Energy flow and carbon flow monitoring, evaluating and optimizing system for aluminum electrolysis production process

Abstract

The invention discloses an energy flow and carbon flow monitoring, evaluating and optimizing system for an aluminum electrolysis production process, and belongs to the technical field of energy conservation, emission reduction and intelligent control in the metallurgical industry. The system design method comprises the following steps: designing an edge layer of the aluminum electrolysis production process energy flow and carbon flow control system, and inputting data collected by the edge layer into an energy flow-carbon flow coupling model in the cloud layer through the communication layer to execute analysis operation; an improved multi-objective genetic optimization algorithm is constructed in the cloud layer, and an optimal process parameter combination is obtained by optimizing an objective function; performing dynamic regulation and control, closed-loop feedback and exception handling operation through an execution mechanism of the edge layer; and performing comparative analysis and economic analysis on the designed aluminum electrolysis production process energy flow and carbon flow control system. Key parameters of an aluminum electrolysis process are monitored in real time through a multi-source data fusion technology, an energy efficiency-carbon emission correlation evaluation system is established in combination with big data analysis, and an electrolytic cell process parameter dynamic adjustment strategy is generated based on a self-adaptive optimization algorithm.

Description

Technical Field

[0001] The present invention belongs to the field of energy conservation, emission reduction and intelligent control technology in the metallurgical industry, and specifically relates to an energy flow and carbon flow monitoring, evaluation and optimization system for an aluminum electrolysis production process. Background Art

[0002] As a typical process industry with high energy consumption and high carbon emissions, aluminum electrolysis accounts for more than 65% of the energy consumption of the global non-ferrous metals industry. The current industry generally adopts a cell control system based on empirical rules. Process parameters (such as inter-electrode spacing and molecular ratio) mostly rely on manual settings or static model adjustments, with an adjustment cycle of up to 2-3 hours, resulting in an annual energy waste of approximately 50,000 kWh per cell. The resulting bottleneck is the separation of energy efficiency and carbon efficiency. Existing technologies focus on single-target optimization (such as reducing power consumption) and ignore the dynamic traceability of carbon footprint. This is especially true in areas with a high proportion of thermal power, where carbon emissions from electricity account for over 60%, but there is a lack of a regulatory mechanism associated with process parameters. Existing technologies are insufficiently data-driven. For example, traditional control relies on a few parameters such as cell voltage and temperature, and has weak capabilities for multi-dimensional data fusion and analysis such as anode current distribution and AlF3 concentration fluctuations, making it difficult to identify "high energy consumption-high emission" operating conditions in real time (such as local overheating leading to accelerated anode consumption). Existing technologies have a delayed dynamic response. For example, when faced with disturbances such as fluctuations in the thermal balance of the electrolytic cell and frequent anode effects, existing methods have long adjustment cycles (usually hours), resulting in cumulative energy waste and redundant carbon emissions.

[0003] While recent research has attempted to incorporate online monitoring, these models are limited to single-dimensional optimization of current efficiency and fail to establish a coupled energy-carbon flow analysis framework. Other technologies, while proposing energy consumption prediction algorithms, lack closed-loop control circuits, making it impossible to achieve dynamic self-correction of process parameters. To achieve the dual carbon goals, an integrated control system is urgently needed that integrates multi-source data perception, energy-carbon collaborative evaluation, and real-time optimization to overcome the core technological bottlenecks of green aluminum electrolysis production. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention provides an energy flow and carbon flow monitoring, evaluation and optimization system for an aluminum electrolysis production process.

[0005] To implement the above technology, the specific steps are as follows:

[0006] S1. Design an edge layer for the energy and carbon flow control system for the aluminum electrolysis production process, including a sensor network, an edge computing unit, and an actuator. After performing data acquisition operations through the sensor network, the edge computing unit performs data processing operations on the collected data. The actuator is used for local control.

[0007] The sensor network is used to perform data acquisition operations, and the network includes: electrical parameter sensors, thermal parameter sensors, chemical parameter sensors and carbon flow sensors;

[0008] The edge computing unit is used to perform data processing operations on the data collected by the sensor network, including: wavelet denoising operations, Kalman filtering operations and anomaly detection operations;

[0009] The actuator is used for local control, including: regulating the anode elevator and AlF3 adding device;

[0010] The electrical parameter sensors include: a Hall sensor with an error of ≤±0.1mV, used to measure the tank voltage within a preset time; a Rogowski coil current transformer, used to measure the series current; and a distributed optical fiber sensor, used to measure the anode current distribution.

[0011] Thermal parameter sensors include: infrared thermometer, used to measure the electrolyte temperature within a preset time; thermocouple, used to measure the tank shell temperature;

[0012] Chemical parameter sensors include: a LIBS laser spectrometer for measuring AlF3 concentration within a preset time; a conversion ratio (CR) for calculating the molecular ratio;

[0013] The carbon flow sensor includes a laser displacement sensor to measure anode consumption and a non-dispersive infrared (NDIR) to continuously monitor gas concentration at a fixed sampling frequency.

[0014] S2. Input the data collected by the edge layer into the energy flow-carbon flow coupling model in the cloud layer through the communication layer to perform analysis operations and obtain dynamic evaluation indicators;

[0015] The steps to perform an analysis operation are as follows:

[0016] S2.1. Construct an energy flow model including the input energy equation, the effective energy equation, and the heat loss equation, including the following steps:

[0017] S2.1.1. Considering the auxiliary system energy consumption, the input energy equation is constructed by collecting the series current and slot voltage and using the preset rectification efficiency. The expression is as follows:

[0018] Q in =I avg ×V cell ×t×η rect +∑P aux ×t

[0019] Where, I avg is the average value of the series current (kA); V cell is the cell voltage acquisition value; t is the acquisition time; η rect is the rectification efficiency; ∑P aux Energy consumption of auxiliary system;

[0020] S2.1.2. Construct the effective energy equation based on the aluminum output and the enthalpy of aluminum liquid formation. The expression is as follows:

[0021] Q Al =m Al ×ΔH Al

[0022] Where m Al is the aluminum production (kg / h); ΔH Al is the enthalpy of formation of molten aluminum;

[0023] S2.1.3. Based on the input energy equation and the effective energy equation, construct the heat loss equation, which is expressed as follows:

[0024] Q loss =Q in -Q Al -Q gas

[0025] Where Q gas To carry away the heat of the flue gas, it is calculated by the flue gas flow meter and temperature sensor;

[0026] S2.2. Construct a carbon flow tracking model including anode carbon consumption equation, carbon emission equation, and electricity carbon footprint equation, including the following steps:

[0027] S2.2.1. Based on the current efficiency and the anode excess consumption coefficient, the anode carbon consumption equation is constructed using the collected series currents. The expression is as follows:

[0028]

[0029] Where, CE is the current efficiency; α is the anode excess consumption coefficient;

[0030] S2.2.2. Based on the carbon oxidation rate, the emission mass of perfluorocarbons (PFCs), and the global warming potential of perfluorocarbons (PFCs), a carbon emission equation is constructed as follows:

[0031]

[0032] Where η os is the carbon oxidation rate; m PFC is the emission mass of perfluorocarbons (PFCs); is the molecular weight ratio of CO2 to C; GWP is the global warming potential of perfluorocarbon (PFC);

[0033] S2.2.3. Construct the electricity carbon footprint equation based on the input energy equation. The expression is as follows:

[0034] E CO2 =Q in ×fgrid

[0035] Where, f grid The real-time carbon emission factor of the power grid (kg-CO2 / kWh) is obtained through the API interface;

[0036] S2.3. Constructing dynamic evaluation indicators including the comprehensive electricity consumption equation per ton of aluminum, the carbon emission intensity equation per ton of aluminum, and the current efficiency equation, including the following steps:

[0037] S2.3.1. Construct the comprehensive power consumption equation for a ton of aluminum. The expression is as follows:

[0038]

[0039] S2.3.2. Construct the carbon emission intensity equation for a ton of aluminum. The expression is as follows:

[0040] C spec =(t-CO2eq / t-Al)

[0041] Where, t-CO2eq is carbon dioxide equivalent; t-Al is aluminum production (tons);

[0042] S2.3.3. Construct the current efficiency equation as follows:

[0043]

[0044] S3. Build an improved multi-objective genetic optimization algorithm (NSGA-II) in the cloud layer and construct an optimization objective function based on dynamic evaluation indicators. Input the optimization objective function into the improved multi-objective genetic optimization algorithm (NSGA-II) and set decision variables and constraints to obtain the optimal process parameter combination. The steps are as follows:

[0045] S3.1. Use the pre-trained LSTM model to predict the electrolyzer state, narrow the search space of the NSGA-II algorithm, and construct an improved multi-objective genetic optimization algorithm (NSGA-II);

[0046] The pre-trained LSTM model was used to predict the electrolytic cell state and narrow the search space of the NSGA-II algorithm. The specific method is as follows: a 2-layer LSTM (hidden units = 64) with Dropout = 0.2 was used as input: historical cell voltage, temperature, and molecular ratio data (time window = 24 hours) was used to output the electrolytic cell state prediction for the next hour to narrow the search space of the NSGA-II algorithm.

[0047] S3.2, construct the optimization objective function based on dynamic evaluation indicators, including minimizing the power consumption E spec , minimize carbon emissions C soecAnd maximize the current efficiency CE, use real number coding to construct the fitness function, that is, the objective function, which is expressed as follows:

[0048]

[0049] Where, E base is the reference value of power consumption; C base is the benchmark value of carbon emissions; CE base is the current efficiency; w1 is the weight of power consumption; w2 is the weight of carbon emissions; w3 is the weight of current efficiency;

[0050] S3.3. Set the decision variables, including pole distance, AlF3 addition rate, anode rise and fall frequency, and chromosome length;

[0051] S3.4. Set constraints, including using penalty function method to limit electrolyte temperature, molecular ratio, and anode effect coefficient; also include equipment constraints, namely anode lifting speed and AlF3 silo capacity limits;

[0052] S3.5. Input the optimization objective function into the improved multi-objective genetic optimization algorithm (NSGA-II) to obtain the optimal process parameter combination;

[0053] The population is initialized using an improved multi-objective genetic optimization algorithm (NSGA-II), and the optimization objective function is input to calculate the optimization objective function value. The Pareto frontier is selected through non-dominated sorting, and a new population is generated through selection, crossover, and mutation to obtain the initial solution. If the number of iterations is less than the maximum number of iterations, the optimization objective function value is recalculated. If the maximum number of iterations is reached, the optimal solution is output.

[0054] The optimal process parameter combination is selected from the non-dominated solution set by using the Pareto frontier screening method, that is, using the fuzzy membership method.

[0055] S4. An improved multi-objective genetic optimization algorithm (NSGA-II) based on the optimal process parameter combination analyzes and calculates the data collected by the edge layer sensor network, and performs dynamic control, closed-loop feedback and exception handling operations through the edge layer actuator;

[0056] Use actuators for dynamic regulation, including anode lifting control and AlF3 addition control;

[0057] Anode lifting control: The servo motor drives the vertical movement of the anode group (the pole pitch accuracy is ±0.5mm), adjusts the anode position to ensure pole pitch stability, and adopts feedforward-feedback composite control to compensate for pole pitch drift caused by anode consumption;

[0058] AlF3 addition is controlled by feeding through a screw feeder with an error of ≤±1%, which is used to cooperate with the vibrator to prevent material blockage. When the molecular ratio deviates from the target value by ±0.1, an emergency feeding program is triggered;

[0059] Closed-loop feedback: Data collection is performed every 5 minutes to generate KPIs (power consumption, carbon emissions, and current efficiency). The trigger condition is when the KPI deviates from the threshold (power consumption ±3%, carbon emissions ±5%, and current efficiency ±1%). The response time is to complete parameter correction within 30 seconds (instructions are issued via the OPC protocol).

[0060] The abnormal handling operation is as follows: if the electrolyte temperature exceeds the limit for three consecutive times, the emergency cooling system will be activated and the operator will be notified; when a sudden increase in PFC emissions (>10ppm) is detected, the electrode distance will be automatically reduced and an alumina covering layer will be injected.

[0061] S5. Conduct comparative analysis and economic analysis on the energy flow and carbon flow control system of the aluminum electrolysis production process with edge layer, cloud layer and communication layer.

[0062] Beneficial effects of the present invention:

[0063] The present invention uses full-factor carbon tracking, that is, incorporating anode consumption, PFC emissions and electricity carbon footprint into real-time optimization targets, breaking through the traditional limitation of focusing only on electricity consumption; by integrating mechanism models and data-driven models (LSTM trend prediction), the model generalization capability is improved; real-time control is guaranteed by the edge layer, and the cloud uses historical data to train adaptive optimization models and update weight coefficients; the temperature constraint range is automatically relaxed according to the aging degree of the electrolytic cell to extend the cell life.

[0064] After applying this system in a certain 300kA prebaked cell series, the following results were achieved: energy efficiency was improved: DC power consumption per ton of aluminum was reduced by 6.2% (12,800→12,000 kWh), with annual electricity savings exceeding 12 million yuan (electricity price 0.5 yuan / kWh); carbon emissions were reduced: carbon emission intensity per ton of aluminum decreased by 10.9% (12.8→11.4 t-CO2 eq), reducing carbon emissions by 75,000 tons annually; process optimization: current efficiency was increased to 92.8%, and the anode effect coefficient was reduced from 0.15 to 0.08 times / cell / day. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 is a flow chart of the steps of the present invention;

[0066] Figure 2 This is a system architecture diagram of the present invention;

[0067] Figure 3 This is a schematic diagram of real-time data collection of the present invention;

[0068] Figure 4It is a dynamic evaluation flow chart of the present invention;

[0069] Figure 5 It is the flow chart of the optimization algorithm of the present invention;

[0070] Figure 6 Schematic diagram of the feedback mechanism of the present invention. DETAILED DESCRIPTION

[0071] The present invention is further described in detail below with reference to specific embodiments.

[0072] like Figure 1 and Figure 2 As shown, a system for monitoring, evaluating and optimizing energy and carbon flows in an aluminum electrolysis production process includes the following steps:

[0073] S1, such as Figure 3 As shown in the figure, an edge layer of an aluminum electrolysis production process energy flow and carbon flow control system is designed with a sensor network, an edge computing unit, and an actuator. After performing data collection operations through the sensor network, the edge computing unit performs data processing operations on the collected data. The actuator is used for local control.

[0074] Specifically, the sensor network is used to perform data acquisition operations, and the network includes: electrical parameter sensors, thermal parameter sensors, chemical parameter sensors and carbon flow sensors;

[0075] Specifically, the edge computing unit is used to perform data processing operations on the data collected by the sensor network, including: wavelet denoising operations, Kalman filtering operations, and anomaly detection operations;

[0076] The edge computing unit hardware includes: a processor (e.g., CPU: ARM Cortex-A72, 8GB of memory) that supports at least Modbus / TCP and OPC UA protocols; a minimum of 512GB of memory (e.g., a 512GB NVMe SSD); communication interfaces: 4 x 1Gbps Ethernet, RS485, and CAN bus; the deployment process includes: installing an explosion-proof cabinet (IP54 protection level) in the control room of the electrolysis workshop; configuring a real-time operating system (e.g., ROS2Galactic); and writing a data acquisition driver (e.g., Python / C++) that supports the Modbus / TCP protocol.

[0077] Specifically, the actuator is used for local control, including: regulating the anode elevator and AlF3 adding device;

[0078] Furthermore, the electrical parameter sensors include: a Hall sensor with an error of ≤±0.1mV, used to measure the cell voltage within a preset time; a Rogowski coil current transformer, used to measure the series current; and a distributed optical fiber sensor, used to measure the anode current distribution;

[0079] In this embodiment, the Hall effect sensor has a range of 0-5V and an accuracy of ±0.05% FS (e.g., Honeywell CSLA2CD). The installation steps include: polishing and cleaning the surface of the cathode busbar of the electrolytic cell and applying conductive grease; securing the sensor with an insulating ceramic bracket to prevent short circuits with the cell shell; and connecting a shielded cable to the edge computing unit with a grounding resistance of ≤1Ω.

[0080] The Rogowski coil current transformer has a range of 0-500kA and an accuracy of ±0.5%;

[0081] The distributed fiber optic sensor has a spatial resolution of 5 cm and is installed on the anode guide rod. The installation steps are as follows: longitudinal grooves are cut along the anode guide rod to embed the optical fiber; high-temperature resistant glue (temperature resistance ≥ 300°C) is used to fix the optical fiber; and the optical signal demodulator is connected to the edge computing unit.

[0082] Furthermore, the thermal parameter sensor includes: an infrared thermometer for measuring the electrolyte temperature within a preset time; a thermocouple for measuring the tank shell temperature;

[0083] In this embodiment, the infrared thermometer has a range of 800-1000°C and a response time of 50ms (such as the Fluke 62Max+ infrared thermometer), and scans the surface temperature field of the melt in the tank every 30 seconds. The installation steps are as follows: open a Φ50mm observation hole on the top of the tank shell and install a quartz glass window; adjust the pitch angle of the infrared thermometer (30°-45°) to focus on the melt surface; calibrate the emissivity (set ε = 0.85, corresponding to molten electrolyte);

[0084] Thermocouples are installed on the outside of the tank shell. 36 thermocouples (e.g., K-type, meeting the accuracy of ±1°C) are used to construct a three-dimensional temperature field model.

[0085] Furthermore, the chemical parameter sensor includes: a LIBS laser spectrometer for measuring the AlF3 concentration within a preset time; and a conversion ratio (CR) for calculating the molecular ratio;

[0086] In this embodiment, the LIBS laser spectrometer analyzes the electrolyte composition online with an accuracy of ±0.01 mm (such as the Keyence LK-G5000 laser displacement sensor) and updates the data every 5 minutes. The installation steps are as follows: install a bracket on the top of the anode group to ensure that the laser irradiates the anode surface vertically; calibrate the initial height (H0) and calculate the real-time consumption ΔH = H0-H(t); combine the anode density (1.55 g / cm 3 ) converted carbon consumption rate;

[0087] Dynamic calculation of the ratio of Al2O3 to Na3AlF6 based on LIBS measurement by calculating the conversion ratio (CR);

[0088] Furthermore, the carbon flow sensor includes: a laser displacement sensor for measuring anode consumption; a non-dispersive infrared (NDIR) for continuously monitoring gas concentration at a fixed sampling frequency;

[0089] In this embodiment, a laser displacement sensor is installed on the top of the anode group with an accuracy of ±0.1 mm, which is used to calculate the real-time consumption rate by combining the anode height change and density parameters;

[0090] Non-dispersive infrared (NDIR) continuously monitors the concentrations of CO2 and CF4 (PFC representative) at a sampling frequency of 1 Hz.

[0091] S2, such as Figure 4 As shown, the data collected by the edge layer is input into the energy flow-carbon flow coupling model in the cloud layer through the communication layer to perform analysis operations and obtain dynamic evaluation indicators;

[0092] The communication layer protocols are: Industrial Ethernet (Modbus / TCP) and 5G wireless network. The requirements are: latency < 100ms, and data encryption is: TLS1.3;

[0093] Dynamic evaluation indicators include: comprehensive electricity consumption per ton of aluminum equation, carbon emission intensity equation per ton of aluminum equation and current efficiency equation;

[0094] The steps to perform an analysis operation are as follows:

[0095] S2.1. Construct an energy flow model including the input energy equation, the effective energy equation, and the heat loss equation, including the following steps:

[0096] S2.1.1. Considering the auxiliary system energy consumption, the input energy equation is constructed by collecting the series current and slot voltage and using the preset rectification efficiency. The expression is as follows:

[0097] Q in =I avg ×V cell ×t×η rect +∑P aux ×t

[0098] Where, I avg is the average value of the series current (kA); V cell is the cell voltage acquisition value; t is the acquisition time, which is collected every 5 minutes; η rectis the rectification efficiency, which is 96%. The DC power consumption is obtained by multiplying the series current, slot voltage acquisition value and acquisition time. The total energy actually obtained from the AC grid is obtained by multiplying the DC power consumption by the rectification efficiency, which is used to compensate for the energy loss in the rectification process. aux Energy consumption of auxiliary systems, including power consumption of auxiliary equipment such as fans and shelling machines;

[0099] S2.1.2. Construct the effective energy equation based on the aluminum output and the enthalpy of aluminum liquid formation. The expression is as follows:

[0100] Q Al =m Al ×ΔH Al

[0101] Where m Al is the aluminum production (kg / h); ΔH Al is the enthalpy of formation of molten aluminum, ΔH Al =8.2kWh / kg;

[0102] S2.1.3. Based on the input energy equation and the effective energy equation, construct the heat loss equation, which is expressed as follows:

[0103] Q loss =Q in -Q Al -Q gas

[0104] Where Q gas To carry away the heat of the flue gas, it is calculated by the flue gas flow meter and temperature sensor;

[0105] S2.2. Construct a carbon flow tracking model including anode carbon consumption equation, carbon emission equation, and electricity carbon footprint equation, including the following steps:

[0106] S2.2.1. Based on the current efficiency and the anode excess consumption coefficient, the anode carbon consumption equation is constructed using the collected series currents. The expression is as follows:

[0107]

[0108] Wherein, CE is the current efficiency; α is the anode excess consumption coefficient, the empirical value is 0.05-0.08, and 0.05 is taken in this embodiment;

[0109] S2.2.2. Based on the carbon oxidation rate, the emission mass of perfluorocarbons (PFCs), and the global warming potential of perfluorocarbons (PFCs), a carbon emission equation is constructed as follows:

[0110]

[0111] Where η oxis the carbon oxidation rate, which is 0.98; m PFC is the emission mass of perfluorocarbons (PFCs); is the molecular weight ratio of CO2 to C; GWP is the global warming potential of perfluorocarbon (PFC) (CF4=6630);

[0112] S2.2.3. Construct the electricity carbon footprint equation based on the input energy equation. The expression is as follows:

[0113] E CO2 =Q in ×f grid

[0114] Where, f grid The real-time carbon emission factor of the power grid (kg-CO2 / kWh) is obtained through the API interface;

[0115] S2.3. Constructing dynamic evaluation indicators including the comprehensive electricity consumption equation per ton of aluminum, the carbon emission intensity equation per ton of aluminum, and the current efficiency equation, including the following steps:

[0116] S2.3.1. Construct the comprehensive power consumption equation for a ton of aluminum. The expression is as follows:

[0117]

[0118] S2.3.2. Construct the carbon emission intensity equation for a ton of aluminum. The expression is as follows:

[0119] C spec =(t-CO2eq / t-Al)

[0120] Where, t-CO2eq is carbon dioxide equivalent; t-Al is aluminum production (tons);

[0121] S2.3.3. Construct the current efficiency equation as follows:

[0122]

[0123] S3. Build an improved multi-objective genetic optimization algorithm (NSGA-II) in the cloud layer and construct an optimization objective function based on dynamic evaluation indicators. Input the optimization objective function into the improved multi-objective genetic optimization algorithm (NSGA-II) and set decision variables and constraints to obtain the optimal process parameter combination. The steps are as follows:

[0124] S3.1. Use the pre-trained LSTM model to predict the electrolyzer state, narrow the search space of the NSGA-II algorithm, and construct an improved multi-objective genetic optimization algorithm (NSGA-II);

[0125] The pre-trained LSTM model was used to predict the electrolytic cell state and narrow the search space of the NSGA-II algorithm. The specific method is as follows: a 2-layer LSTM (hidden units = 64) with Dropout = 0.2 was used as input: historical cell voltage, temperature, and molecular ratio data (time window = 24 hours) was used to output the electrolytic cell state prediction for the next hour to narrow the search space of the NSGA-II algorithm.

[0126] S3.2, construct the optimization objective function based on dynamic evaluation indicators, including minimizing the power consumption E spec , minimize carbon emissions C spec And maximize the current efficiency CE, use real number coding to construct the fitness function, that is, the objective function, which is expressed as follows:

[0127]

[0128] Where, E base The base value of power consumption is 12800kWh / t; C base The benchmark value of carbon emissions is 12.8t-CO2 / t; CE base is the current efficiency, which is 90%; w1 is the weight of power consumption, which is 0.4; w2 is the weight of carbon emissions, which is 0.4; w3 is the weight of current efficiency, which is 0.2;

[0129] S3.3. Set the decision variables, including pole distance (10-50 mm), AlF addition rate (0.5-5 kg / h), anode lifting frequency (0.1-1 times / min), and chromosome length = 3;

[0130] S3.4. Set constraints, including using a penalty function to limit the electrolyte temperature (920-970°C), molecular weight ratio (2.0-2.5), and anode effect coefficient (<0.1 times / tank / day). Also included are equipment constraints, i.e., anode lift speed ≤ 20 mm / min and AlF3 silo capacity limits.

[0131] S3.5. Input the optimization objective function into the improved multi-objective genetic optimization algorithm (NSGA-II) to obtain the optimal process parameter combination;

[0132] like Figure 5 As shown in the figure, the population is initialized by the improved multi-objective genetic optimization algorithm (NSGA-II), and the optimization objective function is input to calculate the optimization objective function value; the Pareto frontier is selected by non-dominated sorting, and a new population is generated through selection, crossover and mutation to obtain the initial solution; if the number of iterations is less than the maximum number of iterations, the optimization objective function value is recalculated; if the maximum number of iterations is reached, the optimal process parameter combination is output;

[0133] In this embodiment, the maximum number of iterations is 500;

[0134] The optimal process parameter combination is selected from the non-dominated solution set by using the Pareto frontier screening method, that is, using the fuzzy membership method.

[0135] S4, such as Figure 6 As shown in the figure, the improved multi-objective genetic optimization algorithm (NSGA-II) based on the optimal process parameter combination analyzes and calculates the data collected by the edge layer sensor network, and performs dynamic control, closed-loop feedback and exception handling operations through the edge layer actuator;

[0136] Use actuators for dynamic regulation, including anode lifting control and AlF3 addition control;

[0137] Anode lifting control is as follows: the servo motor drives the anode group to move vertically (the pole distance adjustment accuracy is ±0.5mm), adjusts the anode position to ensure the stability of the pole distance, and adopts feedforward-feedback composite control to compensate for the pole distance drift caused by anode consumption; its control logic includes: target pole distance d target =40mm, allowable fluctuation ±0.5mm; PID parameter: proportional gain K p =1.2, integral gain K i =0.05, differential gain K d =0.1; the steps are:

[0138] Receive optimization instructions d target ;

[0139] Calculate the anode displacement Δd = d current –d target ;

[0140] Drive the servo motor (with a step accuracy of 0.01mm) to complete position adjustment; feedback the actual pole pitch to the edge computing unit;

[0141] AlF3 addition control is as follows: feeding through a screw feeder with an error of ≤±1%, which is used to cooperate with the vibrator to prevent material blockage. When the molecular ratio deviates from the target value by ±0.1, the emergency feeding program is triggered; its control logic is: target addition rate v target =2kg / h, error ≤±1%; screw feeder speed n(rpm) = 0.1×v target ; The steps are as follows:

[0142] Receive optimization instructions v target ;

[0143] Adjust the inverter output frequency to control the feeder speed;

[0144] Feedback of actual feeding amount through weighing sensor;

[0145] If the deviation is >2%, the vibrator clearing procedure is triggered;

[0146] Closed-loop feedback: Data collection is performed every 5 minutes to generate KPIs (power consumption, carbon emissions, and current efficiency). The trigger condition is when the KPI deviates from the threshold (power consumption ±3%, carbon emissions ±5%, and current efficiency ±1%). The response time is to complete parameter correction within 30 seconds (instructions are issued via the OPC protocol).

[0147] Abnormal handling operations are as follows: if the electrolyte temperature exceeds the limit three times in a row, the emergency cooling system is activated and the operator is notified; if a sudden increase in PFC emissions (>10ppm) is detected, the electrode distance is automatically reduced and an alumina overlay is injected;

[0148] The exception handling cases in this embodiment include Case 1 and Case 2;

[0149] Case 1: Electrolyte temperature exceeds the limit;

[0150] Phenomenon: The temperature of A07 tank is >970℃ for three consecutive times;

[0151] System response: automatically reduce the electrode distance by 5mm, increase the amount of AlF3 added by 20%, and the temperature returns to 950℃ within 30 minutes;

[0152] Case 2: Sudden increase in PFC emissions;

[0153] Phenomenon: CF4 concentration in A12 tank>15ppm;

[0154] System response: Alumina overlay was injected (thickness increased by 2 cm), the inter-electrode distance was reduced by 3 mm, and the concentration dropped to 5 ppm within 1 hour.

[0155] S5. Conduct comparative analysis and economic analysis of the energy flow and carbon flow control system for the aluminum electrolysis production process with the edge layer, cloud layer, and communication layer;

[0156] After the present invention applies this system in a certain 300kA pre-baked tank series:

[0157] Improved energy efficiency: DC power consumption per ton of aluminum decreased by 6.2% (12,800 → 12,000 kWh), with annual electricity savings exceeding 12 million yuan (at an electricity price of 0.5 yuan / kWh);

[0158] Carbon emission reduction: The carbon emission intensity per ton of aluminum decreased by 10.9% (12.8 → 11.4 t-COeq), reducing carbon emissions by 75,000 tons annually;

[0159] Process optimization: Current efficiency increased to 92.8%, and the anode effect coefficient decreased from 0.15 to 0.08 times / cell / day.

[0160] In this embodiment, the optimal process parameter combination obtained is compared with the existing technology. The test object for comparative analysis is: 300kA prebaked anode electrolytic cell series (cell numbers A01-A50) of an electrolytic aluminum enterprise; the test period is: January to March 2023 (the full production cycle); the comparison group is: traditional PID control group (cell numbers B01-B50);

[0161] The results are shown in Table 1 below:

[0162] Table 1: Comparison of results

[0163] index Experimental group (the present invention) Control group (traditional PID) Improvement effect DC power consumption per ton of aluminum 12050kWh 12800kWh ↓5.9％ Carbon emissions per ton of aluminum <![CDATA[11.5t-CO2eq]]> <![CDATA[12.8t-CO2eq]]> ↓10.2％ Current efficiency 92.5％ 90.1％ ↑2.4 percentage points Anode effect coefficient 0.08 times / slot / day 0.15 times / slot / day ↓46.7％

[0164] As shown in the table above, the DC power consumption per ton of aluminum of the present invention is reduced by 5.9% compared with the traditional PID, the carbon emissions per ton of aluminum are reduced by 10.2% compared with the traditional PID, the current efficiency is improved by 2.4% compared with the traditional PID, and the anode effect coefficient is reduced by 46.7% compared with the traditional PID.

[0165] In addition, the economic analysis of the invention is conducted, including investment cost and annual income;

[0166] The investment costs are shown in Table 2;

[0167] Table 2: Investment costs

[0168] project Cost (10,000 yuan) sensor networks 120 Edge computing unit 80 Software System 50 Installation and debugging 30 total 280

[0169] The annual returns are shown in Table 3;

[0170] Table 3: Annual Returns

[0171] project Income (10,000 yuan / year) Power saving benefits 1280 (25.6 million kWh of electricity saved × 0.5 yuan / kWh) Carbon trading revenue 750 (75,000 tons of emission reduction × 100 yuan / ton) Anode saving 300 (anode consumption reduced by 8%) total 2330

[0172] Payback period: 280 / 2330≈0.12 years (about 1.5 months). Sensitivity analysis shows that when the electricity price fluctuates by ±20%, the payback period varies from 1.2 to 1.8 months.

[0173] Although the present invention has been shown and described through embodiments, it should be understood by those skilled in the art that various modifications, variations, or substitutions may be made to these embodiments without departing from the principles and spirit of the present invention. Therefore, the scope of protection of the present invention shall be defined by the appended claims and their equivalents.

Claims

1. A system for monitoring, evaluating and optimizing energy and carbon flows in an aluminum electrolysis production process, characterized in that: The following steps are involved: S1. Design an edge layer for the energy and carbon flow control system for the aluminum electrolysis production process, including a sensor network, an edge computing unit, and an actuator. After performing data acquisition operations through the sensor network, the edge computing unit performs data processing operations on the collected data. The actuator is used for local control. S2. Input the data collected by the edge layer into the energy flow-carbon flow coupling model in the cloud layer through the communication layer to perform analysis operations and obtain dynamic evaluation indicators; The analysis operation is to obtain dynamic evaluation indicators by constructing an energy flow model including an input energy equation, an effective energy equation, and a heat loss equation, and constructing a carbon flow tracking model including an anode carbon consumption equation, a carbon emission equation, and an electricity carbon footprint equation; The dynamic evaluation indicators include: comprehensive power consumption equation per ton of aluminum, carbon emission intensity equation per ton of aluminum and current efficiency equation; S3. Construct an improved multi-objective genetic optimization algorithm in the cloud layer and construct an optimization objective function based on dynamic evaluation indicators; input the optimization objective function into the improved multi-objective genetic optimization algorithm and set decision variables and constraints to obtain the optimal process parameter combination; S4. An improved multi-objective genetic optimization algorithm based on the optimal process parameter combination analyzes and calculates the data collected by the edge layer sensor network, and performs dynamic regulation, closed-loop feedback, and exception handling operations through the edge layer actuator to achieve real-time control of energy flow and carbon flow in the aluminum electrolysis production process; S5. Conduct comparative analysis and economic analysis on the energy flow and carbon flow control system of the aluminum electrolysis production process with edge layer, cloud layer and communication layer.

2. The energy flow and carbon flow monitoring, evaluation and optimization system for an aluminum electrolysis production process according to claim 1, characterized in that: The steps of inputting the data collected by the edge layer through the communication layer into the energy flow-carbon flow coupling model in the cloud layer to perform analysis operations and obtain the dynamic evaluation index are as follows: S2.

1. Construct an energy flow model including the input energy equation, the effective energy equation, and the heat loss equation, including the following steps: S2.1.

1. Considering the auxiliary system energy consumption, the input energy equation is constructed by collecting the series current and slot voltage and using the preset rectification efficiency. The expression is as follows: Q in =I avg ×V cell ×t×η rect +∑P aux ×t Where, I avg is the average value of the series current; V cell is the cell voltage acquisition value; t is the acquisition time; η rect is the rectification efficiency; ∑P aux Energy consumption of auxiliary system; S2.1.

2. Construct the effective energy equation based on the aluminum output and the enthalpy of aluminum liquid formation. The expression is as follows: Q Al =m Al ×Δh Al Where m Al is the aluminum production; ΔH Al is the enthalpy of formation of molten aluminum; S2.1.

3. Based on the input energy equation and the effective energy equation, construct the heat loss equation, which is expressed as follows: Q loss =Q in -Q Al -Q gas Where Q gas Remove heat from flue gas; S2.

2. Construct a carbon flow tracking model including anode carbon consumption equation, carbon emission equation, and electricity carbon footprint equation, including the following steps: S2.2.

1. Based on the current efficiency and the anode excess consumption coefficient, the anode carbon consumption equation is constructed using the collected series currents. The expression is as follows: Where, CE is the current efficiency; α is the anode excess consumption coefficient; S2.2.

2. Based on the carbon oxidation rate, the emission mass of perfluorocarbons, and the global warming potential of perfluorocarbons, a carbon emission equation is constructed as follows: Where η ox is the carbon oxidation rate; m PFC is the emission mass of perfluorocarbons; is the molecular weight ratio of CO2 to C; GWP is the global warming potential of perfluorocarbons; S2.2.

3. Construct the electricity carbon footprint equation based on the input energy equation. The expression is as follows: E CO2 =Q in ×f grid Where, f grid is the real-time carbon emission factor for the power grid; S2.

3. Constructing dynamic evaluation indicators including the comprehensive electricity consumption equation per ton of aluminum, the carbon emission intensity equation per ton of aluminum, and the current efficiency equation, including the following steps: S2.3.

1. Construct the comprehensive power consumption equation for a ton of aluminum. The expression is as follows: S2.3.

2. Construct the carbon emission intensity equation for a ton of aluminum. The expression is as follows: C spec =(t-CO2eq / t-Al) Where, t-CO2eq is carbon dioxide equivalent; t-Al is aluminum production (tons); S2.3.

3. Construct the current efficiency equation as follows:

3. The energy flow and carbon flow monitoring, evaluation and optimization system for an aluminum electrolysis production process according to claim 1, characterized in that: The improved multi-objective genetic optimization algorithm is constructed in the cloud layer, and the optimization objective function is constructed based on the dynamic evaluation index; the optimization objective function is input into the improved multi-objective genetic optimization algorithm and the decision variables and constraints are set to obtain the optimal process parameter combination. The steps are as follows: S3.

1. Use the pre-trained LSTM model to predict the electrolyzer state, narrow the search space of the multi-objective genetic optimization algorithm, and construct an improved multi-objective genetic optimization algorithm; The pre-trained LSTM model was used to predict the electrolytic cell state and narrow the search space of the NSGA-II algorithm. The specific method is as follows: a two-layer LSTM with Dropout = 0.2 was used as input: historical cell voltage, temperature, and molecular ratio data, and the output was a prediction of the electrolytic cell state for the next hour, which was used to narrow the search space of the NSGA-II algorithm. S3.2, construct the optimization objective function based on dynamic evaluation indicators, including minimizing the power consumption E spec , minimize carbon emissions C spec And maximize the current efficiency CE, use real number coding to construct the fitness function, that is, the objective function, which is expressed as follows: Where, E base is the reference value of power consumption; C base is the benchmark value of carbon emissions; CE base is the current efficiency; w1 is the weight of power consumption; w2 is the weight of carbon emissions; w3 is the weight of current efficiency; S3.

3. Set the decision variables, including pole distance, AlF3 addition rate, anode rise and fall frequency, and chromosome length; S3.

4. Set constraints, including using penalty function method to limit electrolyte temperature, molecular ratio, and anode effect coefficient; also include equipment constraints, namely anode lifting speed and AlF3 silo capacity limits; S3.

5. Input the optimization objective function into the improved multi-objective genetic optimization algorithm to obtain the optimal process parameter combination.

4. The energy flow and carbon flow monitoring, evaluation and optimization system for an aluminum electrolysis production process according to claim 1, characterized in that: The improved multi-objective genetic optimization algorithm based on the optimal process parameter combination analyzes and calculates the data collected by the edge layer sensor network, and uses the edge layer actuator to perform dynamic regulation, closed-loop feedback and abnormal processing operations, using the actuator to perform dynamic regulation, including: anode lifting control and AlF3 addition control; Anode lifting control: The servo motor drives the vertical movement of the anode group to adjust the anode position to ensure the stability of the pole distance. Feedforward-feedback composite control is used to compensate for the pole distance drift caused by anode consumption. AlF3 addition is controlled by feeding through a screw feeder with an error of ≤±1%, which is used to cooperate with the vibrator to prevent material blockage. When the molecular ratio deviates from the target value by ±0.1, an emergency feeding program is triggered; The closed-loop feedback system collects data every five minutes to generate KPIs. The trigger condition is when the KPI deviates from the threshold. The response time is to complete parameter correction within 30 seconds. The abnormal handling operation is as follows: if the electrolyte temperature exceeds the limit for three consecutive times, the emergency cooling system will be activated and the operator will be notified; when a sudden increase in PFC emissions is detected, the electrode distance will be automatically reduced and an alumina covering layer will be injected.

Images