivalent and their proportion, etc., which provides the basis for evaluating the carbon emission of production activities.

[0093] Report template: a pre-set report format framework that specifies the content structure, layout style and data presentation method of the carbon emission analysis report, ensuring the standardization and consistency of the report.

[0094] The calculated carbon emission data: refers to the direct carbon emission, indirect carbon emission equivalent and total carbon dioxide emission equivalent calculated in step SB00.

[0095] The approximate process is as follows: 1, data extraction: the report generation unit extracts the calculated carbon emission data from the database, including direct carbon emission, indirect carbon emission equivalent and total carbon dioxide emission equivalent. 2, report generation: fill in the extracted data according to the format and requirements of the report template. For example, fill in the calculated direct carbon emission value in the "direct carbon emission" part of the report; fill in the indirect carbon emission equivalent value in the "indirect carbon emission equivalent" part; fill in the total carbon dioxide emission equivalent value in the "carbon emission equivalent" part, and calculate the proportion of direct and indirect carbon emission equivalent in the total carbon dioxide emission equivalent, and fill in the corresponding position. 3, report storage and output: the generated carbon emission analysis report is stored in the system for easy reference at any time. At the same time, according to the needs, it can be printed or exported as electronic documents (such as PDF, Excel format) to provide relevant personnel or departments, such as enterprise managers for decision analysis.

[0096] Reference Figure 2 According to the preset comprehensive evaluation index and the preset target, the adaptive fitness function is constructed by using the preset adaptive fitness function construction method, including:

[0097] Step S410, according to the comprehensive evaluation index of each link of the whole production system and the preset target, construct the initial fitness function of each system.

[0098] Among them, the comprehensive evaluation index: a series of indicators for measuring the operation and effect of each link of the production system, such as carbon emission indicators, production efficiency indicators, product quality indicators, etc. These indicators reflect the performance of the production system from different aspects.

[0099] The preset target: the target value that the production system is expected to achieve in advance, such as the expected minimum total carbon emissions, the highest production efficiency, etc.

[0100] The initial fitness function of each system: a function constructed for each subsystem (such as coal-fired power generation system, solid waste treatment system, etc.) in the production system, used to evaluate the degree of individual (decision variable combination) in the subsystem. The acquisition method is as follows: according to the comprehensive evaluation index and the preset target, combined with the characteristics and operation mechanism of each system, the fitness function is constructed. For example, for the solid waste treatment system, the fitness function can be constructed according to the indicators such as solid waste treatment capacity, treatment cost and carbon emission.

[0101] The general process is as follows:

[0102] Determine the comprehensive evaluation index of each system: analyze the function and operation characteristics of each subsystem in the production system, and determine the key indicators that affect its performance. For example, for the coal-fired power generation system, the comprehensive evaluation indicators may include power generation efficiency, coal consumption, pollutant emission, etc.

[0103] Define the preset target: set the target value for each comprehensive evaluation index according to the strategic planning and actual needs of the enterprise. For example, set the power generation efficiency target of the coal-fired power generation system to 80%, and the coal consumption target to 0.5 tons of coal per ton of electricity consumption.

[0104] Construct the initial fitness function of each system: according to the comprehensive evaluation index and the preset target, use appropriate mathematical methods to construct the fitness function. Generally speaking, the fitness function should reflect the closeness of the individual (decision variable combination) to the preset target. For example, for a certain evaluation index of a certain system, if the index value is closer to the target value, the fitness function can be constructed in the following simple form:

[0105] Let the evaluation index be x, and the target value be , then the fitness function is When x is equal to , the fitness function value is maximum 1; the greater the difference between x and , the smaller the fitness function value.

[0106] For a system with multiple evaluation indicators, the weighted sum method can be used to construct the fitness function. For example, a system has two evaluation indicators and .

[0107] The weights are and , then the fitness function where and are the fitness functions for and respectively.

[0108] Step S420, initialize population, encode each system decision variable into an individual.

[0109] The general process is as follows:

[0110] Determine the decision variable range: specify the value range of each decision variable. For example, the value range of fly ash particle size is 0.01-0.1 mm, and the value range of carbonation reaction temperature is 50-80℃.

[0111] Select the encoding method: assume that binary encoding is used. For fly ash particle size, if the accuracy requirement is 0.001 mm, there are a total of 90 possible values in the range of 0.01-0.1 mm, and at least 7 binary numbers (because 2^7=128>90) are needed to encode; for carbonation reaction temperature, if the accuracy is 1℃, there are 31 possible values in the range of 50-80℃, and at least 5 binary numbers (2^5=32>31) are needed to encode.

[0112] Generate individuals: randomly generate numerical values within the value range of the decision variables, and then convert these numerical values into code strings according to the selected encoding method to form individuals. For example, randomly generate fly ash particle size as 0.05 mm, which may be converted to binary code as 0110010; carbonation reaction temperature as 65℃, which may be converted to binary code as 100001, and these two code strings are concatenated to form an individual.

[0113] Assemble the population: repeat the step of generating individuals until the preset population size is reached. Assuming that the preset population size is 50, 50 such individuals need to be generated to form the initial population.

[0114] Step S430, calculate the fitness value of the individual under each target, perform non-dominated sorting and crowding degree calculation, and select individuals to enter the next generation.

[0115] Among them, the individual fitness value: the quantitative index of evaluating the individual (combination of decision variables) in achieving the production target, calculated by the fitness function, the higher the value, the better.

[0116] Non-dominated sorting: stratify individuals according to their merits, if individual A is not worse than B in all targets and at least one is better, then A dominates B, and the hierarchy is divided accordingly.

[0117] ​Crowding distance calculation: measure the density of individuals in the same layer, the larger the crowding distance means the sparser the distribution of individuals around, and the more advantageous the selection, which can maintain the diversity of the population.

[0118] The general process is as follows: 1. Calculate the fitness value: put the individual into the fitness function to calculate the comprehensive fitness value. 2. Non-dominated sorting: compare the fitness values of individuals, put the non-dominated individuals in the first layer, and then sort them according to the dominance relationship. 3. Crowding distance calculation: determine the boundary individuals of a layer, and calculate the sum of the distance between the intermediate individuals and adjacent individuals in each target dimension as the crowding distance. 4. Select individuals: select individuals from the front layer, and select individuals with larger crowding distance in the same layer to meet the population size of the next generation.

[0119] Step S440, after a preset number of iterations, the Pareto optimal solution set is obtained, providing multiple sets of trade-off schemes for decision-making.

[0120] Among them, the preset number of iterations: the number of genetic algorithm loop operations set in advance, each iteration includes selection, crossover, mutation and other operations, which promotes the evolution of the population to a better direction. Pareto optimal solution set: in multi-objective optimization problems, a set of solutions that cannot make at least one target better without making other targets worse. In this scenario, it is a series of decision variable combinations that balance multiple targets (such as carbon emissions, cost, production efficiency) in the low-carbon concrete product production system. Trade-off scheme: for multiple conflicting objectives, each solution in the Pareto optimal solution set represents a trade-off scheme, i.e. a balance between targets, providing multiple choices for decision-makers.

[0121] The general process is as follows:

[0122] Iteration operation: according to the set genetic algorithm process, the current population is selected in turn (such as selecting individuals according to non-dominated sorting and crowding distance), crossed (exchanging part of the decision variables of individuals), and mutated (randomly changing part of the decision variables of individuals) to generate a new population. Each time such a operation sequence is completed, a round of iteration is completed.

[0123] Judge the number of iterations: after completing a round of iteration, check whether the preset number of rounds is reached. For example, after 99 iterations, after one more round of operation, it is found that the preset number of 100 rounds is reached.

[0124] Pareto optimal solution set acquisition: After reaching the preset number of rounds, the Pareto optimal solution set is screened from the last generation population. For example, there are 100 individuals in the last generation population. By comparing the performance of these individuals in terms of carbon emissions, cost and production efficiency and other objectives, those individuals whose performance cannot be improved without reducing other objectives are found. The combination of decision variables corresponding to these individuals constitutes the Pareto optimal solution set. These solutions are the trade-off options available to decision makers, such as a scheme with lower carbon emissions but slightly higher cost, and another scheme with higher production efficiency but relatively higher carbon emissions. Decision makers can choose according to actual needs.

[0125] Step S450, use neural network algorithm to establish each system model to predict target values under different decision variables.

[0126] Among them, the neural network algorithm: a computational model that simulates the structure and function of human brain neurons, composed of a large number of nodes (neurons) and edges connecting these nodes, which establishes the complex relationship between input and output through learning data. In this scenario, it is used to model each system.

[0127] Each system model: for coal-fired power generation system, solid waste treatment system, low-carbon concrete production system and other systems in the production system, a mathematical model constructed by using neural network algorithm, used to describe the relationship between decision variables (such as fly ash particle size, carbonation reaction temperature, etc.) and target values (such as carbon emissions, production efficiency, etc.) in the system.

[0128] The general process is as follows:

[0129] Data preparation: collect and organize historical data of decision variables and target values in each system, clean the data and remove outliers and incorrect data. For example, remove obviously unreasonable temperature data caused by sensor failure. Divide the data into training set and test set, generally the training set is used to train the neural network and the test set is used to evaluate the model performance. For example, 70% of the data is used as the training set and 30% as the test set.

[0130] Select neural network structure: select appropriate neural network structure according to system characteristics and data characteristics. For simple systems, you can choose a multilayer perceptron with one or more hidden layers. Determine the number of neurons in the hidden layer, activation function and other parameters. For example, for the solid waste treatment system, a multilayer perceptron with 2 hidden layers, 10 neurons in each layer and ReLU activation function is selected.

[0131] Training model: Train the selected neural network using the training set data. Take the decision variables as input and the target values as output, and adjust the weights and biases of the neural network to make the model's predicted output as close to the actual target values as possible. Gradient descent and other optimization algorithms can be used during training. For example, after 1000 training iterations, the model's prediction error gradually decreases and stabilizes.

[0132] Model evaluation: Evaluate the performance of the trained model using the test set data, and calculate the model's prediction error, such as mean squared error (MSE), mean absolute error (MAE), etc. If the error is within an acceptable range, the model is considered good; if the error is too large, the neural network structure needs to be adjusted or retrained. For example, the MSE of the solid waste treatment system model is calculated to be 0.05, which meets the precision requirements of the production system.

[0133] Model application: Apply the trained and evaluated model to the system, input different decision variable values, and the model can predict the corresponding target values. For example, input a new combination of fly ash particle size and carbonation reaction temperature, and the solid waste treatment system model can predict the corresponding carbon dioxide emissions.

[0134] Step S460, integrate each system fitness function using the weighted sum method to construct the overall fitness function.

[0135] The overall fitness function is as follows: ; wherein, represents the overall fitness function value of the entire low-carbon concrete product production system, is an index variable used to identify different systems in the low-carbon concrete product production system, represents the weight coefficient of the th system in the overall fitness function, is the fitness function value of the th system.

[0136] The general process is as follows: 1. Determine the weight coefficient: determine the weight coefficient of each system fitness function by considering multiple factors. 2. Calculate the overall fitness function: obtain the system fitness function values and calculate the overall fitness function value according to the weighted sum formula. 3. Apply the overall fitness function: the obtained overall fitness function can be used for subsequent genetic algorithm operations, such as evaluating the quality of individuals, selecting individuals to enter the next generation, etc., to guide the system to evolve towards the optimal direction of the overall goal.

[0137] Apply the parameter combination optimized by the genetic algorithm to the actual production equipment to perform real-time control of the production process, including the following steps:

[0138] Step S910, the parameter combination optimized by the genetic algorithm is directly applied to the production equipment. After the parameter setting is completed, the production process is started.

[0139] The production process refers to the entire process of producing low-carbon concrete products from the input of raw materials to a series of processing and handling, covering multiple links such as power generation, solid waste treatment, concrete batching, mixing, pouring, and curing.

[0140] The general process is as follows:

[0141] Parameter matching: the parameters optimized by the genetic algorithm are matched with the adjustable parameters of the production equipment. For example, the optimized carbonation reaction temperature parameter corresponds to the temperature adjustment parameter of the carbonation reaction device in the solid waste treatment system; the optimized fly ash particle size parameter corresponds to the screen size adjustment parameter of the fly ash screening equipment.

[0142] Parameter setting: according to the matching relationship, the optimized parameters are input into the control system of the production equipment. For example, the optimized carbonation curing temperature of 50°C and humidity of 70% are input into the temperature and humidity control system of the carbonation curing room of the low-carbon concrete production system.

[0143] Start production: after the parameter setting is completed, start the production equipment and start the entire production process. For example, start the combustion equipment of the coal-fired power generation system to start power generation and provide energy for other systems, driving the solid waste treatment system and the low-carbon concrete production system to start running in turn.

[0144] Step S920, based on the preset process analysis model, real-time calculation of carbon emissions of each link, and dynamic adjustment of calculation parameters to adapt to production conditions using the preset adjustment strategy.

[0145] The preset process analysis model: a mathematical model set in advance for calculating carbon emissions of each link in the production process, which is based on production technology, chemical reaction principles, and carbon emission-related theories, and can calculate carbon emissions according to input production data.

[0146] Real-time calculation: based on real-time data continuously generated in the production process, carbon emissions are continuously calculated to reflect the dynamic changes in carbon emissions in the production process.

[0147] Carbon emissions of each link: refers to the amount of carbon dioxide emissions generated in each link of the low-carbon concrete product production system, such as coal-fired power generation, solid waste treatment, and low-carbon concrete production.

[0148] The preset adjustment strategy: a method prepared in advance for dynamically adjusting calculation parameters, aiming to make the calculation model better adapt to changes in production conditions and ensure the accuracy of carbon emission calculation.

[0149] Computational Parameters: Variables that affect the calculation of carbon emissions in the process analysis model, such as reaction temperature, raw material usage, gas flow rate, etc. These parameters are adjusted as production conditions change.

[0150] Production Conditions: Factors that affect the accuracy of the carbon emission calculation model, such as raw material quality, equipment operation status, environmental temperature and humidity, etc. Changes in these factors can affect the accuracy of the carbon emission calculation model.

[0151] The general process is as follows:

[0152] Data Collection: The system collects real-time data from various sensors, such as the temperature and carbon dioxide concentration in the carbonation reaction device, and the mass of fly ash on the belt conveying system every minute.

[0153] Initial Calculation: The real-time data collected is input into the pre-set process analysis model to calculate the initial carbon emission values of each link. For example, based on the collected carbonation reaction temperature, carbon dioxide concentration, and reaction time data at a certain time, the model calculates the carbon emissions of the solid waste treatment system at that time.

[0154] Judgment of Production Condition Changes: Compare the current collected data with the previous historical data to determine whether the production conditions have changed. For example, if the content of a certain component in the raw materials exceeds the normal range for several consecutive times, or if the equipment operating temperature fluctuates significantly, it is considered that the production conditions have changed.

[0155] Adjustment of Computational Parameters: When changes in production conditions are found, adjust the computational parameters according to the pre-set adjustment strategy. For example, if the content of a certain component in the raw materials that affects carbon emissions increases, adjust the relevant coefficient of this component in the carbon emission calculation model according to the adjustment strategy.

[0156] Recalculation of Carbon Emissions: Recalculate the carbon emissions of each link using the adjusted computational parameters to obtain more accurate results for subsequent decision-making.

[0157] Step S930: Remove outliers through industrial-level data verification algorithm and transmit the verified data to the next parameter adjustment link in real time.

[0158] Industrial-level data verification algorithm: An algorithm specifically designed for industrial production data processing, which can identify and remove abnormal data that does not conform to the actual production situation or contains errors, ensuring the accuracy and reliability of the data, and providing effective basis for subsequent production decision-making. Common algorithms include statistical-based methods (such as 3σ principle), machine learning-based isolation forest algorithm, etc., which can be selected according to actual situation.

[0159] Outlier: In production data, a data point that deviates significantly from the normal range of data. These data may be caused by sensor failure, data transmission error or sudden abnormality in the production process, which will affect the accuracy of production parameter calculation and analysis.

[0160] Verified data: After processing by industrial-level data verification algorithm, the data after removing outliers can more truly reflect the actual situation of the production process.

[0161] Parameter adjustment link: In the production control process, according to the verified data and production target, the stage of adjusting and optimizing the production parameters to make the production process better.

[0162] Step S940, use PID control algorithm to compare the difference between the actual carbon emissions generated under the current production parameters and the carbon emission target.

[0163] PID control algorithm: Proportion, integral and derivative control algorithm, by proportion, integral and derivative operation on the deviation (the difference between the current value and the target value), output a control quantity, used to adjust the system to reach or approach the target state, widely used in parameter adjustment in industrial production.

[0164] Current production parameters: refers to the parameters running in the production process, such as carbonation reaction temperature in solid waste treatment system, raw material dosage in low-carbon concrete production system, etc. These parameters reflect the actual state of production in real time.

[0165] Carbon emission target: the carbon emission level expected to be achieved in the production process set by the enterprise in advance, which is an important indicator to measure whether the production process is environmentally friendly and efficient.

[0166] Difference: the gap between the actual carbon emissions generated under the current production parameters and the carbon emission target, which is the basis for PID control algorithm to adjust.

[0167] The general process is as follows: 1. Data collection and acquisition: collect current production parameters from sensors and obtain carbon emission targets from production management systems. For example, obtain the carbonation reaction temperature of the solid waste treatment system at this moment as 65℃, and the established carbon emission target of this link as 50kg of carbon dioxide emission per hour. 2. Calculate the actual carbon emission: according to the preset process analysis model in step S920, combine the current production parameters to calculate the actual carbon emission. Suppose that after calculation, under the current 65℃ reaction temperature and other parameters, the actual carbon emission of the solid waste treatment system is 60kg per hour. 3. Calculate the difference: subtract the carbon emission target from the actual carbon emission to obtain the difference value. In this example, the difference is 60-50=10kg. 4. PID operation: input the difference value into the PID control algorithm. The proportional link outputs an adjustment amount proportional to the difference; the integral link integrates the difference to eliminate the steady-state error of the system; the differential link outputs an adjustment amount according to the rate of change of the difference to respond to the change of the system in advance. For example, if the difference is large, the proportional link outputs a large adjustment amount, the integral link gradually accumulates the adjustment to reduce the long-term error, and the differential link adjusts the adjustment speed according to the trend of the difference to generate a control signal together, which provides a basis for subsequent adjustment of production parameters.

[0168] Step S950, according to the difference analysis result, use the preset parameter adjustment formula to adjust the parameter.

[0169] The specific parameter adjustment formula is as follows:

[0170] ;

[0171] Among them, represents the adjusted production parameter value, indicates the production parameter value before adjustment, i.e. the currently used production parameter, is a dynamic adjustment coefficient, is the carbon emission deviation, is a time decay function.

[0172] Step S960, combine real-time raw material prices, energy costs and carbon trading market data to quantify the comprehensive benefits of different parameter adjustment schemes through the preset benefit formula.

[0173] The specific benefit formula is as follows:

[0174] ;

[0175] Among them, represents the comprehensive benefit value, is the maximum carbon emission reduction that can be achieved, which is a preset reference value, is the maximum cost increase that may occur, which is also a preset reference value, To reduce carbon emissions, is the cost increase, and is the weight coefficient.

[0176] The general process is as follows:

[0177] Data Collection: Real-time data on raw material prices, energy costs, and carbon trading markets is collected, and the carbon emission reductions and cost increases under the parameter adjustment plan are calculated. For example, after implementing a parameter adjustment plan, the model calculates that carbon emissions decreased by 10 tons, while costs increased by 5,000 yuan based on changes in raw material and energy use.

[0178] Determine the weight coefficient: clearly define the weight coefficients of the preset carbon emission reduction and cost increase. Assuming that the company pays more attention to carbon emission reduction, set the weight coefficient of carbon emission reduction The weight coefficient of cost increase is 0.7 is 0.3.

[0179] Calculate comprehensive benefits: Substitute the carbon emission reduction, cost increase and corresponding weight coefficient into the preset benefit formula for calculation.

[0180] Evaluation plan: Based on the calculated comprehensive benefit value, different parameter adjustment plans are evaluated and compared. The plan with the higher benefit value is usually the better.

[0181] Step S970: determine the non-inferior solution through Pareto frontier analysis, and then screen out the optimal parameter combination based on the preset enterprise goals.

[0182] Among them, Pareto frontier analysis is a multi-objective optimization analysis method that seeks a set of optimal solutions among multiple conflicting objectives (such as reducing carbon emissions, reducing costs, improving production efficiency, etc.). Among these solutions, there is no solution that can make at least one objective better without making other objectives worse. The set of these optimal solutions is the Pareto frontier.

[0183] Non-inferior solution: also called Pareto optimal solution, is a solution in the Pareto frontier. That is, in a multi-objective optimization problem, for a certain solution, if there is no other solution that is better than it in all objectives, then this solution is a non-inferior solution.

[0184] Preset corporate goals: Goals that companies set in advance to achieve during the production process based on factors such as the current market environment and their own development status. For example, these goals include reducing carbon emissions to a certain level within a certain period of time, or controlling production costs within a certain budget.

[0185] Optimal parameter combination: from the non-inferior solution set, combined with the preset enterprise target, a set of production parameters that can best meet the current production needs and goals of the enterprise is screened out, such as the combination of temperature, raw material dosage and other parameters.

[0186] Step S980, apply the adjusted parameters to production.

[0187] The preset adjustment strategy includes the following steps:

[0188] Step S921, use Kalman filtering algorithm to fuse multi-source data in the production process, and use principal component analysis method to extract key indicators from the fused data.

[0189] Among them, Kalman filtering algorithm: an algorithm that uses linear system state equation to make optimal estimation of system state through system input and output observation data.

[0190] The general process is as follows: 1. Data collection: use various sensors distributed in the production system, such as temperature sensors to collect reaction temperature data, pressure sensors to collect internal pressure data, etc., to obtain multi-source data in real time. 2. Kalman filtering data fusion: input the collected multi-source data into the Kalman filtering algorithm, and the algorithm processes the data according to the dynamic change characteristics and noise model of the data, removes noise interference, and fuses to obtain more accurate and stable data sequence. For example, Kalman filtering can smooth the fluctuating data of the temperature sensor due to environmental interference, and give an estimated value closer to the true temperature. 3. Principal component analysis to extract key indicators: input the fused data into the principal component analysis model, the model calculates the correlation between variables, and through dimension reduction operation, extracts the key indicators that can represent the main information of the data.

[0191] Step S922, use a hybrid prediction model combining long short-term memory network and support vector regression trained in advance to predict the key indicators under the current production conditions and output the prediction results.

[0192] Hybrid prediction model: a model combining long short-term memory network and support vector regression, which fully utilizes the ability of LSTM to process long-term dependencies of sequence data and the advantage of SVR to handle nonlinear relationships, and improves the accuracy of key indicator prediction.

[0193] The general process is as follows: data preparation: collect historical data of the key indicators extracted in step S921, and preprocess the data, such as normalization, division of training set and test set, etc. For example, scale the data of temperature, pressure and other key indicators to the interval [0, 1], and divide the training set and test set according to the ratio of 80% and 20%.

[0194] Model Training: Input the training set data into the hybrid prediction model for training. The LSTM part learns the time series features of the key indicators data, and the SVR part learns the nonlinear relationships in the data. During the training process, the parameters of the model (such as the number of hidden layer neurons of LSTM, the kernel function parameters of SVR, etc.) are constantly adjusted to minimize the prediction error.

[0195] Model Evaluation: Evaluate the trained model using the test set data, calculate the error indicators (such as mean square error, mean absolute error, etc.) between the predicted results and the actual values. If the error indicators do not meet the requirements, adjust the model parameters or increase the amount of training data, and retrain.

[0196] Prediction Output: Input the key indicator data under the current production condition into the trained and evaluated hybrid prediction model, and the model outputs the prediction results of the key indicators. For example, predict the values of temperature, pressure, etc. of the key indicators in the future one hour.

[0197] Step S923, based on the prediction results and the uncertainty factors in the production process, use the Monte Carlo simulation method to generate multiple production scenarios.

[0198] Among them, the prediction results: from step S922, use the hybrid prediction model combining long short-term memory network and support vector regression to predict the values of key indicators under the current production condition, such as production efficiency, carbon emissions, etc. in the future period.

[0199] Uncertainty factors in the production process: variables that are difficult to accurately predict or control in production, such as subtle differences in raw material quality, probability of equipment failure, changes in external environment temperature and humidity, etc.

[0200] Production scenario: description of the possible state of the production process under different combinations of uncertainty factors, including equipment running state, raw material usage, product output, and carbon emission level, etc.

[0201] The general process is as follows:

[0202] Determine the range of uncertainty factors: according to the collected historical data and experience, determine the value range or probability distribution of each uncertainty factor.

[0203] Set the number of simulations: according to the required accuracy and computing resources, determine the number of Monte Carlo simulations.

[0204] Simulate sampling: for each simulation, randomly sample within the value range of each uncertainty factor. For example, for the impurity content of raw materials, randomly select a value within the uniform distribution of 1%-5%; for equipment failure, randomly determine whether it occurs according to the probability of 0.05.

[0205] Generating production scenarios: Substitute the uncertainty factor value obtained in each sampling into the production process model combined with the prediction result of step S922 to generate a production scenario. For example, according to the impurity content of the extracted raw materials, adjust the production process parameters, combine the predicted production efficiency, and obtain the product yield, quality, and carbon emissions, etc. to form a complete production scenario description. Repeat the above steps until the set number of simulations is completed to generate multiple production scenarios. In the low-carbon concrete product production system, the production process model integrates raw materials, production equipment, production process, and carbon emissions, etc. The raw material module calculates the initial production conditions according to the characteristics and ratio of various materials; the production equipment module describes the influence of equipment operating parameters and state on production; the production process module reflects the effect of different process parameters on concrete performance and carbon emissions and calculates intermediate variables; the carbon emission module integrates each link to calculate direct and indirect carbon emissions. This model is the basis for Monte Carlo simulation to generate production scenarios. Substitute the uncertainty factors and prediction results into it to obtain specific production conditions, and also provide a basis for production optimization. Through the analysis of a large number of scenarios, it helps decision-makers to find problems and make targeted improvements to achieve efficient and stable operation of the production system.

[0206] Step S924, risk assessment is performed on each scenario using a risk degree quantification analysis method to calculate the probability of carbon emissions exceeding the target value and the possible economic losses under different scenarios.

[0207] Risk degree quantification analysis method: A method of using mathematical models and statistical means to present the risks in the production scenario in the form of specific numerical values, so as to intuitively evaluate the risk size.

[0208] Economic loss: The sum of economic negative effects such as additional carbon emission reduction costs and losses caused by production delays due to carbon emissions exceeding the target value.

[0209] Probability of carbon emissions exceeding the target value: The possibility of carbon emissions being higher than the target value set by the enterprise in a certain production scenario, expressed in percentage.

[0210] Step S925, carbon emission control, cost control, and production efficiency improvement are taken as multiple targets to establish a multi-objective optimization model, and the optimal solution is obtained according to the multi-objective optimization model, and a production parameter adjustment plan is made.

[0211] Let the carbon emissions be , the cost be , and the production efficiency be , the objective function can be represented as:

[0212] Minimize: ;

[0213] Constraints: ;

[0214] wherein, is a weight coefficient, reflecting the importance of each target; is the carbon emission target, is the cost budget, is the minimum production efficiency requirement.

[0215] According to the multi-objective optimization model, the optimal solution is obtained, and the production parameter adjustment plan is formulated as follows: using optimization software or programming to realize the algorithm, solving the model to obtain the optimal solution, and then formulating a detailed plan according to the optimal solution obtained by the model. The adjustment time node is determined, such as adjusting the plan to start next Monday; the responsible person is specified, such as the production workshop supervisor; the monitoring points are determined, such as monitoring the production efficiency and carbon emission data every hour during the adjustment process.

[0216] The risk assessment is carried out by using the risk degree quantification analysis method for each scenario, and the probability of carbon emission exceeding the target value and the possible economic loss under different scenarios are calculated, including the following steps:

[0217] Step S9241, collect data covering energy consumption, equipment operation, raw materials, environment and carbon trading market mechanism parameters.

[0218] Among them, the energy consumption data: the use amount and consumption rate of various types of energy (such as electricity, coal, natural gas, etc.) in the production process, etc. Data reflecting energy use, which plays an important role in analyzing carbon emissions and costs. Equipment operation data: including the running state of the equipment (such as the length of time on, the number of shutdowns), operating parameters (such as temperature, pressure), etc., which can be used to evaluate the performance of the equipment and its impact on the production process. Raw material data: involving the types, quality, quantity of raw materials and the characteristic data of raw materials (such as chemical composition, physical properties), which will affect the production process and carbon emissions. Environmental data: mainly refers to the relevant data of the production environment, such as environmental temperature, humidity, air pressure, etc., which may have an impact on equipment operation and production process, and then affect carbon emissions. Carbon trading market mechanism parameter data: including carbon trading market price fluctuation data, which is crucial for assessing the economic loss caused by carbon emission exceeding the standard.

[0219] Step S9242, taking equipment aging, energy use, and raw material characteristics as key nodes, and constructing the connection relationship among them, and constructing a Bayesian network according to the historical data statistics of the conditional probability between nodes.

[0220] The general process is as follows: 1. Connectivity Construction: Based on production principles and practical experience, analyze the causal relationships between key nodes and construct connectivity. For example, changes in energy usage may lead to accelerated equipment aging, so a directed edge exists from the energy usage node to the equipment aging node. Different raw material properties may affect energy usage patterns and efficiency, thus establishing a connection between raw material properties and energy usage. 2. Historical Data Collection and Organization: Collect historical data related to key nodes over a period of time (such as the past year or several years) and perform cleansing and preprocessing to ensure data accuracy and completeness. For example, organize equipment maintenance records to extract information such as the age and type of failure at each repair. Compile energy consumption records to record the amount of energy used by each type over different time periods. 3. Conditional Probability Statistics: Utilize the organized historical data and apply statistical methods to calculate conditional probabilities between nodes. For example, calculate the probability that energy usage exceeds a certain threshold when the equipment reaches a certain stage of aging, or the probability that equipment will fail (a sign of equipment aging) under certain raw material properties. Frequency estimation can be used to estimate conditional probabilities by calculating the proportion of data samples that meet specific conditions relative to the total number of samples. 4. Bayesian Network Construction: Using data analysis tools, input the determined nodes, connection relationships, and calculated conditional probabilities to construct a Bayesian network model. For example, using the Python pgmpy library, define node variables, establish directed edges between nodes, and set up a conditional probability table to generate a complete Bayesian network structure for subsequent risk assessment and analysis.

[0221] Step S9243: Select an appropriate Copula function based on the historical data characteristics of energy prices and production process parameters, and use the maximum likelihood estimation method to determine the function parameters.

[0222] Copula function: A function used to describe the correlation structure between multiple random variables. It can link the marginal distributions of random variables. In risk assessment, copula functions can capture nonlinear correlations between different factors (such as energy prices and production process parameters) without relying on the specific distribution form of the variables.

[0223] Maximum Likelihood Estimation: A statistical method used to estimate model parameters given known sample data. The basic idea is to find a set of parameter values ​​that maximizes the probability of observing the sample data under those parameters, thereby determining the optimal estimate of the unknown parameters in the model.

[0224] Copula function type selection: Based on the characteristics of energy price and production process parameter data (such as the distribution shape of data, linear or nonlinear relationship between variables, etc.), the appropriate Copula function type is preliminarily selected. Common Copula functions include Gaussian Copula, t-Copula, Archimedean Copula, etc. For example, if the data shows an approximate elliptical distribution and the correlation between variables is relatively symmetric, the Gaussian Copula function may be selected; if the data has thick tail phenomenon, the t-Copula function may be more appropriate. Scatter plots between variables, correlation coefficient matrix calculation, etc. can be used to assist in judgment.

[0225] Application of maximum likelihood estimation method: Use the selected Copula function to model the data after sorting. Taking the Python scipy.stats library as an example, call the corresponding function, input the energy price and production process parameter data, and use the maximum likelihood estimation method to estimate the parameters in the Copula function. For example, for the Gaussian Copula function, the parameters of its correlation coefficient matrix need to be estimated. During the estimation process, the software will adjust the parameter values through iterative calculation, so that the likelihood function value of the observed data under the Copula function model is maximized.

[0226] Step S9244, input the specific factor values of each scenario into the Bayesian network and Copula function model, perform Monte Carlo simulation for more than a preset number of times, distribute sample weights according to the importance of factors on carbon emissions, count the number of times that carbon emissions exceed the target value in the simulation, and then calculate the exceedance probability.

[0227] The approximate process is as follows:

[0228] Extract scenario factors: Extract energy, equipment, and raw material related data from each scenario, such as energy consumption, equipment running time, raw material composition, etc.

[0229] Set simulation parameters: Determine the number of simulations (such as 10000 times), and assign sample weights to each factor based on historical analysis and expert experience, such as energy consumption weight 0.4, equipment aging weight 0.2.

[0230] Carry out simulation calculation: Input the factor values into the Bayesian network and Copula function model, and use Monte Carlo tools to simulate for a preset number of times. Each time a combination of factors is randomly generated, and the carbon emissions are calculated.

[0231] Statistical exceedance times: After each simulation, compare the carbon emissions with the target value. If it exceeds the target, it is recorded and the total number of exceedances is counted, such as 2000 times out of 10000 simulations.

[0232] Calculate the probability of exceeding the standard: the number of times the standard is exceeded divided by the total number of simulations, such as 2000 ÷ 10000 = 20%, to assess the risk of carbon emissions exceeding the standard in each scenario.

[0233] Step S9245, first, the real-time price of the carbon trading market, the economic cost benchmark of environmental violations, the device status and rectification demand data are counted, and then the carbon emission prediction result of the current scenario is input into the trained economic loss prediction model together with the above statistical data, and finally the economic loss that may be generated in this scenario is output.

[0234] Economic loss prediction model: a mathematical model trained using historical data and related algorithms, which can predict the economic loss that a company may face based on input carbon emission data, market prices, and device-related information, including equipment rectification costs, etc.

[0235] The model calculates according to the algorithm and logic set inside it based on the input data. For example, combined with device status and rectification demand data, it calculates the economic loss of equipment rectification and other items.

[0236] The model outputs the economic loss value that the enterprise may face in the current scenario, providing a reference basis for enterprise decision-making. For example, the output result shows that the total economic loss is 800,000 yuan in the current scenario.

[0237] According to the calculation results, the carbon emission accounting report is automatically generated, including:

[0238] Step SC10, collect various carbon emission-related data to form a data set.

[0239] Among them, carbon emission-related data: covers data directly or indirectly related to carbon emissions in the production process. Data set: a collection formed by integrating and arranging various carbon emission-related data collected for subsequent analysis, modeling, and other operations.

[0240] Step SC20, use the density peak clustering-based emission source classification algorithm to divide data points into different emission source categories and calculate the carbon emission contribution proportion of each category.

[0241] Density peak clustering-based emission source classification algorithm: an algorithm for dividing data points into different categories (i.e., emission source categories). It identifies cluster centers based on the local density of data points and the distance from high-density points, and then classifies data points into corresponding clusters to classify different emission sources. In carbon emission data processing, data with similar carbon emission characteristics can be classified into the same emission source category.

[0242] The process of calculating the proportion is as follows: for each emission source category, the carbon emission amount and total carbon emission amount are calculated based on energy consumption, carbon content of raw materials, and carbon emission coefficient. The emission amount of each category is divided by the total amount to obtain the carbon emission proportion. For example, if a certain category emits 100 tons and the total amount is 500 tons, the proportion is 20%.

[0243] Step SC30, for each emission source category, analyze its unique characteristics, and collect and fuse data related to its unique characteristics based on its unique characteristics.

[0244] Unique characteristics: Each emission source has characteristics that distinguish it from other categories. These characteristics can be reflected in energy consumption patterns, equipment operation characteristics, and raw material usage. For example, a certain emission source may have the characteristic of energy consumption concentrated in a certain time period, or the characteristic of high carbon content of raw materials used.

[0245] Related data collection: To analyze each emission source in depth, collect more detailed data related to its unique characteristics. These data may include additional energy consumption details, specific parameters of equipment operation, and special property data of raw materials.

[0246] Data fusion: Integrate the various data collected related to the unique characteristics of the emission source category to form a comprehensive and integrated data set for more accurate analysis and understanding of the carbon emission of the emission source.

[0247] Step SC40, extract key features from the fused data using principal component analysis method.

[0248] Key features: Information extracted from the fused data that plays a key role in describing and understanding the carbon emission of each emission source.

[0249] Step SC50, construct independent prediction models for each emission source category, and weight and sum the output results of each emission source category prediction model according to the carbon emission proportion to obtain the overall carbon emission prediction value.

[0250] Independent prediction model: A separate model constructed for each emission source using its related data to predict the carbon emission of that emission source. Different categories of emission sources need to be modeled independently due to their unique characteristics to more accurately reflect their carbon emission patterns.

[0251] Overall carbon emission prediction value: The value obtained by summing the output results of the independent prediction models of each emission source category, representing the predicted amount of future carbon emissions of the entire production process.

[0252] Step SC60, integrate the emission source classification results, characteristics and proportions of each category, and carbon emission trend prediction curve to form the main body of the report.

[0253] Carbon emission trend prediction curve: generated with data analysis software (such as Python's Matplotlib library, Excel).

[0254] Content filling layout: the classification results are displayed in a table showing the emission source categories and names, such as high-energy consumption equipment emission sources corresponding to large furnaces, and low-energy consumption high-frequency equipment emission sources corresponding to small fans. The category characteristics section details its unique features, such as the high-power operation and large energy consumption of high-energy consumption equipment emission sources. The proportion is presented in a pie chart or table, such as high-energy consumption equipment emission sources accounting for 40%, and low-energy consumption high-frequency equipment emission sources accounting for 30%. The prediction curve section inserts an image and briefly explains it, such as "This curve shows that in the next year, with the expansion of production scale, the total carbon emissions will increase." Uniform font, font size, line spacing, and standard report layout.

[0255] The emission source classification algorithm based on density peak clustering includes the following steps:

[0256] Step SC21, standardize the collected data related to carbon emissions to ensure that data of different dimensions are on the same scale.

[0257] Step SC22, calculate the distance of all data points, first set the cutoff distance as the average value, count the proportion of data points with higher data density than the average value, and according to the mapping relationship between the proportion of data points with higher data density than the average value and the cutoff distance adjustment mode, determine the corresponding cutoff distance adjustment mode and adjust it.

[0258] Determine the cutoff distance adjustment mode and adjust it: according to the pre-set mapping relationship, find the cutoff distance adjustment mode corresponding to the proportion of data points with higher data density than the average value. For example, if the proportion is greater than 50%, the mapping relationship stipulates that the cutoff distance is increased by 10%; if the proportion is less than 30%, the cutoff distance is reduced by 15%. Adjust the cutoff distance according to the determined adjustment mode. Assuming that the current cutoff distance dc is 10, the proportion is 60%, and according to the mapping relationship, it needs to be increased by 10%, then the adjusted dc=10*(1+0.1)=11.

[0259] Step SC23, calculate the local density of each data point using the local density formula, which is as follows:

[0260] ;

[0261] Where, is the total number of data points, is the distance between data points and , is the previously determined cutoff distance, is the local density;

[0262] Step SC24, calculate the average local density of all data points, that is, the global density, that is, .

[0263] Step SC25, adjust the local density of each data point according to the global density, and the adjusted local density is as follows: , wherein, is the adjustment coefficient, is the adjusted local density.

[0264] Step SC26, calculate the distance between each data point and the nearest point in the data points with higher density than it , for the data point with the highest density, its value is set to the maximum value among all distances;

[0265] Step SC27, draw the local density of each data point and the calculated distance on a graph to form a scatter plot, and select the points in the upper right corner of the scatter plot as possible cluster centers.

[0266] Step SC28, for these possible cluster centers, calculate the fluctuation of the density of the data points around them. Select the point with the smallest fluctuation as the final cluster center.

[0267] Scatter plot: a graph that plots data in the form of points in a two-dimensional coordinate system. In this step, the local density is taken as the vertical axis and the distance is taken as the horizontal axis. Each data point corresponds to a point in the graph. The distribution of the scatter plot is used to intuitively analyze the characteristics of the data and assist in determining the cluster center.

[0268] Cluster center: in the clustering algorithm, the core data point representing a cluster. The method of selecting the point with the smallest fluctuation as the final cluster center is as follows: for each candidate cluster center, take its K nearest neighbors, calculate the standard deviation of the local density of the data points in the neighborhood, and select the candidate point with the smallest standard deviation as the final cluster center.

[0269] Step SC29, divide each data point into the category where the nearest cluster center is located, calculate the density similarity between each data point and the cluster center of the category it belongs to, if the density similarity between a data point and the cluster center of the category it belongs to is lower than a set threshold, mark the data point as an adjustment point, calculate the comprehensive distance between the adjustment point and other cluster centers, and reclassify it to the category where the nearest cluster center is located.

[0270] Step SC2A, for any two categories, calculate the similarity by using a pre-set similarity calculation formula that comprehensively considers the distance between the category centers and the density distribution of the data points within the categories, and the specific similarity calculation formula is as follows:

[0271] ;

[0272] wherein, represents the similarity between class A and class B, and the value range is generally between 0 and 1, and the closer the value is to 1, the more similar the two classes are; is a weight coefficient, used to adjust the importance of the class center distance factor in the similarity calculation, is also a weight coefficient, used to adjust the importance of the data dispersion degree factor in the similarity calculation, and ; is the distance between the center of class A and the center of class B, used to measure the closeness of the two classes in the spatial position, is the maximum value of all class center distances, used to normalize the class center distance, so that the calculation result is within a suitable range; is the standard deviation of the local density of the data points in class A, is the standard deviation of the local density of the data points in class B, is the larger value of and , used to normalize the standard deviation of the two classes.

[0273] Step SC2B, analyze whether the similarity of the two classes exceeds the preset similarity threshold. If yes, execute step SC2C; if no, execute step SC2D.

[0274] Preset similarity threshold: a reference value set in advance, used to determine whether the similarity of the two classes is high enough to decide whether to perform a merging operation on the two classes. The threshold is set according to the purpose of data analysis, the characteristics of the data, and the expectation of the clustering result.

[0275] Step SC2C, merge the two classes into one class, and for the new class after merging, recalculate the position of the clustering center, and recalculate the local density of each data point and the distance to the nearest point among the data points with higher density than it.

[0276] Class merging: in the emission source classification algorithm based on density peak clustering, the operation of integrating two emission source classes with high similarity (exceeding the preset threshold) into a new class, aiming to optimize the clustering result and make the class division more reasonable.

[0277] Distance to the nearest point among the data points with higher density than it: a value used to describe the relative position relationship between the data point and other high-density points. After class merging, the density ranking and distribution of the data points change, so the distance needs to be recalculated.

[0278] For example, originally there are category A and category B, the similarity is calculated as 0.9 (exceeding the preset threshold 0.8) through step SC2A, and step SC2C is executed. First, the data of categories A and B are merged, and the cluster center is recalculated. Assuming that the mean of category A is 50 and the mean of category B is 55 in the energy consumption dimension, the mean of the new category after merging is 52.5; then the local density of each data point and the distance of the nearest point among the data points with greater density than it are recalculated, the new category data update is completed, and accurate data is provided for subsequent clustering analysis.

[0279] Step SC2D, continue the subsequent steps.

[0280] Step SC2E, output the final emission source category division result.

[0281] In addition, the application also relates to a low-carbon concrete product production system, which has the following steps:

[0282] The coal-fired power generation system is located in the concrete production area, at the center of the concrete production area, and is equipped with coal storage facilities, coal conveying equipment, coal combustion treatment system, combustion equipment, power generation equipment, cooling equipment, gas communication equipment, dust collection equipment, and desulfurization and denitrification equipment.

[0283] The coal storage facilities are used for storing coal. The coal conveying equipment conveys the coal from the storage facilities to the coal treatment system; and conveys the coal powder generated after treatment to the combustion equipment and sprays the coal powder into the combustion equipment. The coal combustion treatment system includes coal crushing, screening, and grinding, which grinds the coal into fine powder. The combustion equipment is used for coal powder combustion. The power generation equipment converts the heat energy generated by coal combustion into mechanical energy and then into electrical energy, and is connected with the power transmission system. The cooling equipment is used for cooling the power generation equipment and the combustion equipment. The gas communication equipment connects the gas conveying pipeline and conveys the gas byproducts to the gas conveying pipeline. The dust collection equipment collects the fly ash in the flue gas generated by combustion and is connected with the belt conveying system. The desulfurization and denitrification equipment is used for removing sulfur oxides and nitrogen oxides in the flue gas.

[0284] The solid waste treatment system is located in the concrete production area and adjacent to the coal-fired power generation system, and includes fly ash screening equipment, fly ash storage facilities, fly ash conveying equipment, carbonization reaction device, temperature sensor, temperature control device, gas control system, and carbonized fly ash collection device.

[0285] The fly ash screening device is connected with the belt conveying system, and the fly ash is screened to remove large particle impurities. The particle size analyzer is used to monitor the particle size distribution of the screened fly ash in real time. The fly ash storage facility is used to store the screened fly ash. The fly ash conveying device is connected with the fly ash storage device, the fly ash carbonization reaction device, the carbonized fly ash collection device and the carbonized fly ash storage facility. The screened fly ash is conveyed from the fly ash storage facility to the fly ash carbonization reaction device, and the carbonized fly ash is conveyed from the carbonized fly ash collection device to the carbonized fly ash storage facility. The carbonization reaction device sprays the screened fly ash to contact with carbon dioxide to generate carbonization reaction. The temperature sensor is installed in the carbonization reaction device to monitor the temperature in the carbonization reaction process in real time. The temperature control device is used to regulate the temperature of the fly ash carbonization reaction in the carbonization reaction device. The gas control system is connected with the gas conveying pipeline to control the gas environment of the fly ash carbonization reaction device. The carbonized fly ash collection device is used to collect the carbonized fly ash in the fly ash carbonization reaction device and is connected with the belt conveying system.

[0286] The low-carbon concrete production system located in the concrete production area is adjacent to the solid waste treatment system, and includes a raw material storage facility, a low-carbon concrete batching system, a concrete mixing device, an alkali activator adding device, a concrete conveying device, an automatic pouring device, a vibration table, a mold moving device, a carbonation curing chamber, a multi-parameter sensor, a temperature and humidity control system, a gas control system, and a curing period control system.

[0287] The raw material storage facility is connected with the belt conveying system to store the raw materials of the carbonized fly ash and other low-carbon concrete products. The low-carbon concrete batching system calculates the amount of raw materials according to the strength requirement and weighs the low-carbon concrete raw materials according to the amount. The concrete mixing device is connected with the low-carbon concrete batching system, mixes the concrete raw materials, and automatically controls the mixing process. The alkali activator adding device is connected with the concrete mixing device, adds the alkali activator in the concrete mixing device, and automatically controls the adding time and amount. The data transmission interface wirelessly transmits the type and amount data of the weighed raw materials and the type of the alkali activator to the carbon emission intelligent accounting system. The concrete conveying device is connected with the concrete mixing device to convey the mixed concrete to the automatic pouring device. The automatic pouring device is connected with the concrete conveying device to uniformly pour the concrete into the mold. The vibration table is used to remove the air bubbles in the concrete to ensure the compactness of the concrete. The mold moving device moves the poured mold to the vibration table, and moves the vibrated mold to the carbonation curing chamber. The carbonation curing chamber is connected with the gas conveying pipeline to perform carbonation curing on the prefabricated member. The multi-parameter sensor is used to monitor the temperature and humidity data in the carbonation curing chamber in real time. The temperature and humidity control system is used to regulate the temperature and humidity in the carbonation curing chamber. The gas control system is connected with the gas conveying pipeline to control the gas environment of the carbonation curing chamber. The curing period control system is used to automatically control the residence time of the prefabricated member in the curing chamber.

[0288] A gas delivery system located in the concrete production area, including a pipe connector, a delivery pipe, an insulation layer, an expansion joint, a flow control valve, a carbon dioxide concentration sensor, a flow rate sensor, a data transmission interface, a recording and storage device, a cleaning device, an emergency shut-off valve, an alarm system.

[0289] The pipe connector is used to connect the delivery pipe with the gas communication equipment in the coal-fired power generation system, the gas control system in the solid waste treatment system, and the gas control system in the low-carbon concrete production system. The delivery pipe is used to deliver the gas generated by the coal-fired power generation system to the solid waste treatment system and the low-carbon concrete production system. The insulation layer is used to maintain the temperature of the gas in the delivery pipe. The expansion joint is used to compensate for the expansion and contraction of the delivery pipe due to temperature changes. The flow control valve is used to regulate the flow of gas in the delivery pipe. The carbon dioxide concentration sensor is used to measure the concentration of carbon dioxide in the gas output by the solid waste treatment system and the low-carbon concrete production system in real time. The flow rate sensor is used to measure the flow rate of the gas output by the solid waste treatment system and the low-carbon concrete production system in real time. The data transmission interface is used to wirelessly transmit the data collected by the carbon dioxide concentration sensor and the flow rate sensor to the carbon emission intelligent accounting system. The recording and storage device is used to record and store the carbon dioxide concentration, gas flow rate, and related operation and maintenance history data. The cleaning device is used to periodically clean the delivery pipe to prevent pipe blockage. The emergency shut-off valve quickly shuts off the gas flow when the delivery pipe fails. The alarm system: when the pressure, temperature or other parameters of the delivery pipe are abnormal, an alarm is issued.

[0290] A belt delivery system located in the concrete production area, including a delivery belt, a belt pulley, an electronic belt scale, a data transmission interface, a closed delivery channel, an observation window, a dust collection device, a light load sensor, a speed controller, a starting hopper, a terminal hopper, a material level sensor, a deviation prevention device, an anti-skid device, for delivering fly ash to the solid waste treatment system and carbonized fly ash to the low-carbon concrete production system.

[0291] The conveying belt is used to transport the fly ash generated by the coal-fired power generation system to the solid waste treatment system, and transport the carbonized fly ash generated by the solid waste treatment system to the low-carbon concrete production system. The pulley is used to drive the movement of the conveying belt and keep the tension of the conveying belt. The electronic belt scale is used to measure the mass of the fly ash transported by the belt. The data transmission interface is used to wirelessly transmit the data of the electronic belt scale to the carbon emission intelligent accounting system. The closed conveying channel is used to surround the conveying belt, reduce dust and loss during transportation. The observation window is used to observe the working state of the conveying belt without affecting the sealing effect. The dust collection device is used to collect the dust in the closed conveying channel. The light load sensor is used to monitor the load of the solid waste on the conveying belt, prevent the conveying belt from running empty or overloaded. The speed controller is used to adjust the speed of the conveying belt. The starting hopper is used to receive the fly ash in the dust collection device, accumulate the weight, and evenly distribute it on the conveying belt; receive the carbonized fly ash in the carbonized fly ash collection device, and evenly distribute it on the conveying belt. The terminal hopper is used to collect the fly ash transported to the solid waste treatment system; used to collect the carbonized fly ash transported to the low-carbon concrete production system. The level sensor is used to monitor the level of the starting hopper and the terminal hopper, and automatically control the transportation of solid waste. The anti-deviation device is used to automatically correct the running track of the conveying belt, prevent material leakage caused by belt deviation. The anti-skid device is used to increase the friction between the conveying belt and the solid waste, prevent slipping.

[0292] The power transmission system located in the concrete production area includes power transmission lines, substations, distribution devices, cables, relay protection devices, and power monitoring systems, which provide power for the low-carbon concrete production system and the carbon emission intelligent accounting system.

[0293] The power transmission lines are used to transport the power generated by the power generation equipment to the solid waste treatment system, the low-carbon concrete production system, and the belt conveying system. The substation is used to adjust the voltage and current in the power transmission line. The distribution device is used to distribute the power in the power transmission line to the power consumption equipment of each system. The cable connects the substation, distribution device, and power-driven equipment in the solid waste treatment system, low-carbon concrete production system, and belt conveying system. The relay protection device is used to monitor the running state of the power transmission line and distribution device, and quickly cut off the power supply when a fault occurs. The power monitoring system is used to monitor the voltage, current, and power factor of the power transmission system in real time, ensuring the stability and reliability of the power supply.

[0294] The power transmission line further includes a high-voltage power transmission line for long-distance transmission of high-voltage power. A low-voltage power transmission line is used for short-distance transmission of low-voltage power suitable for use in each system. An isolation device is used to isolate power of different voltage levels in the power transmission line. An emergency power supply provides backup power for critical equipment when the main power supply fails.

[0295] The embodiments of the present application are preferred embodiments of the present application, and are not intended to limit the protection scope of the present application, and thus: any equivalent changes made according to the structure, shape, principle of the present application should be covered within the protection scope of the present application.

Claims

1. A carbon emission control method, characterized in that: include: Obtain key parameter data for each link in the entire production system of low-carbon concrete products and perform data preprocessing; The parameters related to carbon emissions and adjustable are used as decision variables of the genetic algorithm; Within the value range of the decision variable, a preset number of individuals are randomly generated to form an initial population, and each individual represents a set of possible parameter combinations; According to the preset comprehensive evaluation indicators and the preset goals, a preset fitness function construction method is used to construct an adaptive fitness function; For each individual in the initial population, calculate its fitness value based on its parameter combination, according to the constructed fitness function, combined with the relevant data and calculation formula of the production system; The tournament selection method is used to select individuals from the current population to enter the next generation population based on their fitness values; Randomly select two individuals from the selected population as parents, perform a crossover operation according to the preset crossover probability, and perform a mutation operation on the individuals in the population with the preset mutation probability; Continuously repeat the fitness function calculation, selection, crossover and mutation operations to form a new population until the set iteration termination condition is reached. Then the iteration stops and the best individual in the current population is output as the optimized parameter combination. Apply the parameter combination obtained by genetic algorithm optimization to actual production equipment to control the production process in real time; After equipment adjustment, continuously collect key parameter data of each link of the production system; Based on the collected key parameter data, calculate the carbon emission equivalent according to the set carbon emission calculation formula. Carbon emission equivalent includes direct carbon emissions and indirect carbon emissions; Automatically generate carbon emission accounting report based on calculation results; Applying the parameter combination optimized by the genetic algorithm to actual production equipment to perform real-time control of the production process includes the following steps: The parameter combination optimized by the genetic algorithm is directly applied to the production equipment. After the parameter setting is completed, the production process is started; Calculate carbon emissions at each stage in real time based on a preset process analysis model, and dynamically adjust calculation parameters to suit production conditions using a preset adjustment strategy; The default adjustment strategy includes the following steps: Use the Kalman filter algorithm to fuse multi-source data in the production process, and apply the principal component analysis method to extract key indicators from the fused data; A hybrid prediction model combining a pre-trained long short-term memory network and support vector regression is used to predict key indicators under current production conditions and output the prediction results. Based on the forecast results and the uncertainties in the production process, a variety of production scenarios are generated using Monte Carlo simulation methods; Conduct risk assessments for each scenario using a risk quantification analysis method, calculating the probability of carbon emissions exceeding target values ​​and the potential economic losses under different scenarios; Taking carbon emission control, cost control and production efficiency improvement as multiple goals, a multi-objective optimization model was established, the optimal solution was obtained based on the multi-objective optimization model, and a production parameter adjustment plan was formulated; Assume carbon emissions , the cost is , the production efficiency is , the objective function can be expressed as: Minimize: ; Constraints: ; in, is the weight coefficient, which reflects the importance of each goal; is the carbon emission target, It is the cost budget, It is the minimum production efficiency requirement.

2. A carbon emission control method according to claim 1, characterized in that: According to the preset comprehensive evaluation indicators and the preset goals, the adaptive fitness function is constructed using the preset fitness function construction method, including: Construct the initial fitness function of each system based on the comprehensive evaluation indicators of each link of the entire production system and the preset goals; Initialize the population and encode each system decision variable into an individual; Calculate the fitness value of individuals under each goal, perform non-dominated sorting and crowding calculations, and select individuals to enter the next generation; After a preset round of iterations, the Pareto optimal solution set is obtained, providing multiple sets of trade-offs for decision-making; Use neural network algorithms to build models of each system and predict target values ​​under different combinations of decision variables; The weighted summation method is used to integrate the fitness functions of each system and construct the overall fitness function. The overall fitness function is as follows: ； in, Represents the overall fitness function value of the entire low-carbon concrete product production system, is an index variable used to identify different systems in the low-carbon concrete production system. Indicates the The weight coefficient of each system in the overall fitness function, It is The fitness function value of a system.

3. A carbon emission control method according to claim 1, characterized in that: The following steps are included after dynamically adjusting the calculation parameters to adapt to production conditions using a preset adjustment strategy: Outliers are eliminated through industrial-grade data verification algorithms, and the verified data is transmitted in real time to the next step of parameter adjustment; Use the PID control algorithm to compare the actual carbon emissions generated under the current production parameters with the carbon emission target; According to the difference analysis results, the parameters are adjusted using the preset parameter adjustment formula. The specific parameter adjustment formula is as follows: ； in, Represents the adjusted production parameter value, Indicates the production parameter value before adjustment, that is, the production parameter currently in use. is the dynamic adjustment coefficient, is the carbon emission deviation, is the time decay function; Combining real-time raw material prices, energy costs, and carbon trading market data, the comprehensive benefits of different parameter adjustment plans are quantified through a preset benefit formula. The specific benefit formula is as follows: ； in, Represents the comprehensive benefit value, The maximum carbon emission reduction that can be achieved is a preset reference value. It is the maximum possible cost increase and also the preset reference value. To reduce carbon emissions, is the cost increase, and is the weight coefficient; Determine the non-inferior solution through Pareto frontier analysis, and then screen out the optimal parameter combination based on the preset enterprise goals; Apply the adjusted parameters to production.

4. A carbon emission control method according to claim 1, characterized in that: The risk assessment is conducted for each scenario using a risk quantification analysis method. The calculation of the probability of carbon emissions exceeding the target value and the possible economic losses under different scenarios includes the following steps: Collect data covering energy consumption, equipment operation, raw materials, environment, and carbon trading market mechanism parameters; Taking equipment aging, energy usage, and raw material characteristics as key nodes, we build connections between them and construct a Bayesian network based on the statistical conditional probabilities between nodes based on historical data. According to the historical data characteristics of energy prices and production process parameters, an adaptive Copula function is selected, and the function parameters are determined using the maximum likelihood estimation method; The specific factor values ​​of each scenario are input into the Bayesian network and Copula function model, and Monte Carlo simulations are carried out more than a preset number of times. Sample weights are assigned according to the importance of the factors on carbon emissions. The number of times carbon emissions exceed the target value in the simulation is counted, and the probability of exceeding the target is calculated. First, the real-time price of the carbon trading market, the economic cost benchmark of environmental violations, the equipment status and rectification needs data are counted. Then, the carbon emission forecast results of the current scenario are input into the trained economic loss prediction model together with the above statistical data, and finally the economic losses that may be incurred under this scenario are output.

5. A carbon emission control method according to any one of claims 1 to 4, characterized in that: Based on the calculation results, the carbon emission accounting report is automatically generated, including: Collect various data related to carbon emissions and form a data set; An emission source classification algorithm based on density peak clustering is used to classify data points into different emission source categories and calculate the carbon emission contribution of each emission source; Analyze the unique characteristics of each emission source category and collect and integrate data related to its unique characteristics based on its unique characteristics; The principal component analysis method is used to extract key features from the fused data; Build an independent prediction model for each emission source category, and weight the output of each emission source category prediction model according to the proportion of carbon emissions to obtain the overall carbon emission prediction value; Integrate the emission source classification results, the characteristics and contribution ratio of each category, and the carbon emission trend forecast curve to form the main body of the report.

6. A carbon emission control method according to claim 5, characterized in that: The emission source classification algorithm based on density peak clustering includes the following steps: Standardize all collected data related to carbon emissions to ensure that data of different dimensions are on the same scale; Calculate the distances of all data points, first set the cutoff distance to the average value, count the percentage of data points with surrounding data density higher than the average, and determine the corresponding cutoff distance adjustment method based on the mapping relationship between the percentage of data points with surrounding data density higher than the average and the cutoff distance adjustment method; The local density formula is used to calculate the local density of each data point. The local density formula is as follows: ； in, is the total number of data points, is a data point and The distance between is the cutoff distance determined previously, is the local density; Calculate the average local density of all data points, that is, the global density, that is, ; According to the global density, the local density of each data point is adjusted. The adjusted local density is as follows: ,in, is the adjustment coefficient, is the adjusted local density; Calculate the distance between each data point and the nearest point among the data points with a higher density than it , for the data point with the highest density, The value is set to the maximum value among all distances; Plot the local density and calculated distance of each data point on a graph to form a scatter plot. In this scatter plot, select the point in the upper right corner as the possible cluster center. For these possible cluster centers, calculate the density fluctuation of the data points around them and select the point with the smallest fluctuation as the final cluster center; Classify each data point into the category of the cluster center closest to it, calculate the density similarity between each data point and the cluster center of its category, and if the density similarity between a data point and the cluster center of its category is lower than the set threshold, mark the data point as a point to be adjusted, calculate the comprehensive distance between the point to be adjusted and other cluster centers, and reclassify it into the category of the cluster center with the closest comprehensive distance; For any two categories, the similarity is calculated using a preset similarity calculation formula that comprehensively considers the distance between category centers and the density distribution of data points within the category. The specific similarity calculation formula is as follows: ； in, Indicates the similarity between category A and category B. The value range is generally between 0 and 1. The closer the value is to 1, the more similar the two categories are. is the weight coefficient, which is used to adjust the importance of factors affecting the category center distance in the similarity calculation. It is also the weight coefficient, which is used to adjust the importance of factors affecting the degree of data dispersion within the category in the similarity calculation, and ; is the distance between the center of category A and the center of category B, which is used to measure the proximity of the two categories in space. It is the maximum value of the distance between all category centers and is used to normalize the category center distance so that the calculation result is within an appropriate range; is the standard deviation of the local density of data points within category A, is the standard deviation of the local density of data points within category B, To obtain and The larger value of is used to normalize the standard deviations of the two categories; Analyze whether the similarity between two categories exceeds the preset similarity threshold; If yes, merge the two categories into one. Recalculate the cluster center position for the new category after the merger. Based on the new category composition, recalculate the local density of each data point and the distance to the nearest point with a higher density than it. If no, continue with the next steps; Output the final emission source classification results.

7. A low-carbon concrete product production system, characterized in that: The method according to any one of claims 1 to 6 comprises: A coal-fired power generation system located in the concrete production area, used to produce electricity, fly ash, and a mixture of carbon dioxide and water vapor; Solid waste treatment systems located in concrete production areas to generate carbonized fly ash; The low-carbon concrete production system located in the concrete production area uses carbonized fly ash, alkali activator, and raw materials to produce low-carbon concrete products; The gas delivery system located in the concrete production area is used to deliver the carbon dioxide and water vapor mixed gas to the solid waste treatment system and the carbonization curing room of the low-carbon concrete production system; A belt conveyor system located in the concrete production area is used to transport fly ash to the solid waste treatment system and carbonized fly ash to the low-carbon concrete production system; The power transmission system located in the concrete production area provides power for the low-carbon concrete product production system and the carbon emission intelligent accounting system.

Images