Off-grid hydrogen production and flexible chemical coupling-based target optimization scheduling method and system

Abstract

The application discloses a target optimization scheduling method and system based on off-grid hydrogen production and flexible chemical coupling, relates to the technical field of energy management, and comprises the following steps: analyzing a standardized data set by using a linear programming model, generating an initial hydrogen production strategy, and adjusting hydrogen production operation parameters of the initial hydrogen production strategy according to a predicted demand to generate an optimized hydrogen production strategy; identifying demand fluctuation characteristics according to the optimized hydrogen production strategy, formulating short-term and long-term energy scheduling schemes respectively, combining the two schemes, and generating a comprehensive scheduling scheme; performing energy distribution and scheduling on the comprehensive scheduling scheme, monitoring an operation state and an energy distribution process in real time, generating real-time operation data and scheduling feedback; adjusting energy distribution and hydrogen production operation parameters according to the real-time operation data and the scheduling feedback, generating an optimal scheduling scheme, performing comprehensive performance evaluation, and generating an optimization scheduling effect report. The application realizes optimization of hydrogen production efficiency and energy utilization rate.

Description

A target optimization scheduling method and system based on the coupling of off-grid hydrogen production and flexible chemical industry Technical Field

[0001] This invention relates to the field of energy management technology, and in particular to a target optimization scheduling method and system based on the coupling of off-grid hydrogen production and flexible chemical engineering. Background Technology

[0002] With the rapid development of renewable energy and the advancement of the "dual carbon" target, hydrogen energy, as a clean, efficient, and renewable secondary energy source, is gradually becoming an important pillar of the future energy system. Off-grid hydrogen production using intermittent energy sources such as wind and solar power can effectively achieve localized energy utilization and emission reduction targets. Meanwhile, flexible chemical processes (such as methanol synthesis, ammonia production, and olefin conversion) have become important targets for coupling with hydrogen production due to their adjustable load and flexible energy consumption. Currently, research on the coupling of off-grid hydrogen production and flexible chemical processes mainly focuses on two levels: energy coupling and material matching, relying on optimization models, process simulations, and data-driven algorithms to improve energy efficiency.

[0003] For the energy coupling process of off-grid hydrogen production and flexible chemical engineering, existing technologies typically employ static or semi-dynamic optimization models, using algorithms such as linear and nonlinear programming to allocate and schedule energy under specific operating conditions. However, these models often rely solely on static input parameters, neglecting the feedback mechanism of real-time operational data, making it difficult to achieve dynamic coordination between energy and chemical processes across multiple time scales. Furthermore, in complex coupled systems, the high heterogeneity, high dimensionality, and significant temporal variations of data make it difficult for existing scheduling models to form reusable standardized datasets, impacting the accuracy and adaptability of optimization algorithms. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a target optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical industry to solve the problems of uneven energy distribution and insufficient dynamic scheduling response in the coupling of off-grid hydrogen production and chemical industry.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] Firstly, this invention provides a target optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical engineering, which includes,

[0008] Collect raw data and preprocess it to generate a standardized dataset;

[0009] Using a linear programming model, a standardized dataset is analyzed to generate an initial hydrogen production strategy. Based on predicted demand, the hydrogen production operation parameters of the initial hydrogen production strategy are adjusted to generate an optimized hydrogen production strategy.

[0010] Based on the optimized hydrogen production strategy, the characteristics of demand fluctuations are identified, and short-term and long-term energy dispatching schemes are formulated and combined to generate a comprehensive dispatching scheme.

[0011] The system performs energy allocation and scheduling for the integrated scheduling scheme, monitors the operation status and energy allocation process in real time, and generates real-time operation data and scheduling feedback.

[0012] Based on real-time operational data and scheduling feedback, energy allocation and hydrogen production operating parameters are adjusted to generate the optimal scheduling scheme, and a comprehensive performance evaluation is conducted to generate an optimized scheduling effect report.

[0013] As a preferred embodiment of the target optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical engineering described in this invention, the specific steps for generating the standardized dataset are as follows:

[0014] Based on the original data, identify and remove outliers, missing values, and noisy data to generate a clean dataset;

[0015] The clean dataset is merged and aligned with the external environment data to generate a merged dataset;

[0016] Performance metrics are extracted from the merged dataset and standardized to generate a standardized dataset.

[0017] As a preferred embodiment of the target optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical industry as described in this invention, the specific steps of generating the initial hydrogen production strategy are as follows:

[0018] Based on the input parameters of the standardized dataset, define decision variables and construct the objective function to generate a linear programming model;

[0019] Based on resource constraints and capacity, constraints are set, and the linear programming model is solved using the simplex method to obtain the optimal energy allocation, hydrogen production rate, and load regulation parameters.

[0020] Based on the optimal energy allocation, hydrogen production rate, and load adjustment parameters, production parameters are extracted from the linear programming model, and unit conversion and time alignment are performed to generate an initial hydrogen production strategy.

[0021] As a preferred embodiment of the target optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical industry described in this invention, the predicted demand is obtained by using a regression model to predict demand based on historical data, changes in the external environment, and real-time operating status, and by analyzing the relationship between historical data and related factors.

[0022] As a preferred embodiment of the target optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical industry as described in this invention, the specific steps for generating the optimized hydrogen production strategy are as follows:

[0023] Based on forecasted demand, analyze future production demand and energy consumption to generate optimization targets;

[0024] The initial hydrogen production strategy was evaluated, the differences between the strategy and the optimization objectives were analyzed, and an evaluation report was generated.

[0025] Based on the predicted demand and assessment report, the hydrogen production operation parameters in the initial hydrogen production strategy are adjusted to generate the adjusted hydrogen production operation parameters.

[0026] The adjusted hydrogen production operating parameters are input into a linear programming model to optimize the production process and generate an optimized hydrogen production strategy.

[0027] As a preferred embodiment of the target optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical industry as described in this invention, the specific steps for generating the comprehensive scheduling scheme are as follows:

[0028] Based on the optimized hydrogen production strategy, short-term and long-term demand fluctuations are analyzed, the relationship between energy supply and production is identified, and demand fluctuation analysis results are generated.

[0029] Based on the results of demand fluctuation analysis, optimize energy allocation and load adjustment for each production cycle to generate a short-term energy dispatch plan;

[0030] Based on the short-term energy dispatch plan and energy supply, analyze the long-term demand trend, energy supply capacity and production load to generate a long-term energy dispatch plan.

[0031] By combining short-term and long-term energy dispatch schemes, energy allocation and load regulation are integrated to generate a comprehensive dispatch scheme.

[0032] As a preferred embodiment of the target optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical engineering described in this invention, the specific steps for generating real-time operating data and scheduling feedback are as follows:

[0033] The energy allocation and scheduling of the comprehensive scheduling plan are carried out, the scheduling operations of the production process are initiated, and a preliminary scheduling execution record is generated;

[0034] Real-time monitoring of the initial scheduling execution records generates real-time operational data.

[0035] Based on real-time operational data, monitor the energy allocation for each production cycle and generate energy allocation feedback;

[0036] By combining real-time operational data with energy allocation feedback and performing deviation analysis, scheduling feedback is generated.

[0037] As a preferred embodiment of the target optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical engineering described in this invention, the specific steps for generating the optimal scheduling scheme are as follows:

[0038] Based on real-time operational data and scheduling feedback, analyze the current gap between energy allocation and production efficiency, and generate a gap analysis report;

[0039] Based on the gap analysis report, energy allocation and hydrogen production operation parameters were adjusted and combined with the initial hydrogen production strategy to generate the optimal scheduling scheme.

[0040] As a preferred embodiment of the target optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical industry as described in this invention, the specific steps for generating the optimization scheduling effect report are as follows:

[0041] Based on the execution status of the optimal scheduling scheme, collect data on energy usage, production efficiency, and operating status to generate an execution dataset;

[0042] Based on the execution dataset, energy consumption, production efficiency, and cost control are evaluated and compared with optimization objectives to generate performance evaluation results;

[0043] Analyze the performance evaluation results, identify the gap between the actual execution and the optimization target, and generate an optimization scheduling effect report.

[0044] Secondly, this invention provides a target optimization scheduling system based on the coupling of off-grid hydrogen production and flexible chemical engineering, including,

[0045] The preprocessing module is used to collect raw data, perform preprocessing, and generate standardized datasets;

[0046] The strategy optimization module is used to analyze the standardized dataset using a linear programming model, generate an initial hydrogen production strategy, and adjust the hydrogen production operation parameters of the initial hydrogen production strategy according to the predicted demand to generate an optimized hydrogen production strategy.

[0047] The scheduling construction module is used to identify demand fluctuation characteristics based on the optimized hydrogen production strategy, formulate short-term and long-term energy scheduling schemes respectively, and combine them to generate a comprehensive scheduling scheme.

[0048] The real-time monitoring module is used to allocate and schedule energy for the integrated scheduling scheme, and to monitor the operating status and energy allocation process in real time, generating real-time operating data and scheduling feedback.

[0049] The performance evaluation module is used to adjust energy allocation and hydrogen production operation parameters based on real-time operation data and scheduling feedback, generate the optimal scheduling plan, conduct a comprehensive performance evaluation, and generate an optimized scheduling effect report.

[0050] The beneficial effects of this invention are as follows: by utilizing linear programming models and simplex optimization, the optimal energy allocation and production strategy can be efficiently determined under multiple constraints. This not only optimizes the production capacity under limited resources, but also balances load regulation and energy efficiency, thereby reducing energy consumption and production costs. It provides a precise basis for subsequent scheduling and adjustment, and achieves the optimization of hydrogen production efficiency and energy utilization. Attached Figure Description

[0051] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 is a flowchart of the target optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical engineering.

[0053] Figure 2 is a schematic diagram of a target optimization scheduling system based on the coupling of off-grid hydrogen production and flexible chemical engineering.

[0054] Figure 3 is a flowchart of the optimized hydrogen production strategy.

[0055] Figure 4 is a flowchart of the integrated scheduling scheme generation process. Detailed Implementation

[0056] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0057] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0058] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0059] Referring to Figures 1-4, an embodiment of the present invention is provided, which offers a target optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical engineering, including the following steps:

[0060] S1. Collect raw data and preprocess it to generate a standardized dataset.

[0061] S1.1 Based on the original data, identify and remove outliers, missing values, and noisy data to generate a clean dataset.

[0062] Specifically, the raw data includes hydrogen production process data, chemical process data, energy supply data, and external environmental data. The 3σ criterion is used to statistically analyze the numerical distribution of the raw data, calculating the mean and standard deviation, identifying and removing outliers exceeding the normal range. Missing items are filled using mean imputation, replacing missing data by calculating the historical average of the corresponding field. Moving averages are used to smooth the time-series data, removing high-frequency noise and random fluctuations. The processed raw data is then sorted by time and formatted to ensure continuity and consistency, generating a clean dataset.

[0063] It should also be noted that outliers refer to values ​​that deviate from the overall distribution pattern of the original data and exceed the normal fluctuation range. They are generally defined as data that exceed the mean by ±3 times the standard deviation. Historical averages refer to the average value calculated based on historical valid data of the same field. Usually, the mean of the field in the time series is taken as the reference value for interpolation or comparison.

[0064] S1.2. Merge and align the clean dataset with the external environment data to generate a merged dataset.

[0065] Specifically, a time series alignment method is used to synchronize data from different sources in the clean dataset according to timestamps, unifying the time resolution to the same sampling period (e.g., every 5 minutes or hour) and performing linear interpolation to fill in missing moments; a field mapping method is used to unify the field names, units, and formats of data from different sources, converting similar indicators (e.g., energy consumption, production, or load data) into unified units of measurement and standardizing field names; Kalman filtering, Bayesian fusion, and multivariate interpolation are used to merge the hydrogen production process data, chemical process data, and energy supply data after time alignment and field unification, establishing index relationships based on timestamps and data sources to generate a unified format dataset; the unified format dataset is then correlated and matched with external environmental data according to time and spatial dimensions, and correlation coefficients or collaborative change trends are calculated through multi-source data fusion analysis to extract information on the impact of environmental changes on energy and production indicators, generating a fused dataset.

[0066] It should also be noted that external environmental data refers to real-time information reflecting the external conditions of hydrogen production and chemical engineering coupled operation, including light intensity, wind speed, temperature, humidity, electricity price, market demand, and maintenance cycle, etc., used to describe the impact of external factors on energy supply and production load.

[0067] S1.3 Extract performance metrics from the merged dataset and perform standardization to generate a standardized dataset.

[0068] Specifically, the merged dataset undergoes field filtering to extract performance indicators that reflect hydrogen production efficiency, energy consumption level, production load, and energy utilization rate. Feature extraction is used to quantify these performance indicators, transforming qualitative information into calculable quantitative data, such as representing operating status numerically. Range standardization is employed to normalize performance indicators of different dimensions, adjusting their value ranges to a unified interval. Finally, the standardized performance indicator data undergoes format standardization and time sequence verification to ensure structural integrity and comparability, generating a standardized dataset.

[0069] S2. Using a linear programming model, analyze the standardized dataset to generate an initial hydrogen production strategy. Based on the predicted demand, adjust the hydrogen production operation parameters of the initial hydrogen production strategy to generate an optimized hydrogen production strategy.

[0070] S2.1. Based on the input parameters of the standardized dataset, define decision variables, construct the objective function, and generate a linear programming model.

[0071] Specifically, based on the input parameters of the standardized dataset, quantifiable information such as energy consumption, production demand, raw material consumption, time interval, and capacity limit are analyzed item by item according to the field list. Using variable naming rules, energy allocation, hydrogen production rate, start-up / shutdown status, and period load are defined as decision variables for the linear programming model. Upper and lower bounds and feasible value sets are set for each decision variable according to the value range of the standardized dataset. An objective function coefficient vector is constructed based on the cost coefficient, energy consumption coefficient, and efficiency weight in the standardized dataset, clearly defining the objective as a linear form that minimizes cost or maximizes energy efficiency (e.g., accumulating energy consumption cost and operating cost over time periods), and establishing a one-to-one correspondence between the objective function and decision variables. Consistency checks are performed on the units and time granularities involved in the standardized dataset. The data structure of the linear programming model is assembled according to the decision variable index and objective function coefficients, outputting a linear programming model containing the set of decision variables, variable bounds, and objective function coefficients.

[0072] It should also be noted that variable naming rules refer to assigning uniform and identifiable symbols or identifiers to different types of input parameters based on the data attributes and meanings in the standardized dataset.

[0073] S2.2. Based on resource constraints and capacity, set constraints and use the simplex method to solve the linear programming model to obtain the optimal energy allocation, hydrogen production rate and load adjustment parameters.

[0074] Specifically, energy supply ceiling, hydrogen production rate ceiling, raw material supply, load demand, time window, and unit conversion factor are extracted from the standardized dataset. Resource constraints and capacity are organized into a list of constraints (e.g., the sum of energy allocations does not exceed the energy supply ceiling, hydrogen production rate in each time period does not exceed the capacity ceiling, and demand in each time period is met). Upper and lower bounds and feasible value ranges are set for each decision variable in the linear programming model. The constraints are checked for consistency, and after unifying the unit and time granularity, the simplex method is selected to solve the linear programming model, setting the solution accuracy and iteration ceiling (e.g., convergence tolerance of 1×10⁻). 6 The maximum number of iterations is 1×10. 6 The initial basic feasible solution is constructed using a two-stage simplex method or artificial basic variables. Constraints involving a mixture of equality and inequality are handled using slack variables or artificial variables. During the simplex method process, pivot operations are performed according to the entry and exit rules until the optimality and feasibility conditions are met. The solution state of the simplex method, the objective function value, and the optimal values ​​of each decision variable are read to obtain the optimal energy allocation, hydrogen production rate, and load adjustment parameters.

[0075] It should also be noted that the upper limit of energy supply refers to the maximum energy value that renewable energy or energy storage devices can provide per unit time; the upper limit of hydrogen production rate refers to the maximum hydrogen production rate that hydrogen production devices can achieve under given operating conditions; the raw material supply refers to the total amount of raw materials that flexible chemical processes can obtain within the scheduling cycle; the load demand refers to the energy or hydrogen demand value of chemical production links in each time period; the time window refers to the continuous time range covered by optimized scheduling; and the unit conversion factor is a proportional coefficient used to convert parameters with different units of measurement (such as kilowatts, kilograms, and standard cubic meters) into a unified dimension.

[0076] The simplex method is an iterative algorithm for solving linear programming problems. It optimizes the objective function value by gradually moving to the neighborhood of a feasible solution until the optimal solution is found.

[0077] The rules for entering and leaving the basis are operations in the simplex method. Entering a basis variable refers to a non-basic variable that does not satisfy optimality in the current basic solution. It enters the basic solution through a pivot operation. Leaving a basis variable refers to the least optimal basic variable in the current basic solution. It will be replaced to maintain the feasibility of the solution.

[0078] Constraints refer to the conditions set by resource limitations and capacity settings when solving a linear programming model to ensure that each decision variable (such as energy allocation, hydrogen production rate, and load adjustment parameters) is within the feasible range and meets the actual operational constraints.

[0079] S2.3. Based on the optimal energy allocation, hydrogen production rate, and load adjustment parameters, extract production parameters from the linear programming model, perform unit conversion and time alignment, and generate an initial hydrogen production strategy.

[0080] Specifically, the system reads the values ​​of the decision variables corresponding to the optimal solution, extracts the energy allocation, hydrogen production rate, start-up and shutdown status, and load according to the time window, and performs unit conversion and time granularity alignment; it judges the start-up and shutdown status, corrects outliers and keeps them within the upper limit of energy supply and upper limit of hydrogen production rate; it summarizes the energy allocation, hydrogen production rate and operating status of each period in chronological order to generate the initial hydrogen production strategy.

[0081] It should also be noted that the process of determining the start-up and shutdown status includes: reading the energy allocation and hydrogen production rate for each time period in the optimal solution; determining "start" when the hydrogen production rate is higher than the start-up threshold and "stop" when it is lower than the shutdown threshold; setting the start-up and shutdown thresholds using a hysteresis decision method to avoid frequent switching; verifying the continuous duration, deleting abnormal start-up and shutdown records with excessively short durations and correcting them according to the adjacent majority status; truncating and adjusting values ​​that exceed the upper limit of energy supply or the upper limit of hydrogen production rate while maintaining the corresponding start-up and shutdown status; finally, verifying the consistency of the status in chronological order and outputting the corrected start-up and shutdown status sequence.

[0082] The start-up and shutdown thresholds are typically set based on historical operating data, performance, and safety standards. The start-up threshold is set so that the hydrogen production rate is considered acceptable for startup only when it exceeds the minimum requirement, while the shutdown threshold is set so that the production rate falls below the minimum operating level. The start-up threshold is higher than the shutdown threshold to avoid frequent startups and shutdowns. The hysteresis method sets a "dead zone" between the start-up and shutdown thresholds. For example, the start-up threshold is 60% and the shutdown threshold is 40%. During production rate changes, startup only occurs when the production rate exceeds 60%, shutdown occurs when it falls below 40%, and the current state is maintained when the production rate is between 60% and 40%, thus avoiding frequent switching due to small fluctuations.

[0083] S2.4 Demand forecasting is based on historical data, changes in the external environment, and real-time operating status. Demand forecasting is performed through regression models, and the relationship between historical data and relevant factors is analyzed.

[0084] Specifically, records with the same time granularity are extracted from historical data, external environment change data, and real-time operational status data. Timestamp alignment and unit unification are performed to construct a feature set including lag terms, rolling averages, peak and valley period identifiers, temperature, wind speed, electricity price, and load indication. Training and validation intervals are divided according to time sequence, and regression models are used for parameter estimation, such as ridge regression. Cross-validation is used to select hyperparameters of the regression model and complete the fitting. In the validation interval, residual statistics and error indices are calculated, and the feature set or regression model hyperparameters are adjusted accordingly to determine the final version of the regression model. The latest features of real-time operational status data and external environment change data are input into the regression model, and the predicted demand sequence arranged by time window is output to form the predicted demand.

[0085] It should also be noted that historical data refers to past production and energy usage records, including hydrogen production rate, energy consumption, and load demand, reflecting past operating patterns and trends. Analyzing historical data can identify cyclical changes, regular fluctuations, and correlations with other factors, helping to predict future demand changes. External environmental changes refer to external factors affecting operations, such as climate change, policy adjustments, and market demand fluctuations, which can impact energy supply, demand, and production efficiency. These changes need to be considered in the forecasting model to improve the accuracy and adaptability of the forecast. Real-time operating status refers to current operating data, including current hydrogen production rate, energy consumption, and load adjustment status, reflecting the real-time situation in actual operation. It can reflect immediate production demand and energy usage, helping to adjust the forecasting model and provide more accurate demand forecasts.

[0086] S2.5. Based on the predicted demand, analyze future production demand and energy consumption, and generate optimization targets.

[0087] Specifically, the process involves reading the time-window sequence of predicted demand and aligning it with production capacity, energy supply ceiling, hydrogen production rate ceiling, raw material supply, and time windows using timestamps and unit unification to create a demand-supply comparison table for the scheduling cycle. Based on the predicted demand, the planned output and corresponding energy consumption for each period are calculated. Potential gaps, surpluses, and peak periods are identified by combining the energy supply ceiling and load demand. For example, periods exceeding the energy supply ceiling are marked as high-risk periods. Evaluation indicators such as cost, energy consumption, unmet demand penalties, start-stop frequency, and load fluctuation are extracted. Range standardization is performed on each indicator to form an indicator vector of the same scale. The analytic hierarchy process (AHP) is used to determine the weight order of each evaluation indicator, generating a weight table. The standardized indicators are then summarized using a weighted summation method to obtain a target priority sequence for both short-term and long-term dimensions. Based on the target priority sequence, an executable optimization target list is generated, clarifying the primary and secondary relationships of targets such as "cost minimization, energy consumption minimization, unmet demand minimization, start-stop frequency minimization, and load fluctuation minimization," thus generating optimization targets.

[0088] It should also be noted that the Analytic Hierarchy Process (AHP) is a multi-level and multi-factor decision analysis method. It decomposes complex problems into hierarchical structures and determines the relative weights of multiple factors by comparing them, thereby helping to prioritize and make decisions.

[0089] S2.6. Evaluate the initial hydrogen production strategy, analyze the differences between the strategy and the optimization objectives, and generate an evaluation report.

[0090] Specifically, the initial hydrogen production strategy and optimization objectives are aligned with time stamps and unified in units according to time windows, establishing a comparison list for corresponding time periods. Based on the comparison list, evaluation indicators such as cost, energy consumption, unmet demand, number of start-ups and shutdowns, and load fluctuations are calculated. Cost and energy consumption are statistically analyzed using cumulative and itemized summaries, unmet demand is statistically analyzed using hourly difference cumulative summation, the number of start-ups and shutdowns is statistically analyzed using state transition counting, and load fluctuations are evaluated using a combination of moving average and adjacent difference methods. The deviation between each evaluation indicator and the optimization objective is calculated item by item, forming a deviation detail table, which is sorted according to the objective priority sequence and marked with high, medium, and low levels of difference. The deviation detail table is checked for consistency, and abnormal periods and extreme values ​​are verified and their causes are recorded (e.g., triggering of energy supply ceiling or hydrogen production rate ceiling). The deviation detail table, the list of key periods, and improvement suggestions are integrated into an evaluation report, generating an evaluation report.

[0091] S2.7. Based on the predicted demand and assessment report, adjust the hydrogen production operation parameters in the initial hydrogen production strategy to generate the adjusted hydrogen production operation parameters.

[0092] Specifically, the deviation data between the predicted demand and the assessment report are aligned according to time windows to determine the gap period and the surplus period; for the gap period, the hydrogen production rate and energy allocation are increased according to the gap ratio, and for the surplus period, the hydrogen production rate and energy allocation are reduced according to the surplus ratio, while remaining within the upper limit of energy supply and the upper limit of hydrogen production rate; the start-up and shutdown status is checked and abnormal short-term switching is corrected; the rate of change constraint is applied to the hydrogen production rate changes of adjacent periods and the time series data is smoothed using the rolling average method; finally, the adjusted hydrogen production rate, energy allocation and start-up and shutdown status are summarized in chronological order to generate the adjusted hydrogen production operation parameters.

[0093] S2.8 Input the adjusted hydrogen production operating parameters into the linear programming model, optimize the production process, and generate an optimized hydrogen production strategy.

[0094] Specifically, the hydrogen production rate, energy allocation ratio, and start-up / shutdown status from the adjusted hydrogen production operating parameters are input into the decision variables of the linear programming model; the upper and lower bounds of the variables and the values ​​of the constraints are updated to ensure that the upper limit of energy supply, load demand, and time window are consistent with the adjusted hydrogen production operating parameters; the simplex method is called to solve the problem, and the energy allocation, hydrogen production rate, and start-up / shutdown status in the optimal solution are read and summarized in chronological order to generate the optimized hydrogen production strategy.

[0095] S3. Based on the optimized hydrogen production strategy, identify the characteristics of demand fluctuations, formulate short-term and long-term energy dispatch plans respectively, and combine them to generate a comprehensive dispatch plan.

[0096] S3.1 Based on the optimized hydrogen production strategy, analyze short-term and long-term demand fluctuations, identify the relationship between energy supply and production, and generate demand fluctuation analysis results.

[0097] Specifically, the hydrogen production rate data, energy allocation ratio data, and load demand data in the optimized hydrogen production strategy are divided into short-term and long-term datasets according to time scale, and timestamp alignment and missing value completion are performed. The moving average method is used to smooth the hydrogen production rate data in the short-term dataset, extracting short-term demand fluctuation characteristics. Trend analysis is used to fit the trend of the energy allocation ratio data in the long-term dataset, identifying long-term demand change patterns, such as identifying the upward or downward trend of energy demand within a certain season or cycle, thereby optimizing future energy allocation strategies. The sliding window method combined with standard deviation calculation is used to quantitatively calculate the amplitude of short-term demand fluctuations. The fast Fourier transform method is used to extract short-term wave... Cyclical characteristics; linear regression is used to fit long-term trend data to obtain the long-term trend slope and rate of change; the differences between short-term fluctuation amplitude, fluctuation cycle and long-term trend are normalized and compared, and the demand fluctuation range is determined according to the set fluctuation threshold range; hydrogen production rate data, energy allocation ratio data and energy supply ceiling data are aligned according to time windows, the correlation coefficient and lag time between energy supply data and hydrogen production rate data are calculated, and correlation coefficient analysis and lag time analysis methods are used to identify the relationship between energy supply and hydrogen production; the short-term demand fluctuation characteristics, long-term demand change patterns and the relationship between energy supply and production are compiled into a unified analysis table to generate demand fluctuation analysis results.

[0098] It should also be noted that setting the fluctuation threshold range typically includes: identifying the normal range of fluctuations by calculating the amplitude and cycle of short-term demand fluctuations; using normalization processing to compare the differences between short-term fluctuation amplitude and cycle and long-term demand change patterns on a unified scale; setting a reasonable fluctuation tolerance range based on historical data and actual production conditions, i.e., determining the upper and lower limits of the fluctuation threshold; and ensuring, through verification and adjustment, that the set fluctuation threshold range can effectively distinguish between normal and abnormal fluctuations, thereby determining the reasonable range of demand fluctuations.

[0099] The process of determining the demand fluctuation range based on the set fluctuation threshold range includes: using the standard deviation of short-term fluctuation amplitude in historical data as a reference threshold benchmark, usually set to ±10% to ±20% of the average value; for long-term trend differences, using ±5% of the trend slope change rate as the long-term fluctuation judgment threshold; when the short-term demand fluctuation amplitude exceeds the upper limit of the reference threshold or the long-term trend change rate exceeds the long-term fluctuation judgment threshold range, it is determined to be a demand fluctuation range.

[0100] S3.2 Based on the demand fluctuation analysis results, optimize the energy allocation and load adjustment for each production cycle and generate a short-term energy dispatch plan.

[0101] Specifically, the short-term demand fluctuation amplitude, energy supply ceiling data, and load demand data from the demand fluctuation analysis results are divided into time periods to determine the production cycle; a rolling optimization method is used to allocate energy supply and load ratio within each cycle; the allocation relationship between hydrogen production rate data and flexible chemical energy consumption data is adjusted using energy supply ceiling data as a constraint; and the gradient descent method is used to iteratively correct the energy allocation differences to generate a short-term energy dispatch plan.

[0102] S3.3 Based on the short-term energy dispatch plan and energy supply, analyze the long-term demand trend, energy supply capacity and production load, and generate a long-term energy dispatch plan.

[0103] Specifically, short-term energy dispatch plans are aggregated on a weekly or monthly basis. Energy supply, load demand, feedstock supply, hydrogen production rate ceiling, and energy storage capacity are aligned and standardized by time window to form a long-term baseline sequence. Based on the energy supply ceiling and load demand, charging and discharging sequences and reserve capacity are arranged. A rolling planning method is used to allocate hydrogen production rate and energy allocation ratios within each time window, setting minimum continuous operating time, minimum continuous downtime, and upper limits for the rate of change between adjacent windows (e.g., a change rate not exceeding 5%). During constraint verification, each item—energy supply ceiling, hydrogen production rate ceiling, feedstock supply, and the actual allocation value of the time window—is checked to ensure it does not exceed the limits. For time periods where limits are detected to be exceeded, the allocation ratio is adjusted back according to energy priority. Scenario analysis is conducted to calculate long-term allocation sequences for high supply, low supply, and maintenance scenarios, and feasible intersections are taken to obtain robust long-term allocation sequences. The long-term allocation sequences are smoothed and time-series verified. A long-term parameter list of hydrogen production rate, energy allocation ratio, and reserve capacity is output by time window to generate a long-term energy dispatch plan.

[0104] It should also be noted that the limited scope refers to the allowable range of values ​​for the upper limit of energy supply, the upper limit of hydrogen production rate, the amount of raw material supply, and the time window. The upper limit of energy supply is limited to the maximum available energy of the energy system, the upper limit of hydrogen production rate is limited to the rated capacity, the amount of raw material supply is limited to the maximum daily available amount of raw material, and the time window is limited to the allowable scheduling period for each cycle.

[0105] Scenario analysis is a method that involves constructing different hypothetical scenarios (such as high supply, low supply, and maintenance), evaluating the results under each scenario, and deriving the most robust decision based on the feasible intersection of multiple scenarios.

[0106] S3.4 Combine short-term and long-term energy dispatch schemes to integrate energy allocation and load regulation, and generate a comprehensive dispatch scheme.

[0107] Specifically, the hydrogen production rate data, energy allocation ratio data, and load adjustment parameters in the short-term energy dispatch plan are aligned with the long-term energy dispatch plan according to time windows, unifying the time base and data format. A weighted fusion method is used to merge the short-term and long-term energy allocation ratio data within the same time period, with short-term weights determined based on demand fluctuation frequency and long-term weights based on energy supply stability. A hierarchical weighted regression method is used to calculate the weighted average of the merged energy allocation ratio and load adjustment parameters, generating an intermediate comprehensive sequence. Moving average and three-sigma anomaly detection methods are used to smooth the intermediate comprehensive sequence and detect abrupt changes, identifying abnormal fluctuations and performing interpolation corrections to ensure the continuity of energy allocation. The smoothed energy allocation ratio data, hydrogen production rate data, and load adjustment parameters are output in chronological order as a unified comprehensive dispatch time series, generating the comprehensive dispatch plan.

[0108] Among them, the moving average method is a method to smooth the fluctuations of time series by calculating the average value of continuous data points within a fixed time window; the three sigma anomaly detection method is a method based on statistical principles to identify outliers by judging whether data points exceed the mean ± 3 times the standard deviation.

[0109] It should be noted that by analyzing the demand fluctuation characteristics in the optimized hydrogen production strategy, a short-term energy allocation and load adjustment scheme is generated. Combined with the long-term energy supply and demand relationship, a cross-time-level scheduling strategy is formed, realizing the unity of short-term flexible adjustment and long-term stable planning. This enables energy allocation to have dynamic adaptability and global coordination. It can achieve efficient coordination of energy flow and material flow in an off-grid environment, balance short-term load and long-term energy consumption targets, significantly improve the overall utilization rate and operating economy of the energy system, and avoid uneven production or excessive redundant energy storage caused by energy fluctuations. This enables continuous optimization and coordinated scheduling of flexible chemical processes and hydrogen production processes.

[0110] S4. Perform energy allocation and scheduling for the integrated scheduling scheme, monitor the operation status and energy allocation process in real time, and generate real-time operation data and scheduling feedback.

[0111] S4.1 Perform energy allocation and scheduling for the comprehensive scheduling plan, initiate scheduling operations in the production process, and generate preliminary scheduling execution records.

[0112] Specifically, the energy allocation ratio data, hydrogen production rate data, and load adjustment parameters in the comprehensive scheduling plan are divided into multiple scheduling cycles according to time windows. Within each scheduling cycle, the corresponding allocable energy is calculated based on the upper limit of energy supply data and the load demand data, and the energy resources are allocated proportionally to the hydrogen production and flexible chemical processes. The scheduling operations of each production process are started sequentially using a sequential execution method, and the start-up time, energy consumption, hydrogen production, and chemical energy consumption data of each process are recorded. The energy consumption fluctuations and load response in the initial stage of operation of each process are sampled and monitored, and the sampling results are aligned with the scheduling plan. The energy allocation records, production process start-up records, and operation monitoring data of each scheduling cycle are summarized into a unified format to generate a preliminary scheduling execution record.

[0113] S4.2. Monitor the preliminary scheduling execution records in real time and generate real-time operation data.

[0114] Specifically, the start-up time, energy allocation, and load adjustment parameters of each production stage are extracted from the preliminary scheduling execution records, and the data is aligned according to time windows; real-time operational status data of each production stage is collected, including actual hydrogen production rate, actual energy consumption, operational status, and load response; the operational data of each stage is updated periodically through real-time data acquisition and compared with the predetermined plan in the preliminary scheduling execution records to calculate deviation values ​​and energy efficiency changes; real-time data is smoothed to remove instantaneous fluctuations, and real-time operational data is updated according to set time intervals; the real-time operational data is summarized by time series to form complete real-time operational data.

[0115] It should also be noted that the predetermined plan refers to the energy allocation, hydrogen production rate, load adjustment parameters, and start-up and end times of each production stage that have been determined in the preliminary scheduling execution record. This plan is used to compare actual operating data to determine execution deviations. The set time interval refers to the fixed cycle for real-time operating data collection and updating. It is usually determined according to the dynamic nature of the process. For example, the energy allocation process uses a sampling interval of 1 to 5 minutes, while the hydrogen production and chemical load processes use a sampling interval of 10 minutes to 1 hour. This is used to ensure the timeliness and data continuity of real-time monitoring.

[0116] S4.3 Based on real-time operating data, monitor the energy allocation for each production cycle and generate energy allocation feedback.

[0117] Specifically, real-time operational data is grouped according to production cycles, and the energy input, hydrogen production rate, and flexible chemical energy consumption data for each cycle are statistically summarized. The total energy input and output for each cycle is calculated using the energy balance method to determine the energy allocation ratio and energy loss value. The energy allocation ratio is compared item by item with the planned allocation ratio in the comprehensive scheduling scheme to calculate the deviation value and identify the time period exceeding the allowable deviation threshold. The source of deviation is determined using the difference analysis method, including energy supply fluctuations, abnormal hydrogen production rate, or load changes, and the corresponding time points and values ​​are recorded. The deviation value, deviation source, and adjustment suggestions are compiled into a structured table to generate energy allocation feedback.

[0118] It should also be noted that the allowable deviation threshold is a limited value used to determine whether the difference between the energy allocation ratio and the planned allocation ratio in the comprehensive dispatch scheme is within an acceptable range, in order to distinguish between normal fluctuations and abnormal deviations. The allowable deviation threshold is usually determined based on the fluctuation range of historical operating data, the stability of the upper limit of energy supply, and the sensitivity of load demand. It is generally taken as ±3% to ±5% of the planned allocation ratio as the allowable range. When the difference between the actual energy allocation ratio and the planned allocation ratio exceeds the upper limit of the allowable deviation threshold, it is determined to be an over-limit period. By setting the allowable deviation threshold, the stability and controllability of energy allocation during the dispatch process can be guaranteed, and sudden deviations in energy utilization rate or hydrogen production rate can be prevented.

[0119] S4.4 Combine real-time operating data with energy allocation feedback and perform deviation analysis to generate scheduling feedback.

[0120] Specifically, real-time operational data and energy allocation feedback are aligned according to a unified time window and merged into a joint dataset according to the production cycle. The difference between real-time energy input, hydrogen production rate data, flexible chemical energy consumption data, and planned values ​​in the energy allocation feedback is calculated using a differential calculation method to obtain an initial deviation sequence. The deviation sequence is then divided into three categories—energy supply deviation, load response deviation, and energy efficiency deviation—using a deviation decomposition method, and the proportion and cumulative impact of each type of deviation are calculated. Time series analysis is used to identify the changing trend of deviations within a continuous period, marking the time periods of continuous deviation from the target range. The results of various deviations and time trend information are compiled into a basis for scheduling adjustments, generating scheduling feedback that includes deviation type, deviation magnitude, duration, and suggested adjustment direction.

[0121] S5. Based on real-time operation data and scheduling feedback, adjust energy allocation and hydrogen production operation parameters, generate the optimal scheduling scheme, conduct a comprehensive performance evaluation, and generate an optimized scheduling effect report.

[0122] S5.1 Based on real-time operation data and scheduling feedback, analyze the current gap between energy allocation and production efficiency, and generate a gap analysis report.

[0123] Specifically, real-time operational data and scheduling feedback are aligned by timestamp and production cycle to form a joint dataset. The energy input data, hydrogen production rate data, flexible chemical energy consumption data, and load response data in the joint dataset are statistically summarized to calculate the energy utilization rate, hydrogen production efficiency, and energy consumption ratio for each production cycle. A comparative analysis method is used to calculate the difference between real-time operational data and planned values ​​in scheduling feedback, determining the energy allocation deviation rate and production efficiency deviation rate. Correlation analysis is used to identify the main influencing factors contributing to the gaps, including energy supply fluctuations, low hydrogen production rates, or uneven load distribution. The gap data, sources of deviation, and degree of impact are organized by time series to generate a gap analysis report containing energy allocation gaps, production efficiency gaps, and causal analysis.

[0124] It should also be noted that correlation analysis is a method that identifies the relationships between different factors by calculating the correlation coefficients between variables, and then finds out the main factors affecting the differences in results.

[0125] S5.2 Based on the gap analysis report, adjust the energy allocation and hydrogen production operation parameters, and combine them with the initial hydrogen production strategy to generate the optimal scheduling scheme.

[0126] Specifically, the energy allocation gap rate, hydrogen production rate deviation rate, and load response deviation rate data are extracted from the gap analysis report to determine the time periods and parameter types where deviations exceed limits. Based on the energy allocation gap rate, the energy supply ceiling data and load demand data for each production cycle are proportionally corrected, and the energy supply is redistributed using a proportional adjustment method. Based on the hydrogen production rate deviation rate, the hydrogen production operation parameters, including electrolyzer current density, electrolysis temperature, and gas flow rate, are optimized and adjusted, and limited within the upper limit of hydrogen production rate. The adjusted energy allocation data and hydrogen production operation parameters are time-synchronized, and the comprehensive scheduling parameters are calculated in the overlapping interval using a linear weighting method. The synchronized energy allocation data and hydrogen production operation parameters are output as a unified scheduling plan table according to the time series, generating the optimal scheduling scheme.

[0127] S5.3. Based on the execution status of the optimal scheduling scheme, collect data on energy usage, production efficiency, and operating status to generate an execution dataset.

[0128] Specifically, the process involves reading the time windows and scheduling plan table of the optimal scheduling scheme, listing the energy usage, production efficiency, and operating status fields to be recorded in each time window, and forming a collection list. According to the collection list, raw data on energy usage (e.g., electricity consumption and hydrogen compression energy consumption), production efficiency (e.g., hydrogen production per unit of energy and qualification rate), and operating status (e.g., start-up / shutdown status and load level) are recorded in each time window, and a timestamp is added to each record. The energy usage, production efficiency, and operating status are time-stamp aligned and units are unified (e.g., conversion between power and energy, and mass and volume conversion) to ensure that the three types of data are comparable within the same time window. Missing values ​​are marked and filled using adjacent time-series interpolation. Outliers are identified using the 3σ criterion and replaced with adjacent valid values. Duplicate records are deduplicated. The aligned, unified, and verified energy usage, production efficiency, and operating status are merged into a structured table in chronological order, labeled with field names, units, and time windows, and output as the execution dataset.

[0129] S5.4. Based on the execution dataset, evaluate energy consumption, production efficiency, and cost control, and compare them with the optimization objectives to generate performance evaluation results.

[0130] Specifically, the energy usage, production efficiency, and operating cost fields in the execution dataset are read and organized by time window. The total energy consumption and unit hydrogen production energy consumption for each time window are calculated using an energy consumption calculation method. Production efficiency indicators (e.g., hydrogen production to energy consumption ratio) and economic indicators (e.g., unit product cost and energy consumption ratio) are calculated. The differences between each indicator and its corresponding target value in the optimization objective are compared to determine the energy consumption deviation rate, production efficiency deviation rate, and cost control deviation rate. The three types of deviations are normalized using a weighted comprehensive scoring method, and a comprehensive performance score is calculated according to the performance evaluation weights. The expression is:

[0131] ;

[0132] in, Indicates time window The overall performance score, Indicates time window The efficiency of hydrogen production. Indicates time window Cost deviation rate, Indicates time window Energy consumption deviation rate Weights representing production efficiency Indicates the weight of cost control. Indicates the weight of energy consumption. Indicates a time window;

[0133] The energy consumption deviation, efficiency deviation, cost deviation, and overall performance score for each time window are summarized in time series to generate a performance evaluation result that includes numerical tables and trend results.

[0134] It should also be noted that the performance evaluation weights are used to determine the relative importance of the three indicators of energy consumption, production efficiency and cost control in the comprehensive evaluation. The weight of energy consumption is 0.4, the weight of production efficiency is 0.4 and the weight of cost control is 0.2. This is used to balance the impact of energy efficiency, output and economy, so that the comprehensive evaluation results can accurately reflect the overall scheduling performance.

[0135] The energy consumption calculation method is a method that calculates production efficiency and economic indicators by statistically analyzing the total energy consumption and unit hydrogen production energy consumption in each time window, in order to evaluate energy use efficiency and production costs.

[0136] S5.5 Analyze the performance evaluation results, identify the gap between the actual execution and the optimization target, and generate an optimization scheduling effect report.

[0137] Specifically, energy consumption deviation data, production efficiency deviation data, and cost control deviation data are extracted from the performance evaluation results and organized and classified according to time windows to generate a deviation dataset. The percentage of energy consumption deviation, production efficiency deviation, and cost control deviation is calculated using a difference comparison method to generate deviation comparison results. These results are then used as input data, and cluster analysis is employed to group and statistically analyze the deviations, identifying the main sources of energy, efficiency, and economic gaps, generating gap classification results. Trend analysis is used to analyze the temporal variation patterns of energy consumption deviation data, production efficiency deviation data, and cost control deviation data, identifying periods of continuous deviation from the optimization target or excessive fluctuations, generating gap trend results. Based on the gap trend results, the impact and duration of each gap are calculated, determining the priority of gap adjustments and generating a gap priority list. Finally, the gap priority list, gap trend results, and gap classification results are combined to form a comprehensive analysis including the source of deviation, deviation magnitude, duration, and adjustment direction, generating an optimization scheduling effect report.

[0138] It should also be noted that trend analysis is a method that identifies periods of continuous deviation from the optimization target or excessive fluctuation by analyzing the trend of data changes over time, thereby revealing the patterns of data change and potential problems.

[0139] In summary, this invention utilizes linear programming models and the simplex method to efficiently determine the optimal energy allocation and production strategy under multiple constraints. This not only optimizes production capacity under limited resources but also balances load regulation and energy efficiency, thereby reducing energy consumption and production costs. It provides a precise basis for subsequent scheduling and adjustment, and achieves optimization of hydrogen production efficiency and energy utilization.

[0140] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A multi-objective optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical engineering, characterized by: The process includes: collecting raw data and preprocessing it to generate a standardized dataset; analyzing the standardized dataset using a linear programming model to generate an initial hydrogen production strategy. The specific steps are as follows: defining decision variables and constructing an objective function based on the input parameters of the standardized dataset to generate a linear programming model; setting constraints based on resource limitations and capacity, and solving the linear programming model using the simplex method to obtain optimal energy allocation, hydrogen production rate, and load adjustment parameters; extracting production parameters from the linear programming model based on the optimal energy allocation, hydrogen production rate, and load adjustment parameters, performing unit conversion and time alignment to generate the initial hydrogen production strategy; adjusting the hydrogen production operation parameters of the initial hydrogen production strategy based on predicted demand to generate an optimized hydrogen production strategy; and identifying demand fluctuation characteristics based on the optimized hydrogen production strategy, developing short-term and long-term energy dispatch schemes and combining them to generate a comprehensive dispatch scheme. The specific steps are as follows: analyzing short-term and long-term demand fluctuations based on the optimized hydrogen production strategy, identifying the relationship between energy supply and production, and generating demand fluctuation analysis results. Based on the results of demand fluctuation analysis, optimize energy allocation and load adjustment for each production cycle to generate a short-term energy dispatch plan; Based on the short-term energy dispatch plan and energy supply, analyze the long-term demand trend, energy supply capacity and production load to generate a long-term energy dispatch plan. By combining short-term and long-term energy dispatch schemes, energy allocation and load regulation are integrated to generate a comprehensive dispatch scheme; energy is allocated and dispatched according to the comprehensive dispatch scheme, and the operating status and energy allocation process are monitored in real time to generate real-time operating data and dispatch feedback. Based on real-time operational data and scheduling feedback, energy allocation and hydrogen production operating parameters are adjusted to generate the optimal scheduling scheme, and a comprehensive performance evaluation is conducted to generate an optimized scheduling effect report.

2. The multi-objective optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical engineering as described in claim 1, characterized in that: The specific steps for generating the standardized dataset are as follows: Based on the original data, identify and remove outliers, missing values, and noisy data to generate a clean dataset; merge and align the clean dataset with the external environment data to generate a merged dataset. Performance metrics are extracted from the merged dataset and standardized to generate a standardized dataset.

3. The multi-objective optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical engineering as described in claim 1, characterized in that: The predicted demand is obtained by using a regression model to predict demand based on historical data, changes in the external environment, and real-time operating status, and by analyzing the relationship between historical data and relevant factors.

4. The multi-objective optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical engineering as described in claim 1, characterized in that: The specific steps for generating the optimized hydrogen production strategy are as follows: Based on the predicted demand, analyze future production demand and energy consumption to generate optimization targets; evaluate the initial hydrogen production strategy, analyze the differences between it and the optimization targets, and generate an evaluation report; adjust the hydrogen production operation parameters in the initial hydrogen production strategy based on the predicted demand and the evaluation report to generate the adjusted hydrogen production operation parameters. The adjusted hydrogen production operating parameters are input into a linear programming model to optimize the production process and generate an optimized hydrogen production strategy.

5. The multi-objective optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical engineering as described in claim 1, characterized in that: The specific steps for generating real-time operational data and scheduling feedback are as follows: energy allocation and scheduling are performed on the comprehensive scheduling scheme; scheduling operations in the production process are initiated to form a preliminary scheduling execution record; the preliminary scheduling execution record is monitored in real time to generate real-time operational data. Based on real-time operational data, monitor the energy allocation for each production cycle and generate energy allocation feedback; By combining real-time operational data with energy allocation feedback and performing deviation analysis, scheduling feedback is generated.

6. The multi-objective optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical engineering as described in claim 1, characterized in that: The specific steps for generating the optimal scheduling scheme are as follows: Based on real-time operating data and scheduling feedback, analyze the gap between the current energy allocation and production efficiency, and generate a gap analysis report; based on the gap analysis report, adjust the energy allocation and hydrogen production operating parameters, and combine them with the initial hydrogen production strategy to generate the optimal scheduling scheme.

7. The multi-objective optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical engineering as described in claim 1, characterized in that: The specific steps for generating the optimized scheduling effect report are as follows: Based on the execution status of the optimal scheduling scheme, collect energy usage, production efficiency, and operating status to generate an execution dataset; Based on the execution dataset, evaluate energy consumption, production efficiency, and cost control, and compare them with the optimization target to generate performance evaluation results; Analyze the performance evaluation results to identify the gap between the actual execution status and the optimization target, and generate the optimized scheduling effect report.

8. A multi-objective optimization scheduling system based on the coupling of off-grid hydrogen production and flexible chemical engineering, characterized in that: The method for implementing the multi-objective optimization scheduling method based on the coupling of off-grid hydrogen production and flexible chemical industry as described in any one of claims 1 to 7 includes: a preprocessing module for collecting raw data and preprocessing it to generate a standardized dataset; a strategy optimization module for analyzing the standardized dataset using a linear programming model to generate an initial hydrogen production strategy, and adjusting the hydrogen production operation parameters of the initial hydrogen production strategy according to predicted demand to generate an optimized hydrogen production strategy; a scheduling construction module for identifying demand fluctuation characteristics based on the optimized hydrogen production strategy, formulating short-term and long-term energy scheduling schemes respectively, and combining them to generate a comprehensive scheduling scheme; a real-time monitoring module for allocating and scheduling energy according to the comprehensive scheduling scheme, and monitoring the operating status and energy allocation process in real time to generate real-time operating data and scheduling feedback; and a performance evaluation module for adjusting energy allocation and hydrogen production operation parameters based on real-time operating data and scheduling feedback to generate an optimal scheduling scheme, and conducting a comprehensive performance evaluation to generate an optimized scheduling effect report.

Images