Real-time coal gas scheduling system and method by utilizing peak load shifting of gas holder

Abstract

The invention relates to a real-time gas scheduling system and method utilizing peak load shifting of a gas holder, and belongs to the technical field of energy scheduling devices and methods of iron and steel enterprises. According to the technical scheme, a data acquisition module acquires data and transmits the data to a data preprocessing module; the data preprocessing module forms a high-quality data set; the prediction module predicts future gas production and consumption and power load demands based on the preprocessed data; the scheduling model construction module constructs a dynamic scheduling model according to the prediction data and the real-time data; the execution control module converts the scheduling scheme into a control instruction; and the feedback adjustment module adjusts the scheduling scheme in real time according to the execution effect of the execution control module. The method has the advantages that through real-time monitoring, accurate prediction and dynamic optimization scheduling of gas production and consumption, the peak load shifting function of the gas holder is fully played, efficient utilization of gas resources is achieved, the energy cost of enterprises is reduced, and the energy management level and economic benefits of iron and steel enterprises are improved.

Description

Technical Field

[0001] This invention relates to a real-time gas dispatching system and method that utilizes gas holders for peak shaving and valley filling, belonging to the technical field of energy dispatching devices and methods for steel enterprises. Background Technology

[0002] The steel industry is a pillar industry of the national economy and a major energy consumer. Coal gas, as an important secondary energy source generated during steel production (including blast furnace gas, coke oven gas, and converter gas), occupies a significant position in the energy structure of enterprises. However, the steel production process is complex, and the production and consumption of coal gas at each stage exhibit significant dynamism and uncertainty. Blast furnace gas is affected by furnace conditions and raw materials, coke oven gas is related to coking time and coal blending scheme, and converter gas is produced intermittently due to blowing cycle.

[0003] Gas consumption in processes such as steelmaking and rolling fluctuates irregularly depending on production plans and equipment status.

[0004] This imbalance between production and consumption often leads to supply and demand imbalances in the gas system: when there is an overproduction, the gas is released, causing waste and pollution; when there is an underproduction, production is affected or external energy purchases are needed, increasing costs.

[0005] While gas holders play a crucial role in regulating supply and demand and achieving peak shaving and valley filling, many steel companies rely on manual scheduling, lacking accurate forecasting and dynamic adjustment capabilities, thus failing to fully realize their function. Furthermore, existing systems do not integrate peak-valley electricity pricing mechanisms for optimized scheduling, resulting in the gas holders' peak-shaving and valley filling value not being fully realized. Therefore, developing a real-time scheduling system that integrates peak-valley electricity pricing mechanisms and deeply utilizes gas holders for peak shaving and valley filling is of great significance. Summary of the Invention

[0006] The purpose of this invention is to provide a real-time gas dispatching system and method that utilizes gas holders for peak shaving and valley filling. Closely integrated with the peak-valley electricity pricing mechanism, this system fully leverages the peak shaving and valley filling function of gas holders through real-time monitoring, accurate prediction, and dynamic optimization of gas production and consumption. This achieves efficient utilization of gas resources, reduces enterprise energy costs, improves the energy management level and economic benefits of steel enterprises, and effectively solves the aforementioned problems existing in the background technology.

[0007] The technical solution of this invention is: a real-time gas dispatching system utilizing gas holders for peak shaving and valley filling, comprising a data acquisition module, a data preprocessing module, a prediction module, a dispatching model construction module, a dispatching scheme generation module, an execution control module, and a feedback adjustment module. The data acquisition module collects real-time data on gas production and consumption, gas holder operation, electricity, and production plans, and transmits this data to the data preprocessing module. The data preprocessing module performs outlier removal, missing value imputation, standardization, and correlation processing on the collected data to form a high-quality dataset. The prediction module, based on the preprocessed data, uses a combined prediction algorithm to predict future gas production and consumption and electricity load demand. The dispatching model construction module constructs a dynamic dispatching model with the goal of minimizing total energy costs based on the predicted data and real-time data. The execution control module converts the dispatching scheme into control commands to control the operation of the gas holder and related equipment. The feedback adjustment module adjusts the dispatching scheme in real-time based on the execution effect of the execution control module to ensure stable system operation.

[0008] A real-time gas dispatching method utilizing gas holders for peak shaving and valley filling, characterized by comprising the following steps: (1) Data acquisition: The data acquisition module collects various types of data in real time and transmits them to the data preprocessing module. Among them, the data acquisition related to peak shifting and valley filling of gas holders is the key. (2) Data preprocessing: The data preprocessing module processes the raw data to provide high-quality data for subsequent analysis and decision-making related to peak shifting and valley filling of gas holders; (3) Production and load forecasting: The forecasting module uses a combined forecasting algorithm based on preprocessed data to predict future gas production and consumption and power load demand, providing a basis for planning peak shifting and valley filling operations of gas holders in advance. (4) Construct a scheduling model. The scheduling model construction module constructs a dynamic scheduling model based on the predicted data and real-time data, with the goal of minimizing total energy costs and highlighting the peak shifting and valley filling function of the gas holder. (5) Generate scheduling scheme. The scheduling scheme generation module generates the optimal scheduling scheme, which includes the specific operations of peak shifting and valley filling of the gas holder, based on the model solution results. (6) Execution and control: The execution and control module converts the peak shifting and valley filling operations of the gas holder in the scheduling scheme into control commands and sends them to the execution equipment to realize real-time control of the gas system; (7) Feedback and adjustment: The feedback and adjustment module monitors the execution effect and system operation status in real time, focusing on the actual effect of the gas holder peak shifting and valley filling operation to form a closed-loop control.

[0009] In step (1), the content collected includes: The generation, consumption and calorific value of gas generated by blast furnace and coke oven gas generating equipment and gas consumption equipment of steel rolling heating furnace are collected by flow sensor and calorific value analyzer. The sampling frequency is not less than 1 minute / time, which provides a basis for judging the gas supply and demand status and deciding whether the gas holder should store gas or release gas. The gas holder's storage and venting capacity is monitored in real time by collecting data on the cabinet height, internal pressure, and inlet and outlet gas flow rates using liquid level and pressure sensors, with a sampling frequency of no less than 1 second per time. Data collection on substation power load and purchased electricity, as well as the division of peak, flat, and valley electricity pricing periods and the electricity prices for each period, provides a reference for adjusting the peak-shifting and valley-filling strategies of gas holders in conjunction with the electricity pricing mechanism; Collect production plans and equipment operating status to support the prediction of gas production and consumption and the advance planning of peak shaving and valley filling operations for gas holders; Industrial Ethernet and wireless sensor network communication are adopted, and data encryption and verification technologies are used to ensure secure and reliable data transmission, thereby ensuring the accuracy and timeliness of data related to peak shifting and valley filling in gas holders.

[0010] The specific steps in step (2) are as follows: The 3σ criterion and box plot method were used to remove outliers from the gas holder operation data and gas production and consumption data, and duplicate records were deduplicated to avoid outliers affecting the judgment of the timing of peak shifting and valley filling of the gas holder. For missing values ​​in the gas holder operation data and production and consumption data, linear interpolation and Lagrange interpolation methods are used to fill them in based on the characteristics and correlation of time series data, ensuring data continuity and enabling continuous analysis of gas supply and demand trends, thus providing coherent data support for peak shaving and valley filling operations. Standardize and convert gas holder data and gas production and consumption data of different units and formats, and normalize or standardize them to the range of [0,1] or [-1,1] to facilitate the unified processing of gas holder peak shifting and valley filling related data in subsequent models. By linking coal gas production and consumption data with relevant data from different data sources such as gas holder operation data and electricity data, a complete dataset is constructed. The processed data is stored in a distributed database, with a focus on ensuring the storage quality and access speed of gas holder operation data and gas production and consumption data, so as to meet the need for rapid access to peak shifting and valley filling data in subsequent real-time scheduling.

[0011] The specific steps in step (3) are as follows: By employing a combination algorithm of long short-term memory network and gradient boosting tree, combined with production process and equipment status data, the model parameters are dynamically adjusted to make 1-hour and 4-hour predictions, accurately capturing the time series characteristics and nonlinear features of gas production and consumption, and predicting in advance when there will be a gas surplus and when there will be a gas shortage. Using a similar combined algorithm, combined with production plan, historical load and peak-to-valley time information, the demand for electricity load in each period can be predicted, providing a reference for optimizing the peak-shifting and valley-filling strategy of gas holders in conjunction with the electricity price mechanism. During peak electricity periods, gas can be released from the gas holders first to meet the gas demand and reduce the purchase of electricity. The root mean square error and mean absolute percentage error are calculated to evaluate the prediction results. If the accuracy is lower than the threshold, the model is retrained or the algorithm is adjusted to ensure the accuracy of the prediction of gas production and consumption, thus providing a guarantee for the accurate planning of peak shifting and valley filling operations of gas holders.

[0012] In step (4), the objective function is Minimize C=C1+C2+C3+C4, where C is the total energy cost, C1 is the cost of gas venting loss, C2 is the cost of purchased electricity, C3 is the operating cost of the gas holder, and C4 is the operating and maintenance cost of equipment such as gas turbines. Reducing the cost of gas venting loss and the cost of purchased electricity through the peak shifting and valley filling function of the gas holder is an important way to achieve the objective. The cost of purchased electricity is related to the peak, flat and valley electricity price. During peak hours, gas is released through the gas holder, i.e., peak shifting, to increase gas power generation and reduce purchased electricity. During valley hours, gas is stored through the gas holder, i.e. valley filling, to store excess gas for use during peak hours. The function constraints include gas flow balance, gas holder operation, equipment operation, power balance, and peak-valley electricity price. The scheduling model distinguishes between normal and abnormal production conditions. For normal conditions, an improved particle swarm optimization algorithm is used to solve the model, while for abnormal conditions, a three-layer decision tree model is constructed to support the model's rapid response.

[0013] In step (5), The gas inlet and outlet volumes of the gas holders at different times should be clearly defined, i.e., the valley filling operation and the peak shifting operation, as well as the gas distribution and supply of each piece of equipment, the output of the gas turbine equipment and the purchased electricity. Among these, the gas inlet and outlet volumes of the gas holders are the core content of the plan and need to be accurately determined based on the peak, flat and valley periods and the gas production and consumption forecast to ensure the timeliness and accuracy of the peak shifting and valley filling operations. It has multi-scenario analysis capabilities, simulating peak shifting and valley filling operation schemes for gas holders under different gas production and consumption fluctuations and electricity price changes, providing decision support and emergency plans for dispatchers, and ensuring that peak shifting and valley filling can be achieved through reasonable operation of gas holders under various circumstances; The dispatching scheme distinguishes between the specific implementation of normal and abnormal operating conditions. Under normal operating conditions, it is implemented in different electricity price stages. The priority of handling abnormal operating conditions is higher than that of electricity price time periods. Abnormal operating conditions adopt a hierarchical response mechanism with built-in contingency plans. A solution is generated within 3 seconds, prioritizing the basic peak shifting and valley filling functions of the gas holder.

[0014] The specific steps in step (6) are as follows: The key is to control the inlet and outlet of the gas holder and the opening of the pipeline regulating valves. According to the scheduling plan, the gas intake and output of the gas holder are precisely adjusted. During the valley filling stage, the gas intake is increased and during the peak shifting stage, the gas output is increased to achieve precise control of gas storage and flow distribution. Control the fuel supply and combustion intensity of the gas turbine and boiler equipment, and adjust the output to coordinate with the peak shifting and valley filling operation of the gas holder. When the gas holder releases gas, increase the gas supply of the gas turbine to increase the power generation. It communicates with the power sector's dispatch system and automatically adjusts the purchased electricity volume according to the peak-shaving and valley-filling situation of the gas holder, reducing the amount of purchased electricity when the gas holder releases a large amount of gas; It adopts PLC and DCS industrial control technology and has manual and automatic switching functions to ensure the accuracy of peak shaving and valley filling operation control of the gas holder and the stability of the system. In case of abnormality in automatic control, the gas holder can be manually operated to ensure that the peak shaving and valley filling function is not interrupted.

[0015] The specific steps in step (7) are as follows: By comparing the actual gas inlet and outlet volume and tank height with the target values ​​of the scheme, as well as the deviation between the actual gas production and consumption and the predicted values, the impact of the deviation on the peak shifting and valley filling effect and the reasons are analyzed. Based on the cause of the deviation, the scheduling model parameters are corrected or the gas holder's inlet and outlet gas scheduling scheme is directly adjusted and sent to the execution control module. The time granularity is consistent with the scheduling plan to ensure timely correction of deviations and maintain the optimal operating state of the gas holder's peak-shifting and valley-filling function.

[0016] The beneficial effects of this invention are: by closely integrating the peak-valley electricity pricing mechanism, through real-time monitoring, accurate prediction and dynamic optimization scheduling of gas production and consumption, the peak-shifting and valley-filling function of the gas holder is fully utilized, achieving efficient utilization of gas resources, reducing enterprise energy costs, and improving the energy management level and economic benefits of steel enterprises. Detailed Implementation

[0017] To make the purpose, technical solutions, and advantages of the invention's embodiments clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only a small part of the embodiments of the present invention, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the protection scope of the present invention.

[0018] A real-time gas dispatching system utilizing gas holders for peak shaving and valley filling includes a data acquisition module, a data preprocessing module, a prediction module, a dispatching model construction module, a dispatching scheme generation module, an execution control module, and a feedback adjustment module. The data acquisition module collects real-time data on gas production and consumption, gas holder operation, electricity supply, and production plans, and transmits this data to the data preprocessing module. The data preprocessing module performs outlier removal, missing value imputation, standardization, and correlation processing on the collected data to form a high-quality dataset. The prediction module, based on the preprocessed data, uses a combined prediction algorithm to predict future gas production and consumption and electricity load demand. The dispatching model construction module constructs a dynamic dispatching model with the goal of minimizing total energy costs based on the predicted and real-time data. The execution control module converts the dispatching scheme into control commands to control the operation of the gas holder and related equipment. The feedback adjustment module adjusts the dispatching scheme in real-time based on the execution effect of the execution control module to ensure stable system operation.

[0019] A real-time gas dispatching method utilizing gas holders for peak shaving and valley filling, characterized by comprising the following steps: (1) Data acquisition: The data acquisition module collects various types of data in real time and transmits them to the data preprocessing module. Among them, the data acquisition related to peak shifting and valley filling of gas holders is the key. (2) Data preprocessing: The data preprocessing module processes the raw data to provide high-quality data for subsequent analysis and decision-making related to peak shifting and valley filling of gas holders; (3) Production and load forecasting: The forecasting module uses a combined forecasting algorithm based on preprocessed data to predict future gas production and consumption and power load demand, providing a basis for planning peak shifting and valley filling operations of gas holders in advance. (4) Construct a scheduling model. The scheduling model construction module constructs a dynamic scheduling model based on the predicted data and real-time data, with the goal of minimizing total energy costs and highlighting the peak shifting and valley filling function of the gas holder. (5) Generate scheduling scheme. The scheduling scheme generation module generates the optimal scheduling scheme, which includes the specific operations of peak shifting and valley filling of the gas holder, based on the model solution results. (6) Execution and control: The execution and control module converts the peak shifting and valley filling operations of the gas holder in the scheduling scheme into control commands and sends them to the execution equipment to realize real-time control of the gas system; (7) Feedback and adjustment: The feedback and adjustment module monitors the execution effect and system operation status in real time, focusing on the actual effect of the gas holder peak shifting and valley filling operation to form a closed-loop control.

[0020] In step (1), the content collected includes: The generation, consumption and calorific value of gas generated by blast furnace and coke oven gas generating equipment and gas consumption equipment of steel rolling heating furnace are collected by flow sensor and calorific value analyzer. The sampling frequency is not less than 1 minute / time, which provides a basis for judging the gas supply and demand status and deciding whether the gas holder should store gas or release gas. The gas holder's storage and venting capacity is monitored in real time by collecting data on the cabinet height, internal pressure, and inlet and outlet gas flow rates using liquid level and pressure sensors, with a sampling frequency of no less than 1 second per time. Data collection on substation power load and purchased electricity, as well as the division of peak, flat, and valley electricity pricing periods and the electricity prices for each period, provides a reference for adjusting the peak-shifting and valley-filling strategies of gas holders in conjunction with the electricity pricing mechanism; Collect production plans and equipment operating status to support the prediction of gas production and consumption and the advance planning of peak shaving and valley filling operations for gas holders; Industrial Ethernet and wireless sensor network communication are adopted, and data encryption and verification technologies are used to ensure secure and reliable data transmission, thereby ensuring the accuracy and timeliness of data related to peak shifting and valley filling in gas holders.

[0021] The specific steps in step (2) are as follows: The 3σ criterion and box plot method were used to remove outliers from the gas holder operation data and gas production and consumption data, and duplicate records were deduplicated to avoid outliers affecting the judgment of the timing of peak shifting and valley filling of the gas holder. For missing values ​​in the gas holder operation data and production and consumption data, linear interpolation and Lagrange interpolation methods are used to fill them in based on the characteristics and correlation of time series data, ensuring data continuity and enabling continuous analysis of gas supply and demand trends, thus providing coherent data support for peak shaving and valley filling operations. Standardize and convert gas holder data and gas production and consumption data of different units and formats, and normalize or standardize them to the range of [0,1] or [-1,1] to facilitate the unified processing of gas holder peak shifting and valley filling related data in subsequent models. By linking coal gas production and consumption data with relevant data from different data sources such as gas holder operation data and electricity data, a complete dataset is constructed. The processed data is stored in a distributed database, with a focus on ensuring the storage quality and access speed of gas holder operation data and gas production and consumption data, so as to meet the need for rapid access to peak shifting and valley filling data in subsequent real-time scheduling.

[0022] The specific steps in step (3) are as follows: By employing a combination algorithm of long short-term memory network and gradient boosting tree, combined with production process and equipment status data, the model parameters are dynamically adjusted to make 1-hour and 4-hour predictions, accurately capturing the time series characteristics and nonlinear features of gas production and consumption, and predicting in advance when there will be a gas surplus and when there will be a gas shortage. Using a similar combined algorithm, combined with production plan, historical load and peak-to-valley time information, the demand for electricity load in each period can be predicted, providing a reference for optimizing the peak-shifting and valley-filling strategy of gas holders in conjunction with the electricity price mechanism. During peak electricity periods, gas can be released from the gas holders first to meet the gas demand and reduce the purchase of electricity. The root mean square error and mean absolute percentage error are calculated to evaluate the prediction results. If the accuracy is lower than the threshold, the model is retrained or the algorithm is adjusted to ensure the accuracy of the prediction of gas production and consumption, thus providing a guarantee for the accurate planning of peak shifting and valley filling operations of gas holders.

[0023] In step (4), the objective function is Minimize C=C1+C2+C3+C4, where C is the total energy cost, C1 is the cost of gas venting loss, C2 is the cost of purchased electricity, C3 is the operating cost of the gas holder, and C4 is the operating and maintenance cost of equipment such as gas turbines. Reducing the cost of gas venting loss and the cost of purchased electricity through the peak shifting and valley filling function of the gas holder is an important way to achieve the objective. The cost of purchased electricity is related to the peak, flat and valley electricity price. During peak hours, gas is released through the gas holder, i.e., peak shifting, to increase gas power generation and reduce purchased electricity. During valley hours, gas is stored through the gas holder, i.e. valley filling, to store excess gas for use during peak hours. The function constraints include gas flow balance, gas holder operation, equipment operation, power balance, and peak-valley electricity price. The scheduling model distinguishes between normal and abnormal production conditions. For normal conditions, an improved particle swarm optimization algorithm is used to solve the model, while for abnormal conditions, a three-layer decision tree model is constructed to support the model's rapid response.

[0024] In step (5), The gas inlet and outlet volumes of the gas holders at different times should be clearly defined, i.e., the valley filling operation and the peak shifting operation, as well as the gas distribution and supply of each piece of equipment, the output of the gas turbine equipment and the purchased electricity. Among these, the gas inlet and outlet volumes of the gas holders are the core content of the plan and need to be accurately determined based on the peak, flat and valley periods and the gas production and consumption forecast to ensure the timeliness and accuracy of the peak shifting and valley filling operations. It has multi-scenario analysis capabilities, simulating peak shifting and valley filling operation schemes for gas holders under different gas production and consumption fluctuations and electricity price changes, providing decision support and emergency plans for dispatchers, and ensuring that peak shifting and valley filling can be achieved through reasonable operation of gas holders under various circumstances; The dispatching scheme distinguishes between the specific implementation of normal and abnormal operating conditions. Under normal operating conditions, it is implemented in different electricity price stages. The priority of handling abnormal operating conditions is higher than that of electricity price time periods. Abnormal operating conditions adopt a hierarchical response mechanism with built-in contingency plans. A solution is generated within 3 seconds, prioritizing the basic peak shifting and valley filling functions of the gas holder.

[0025] The specific steps in step (6) are as follows: The key is to control the inlet and outlet of the gas holder and the opening of the pipeline regulating valves. According to the scheduling plan, the gas intake and output of the gas holder are precisely adjusted. During the valley filling stage, the gas intake is increased and during the peak shifting stage, the gas output is increased to achieve precise control of gas storage and flow distribution. Control the fuel supply and combustion intensity of the gas turbine and boiler equipment, and adjust the output to coordinate with the peak shifting and valley filling operation of the gas holder. When the gas holder releases gas, increase the gas supply of the gas turbine to increase the power generation. It communicates with the power sector's dispatch system and automatically adjusts the purchased electricity volume according to the peak-shaving and valley-filling situation of the gas holder, reducing the amount of purchased electricity when the gas holder releases a large amount of gas; It adopts PLC and DCS industrial control technology and has manual and automatic switching functions to ensure the accuracy of peak shaving and valley filling operation control of the gas holder and the stability of the system. In case of abnormality in automatic control, the gas holder can be manually operated to ensure that the peak shaving and valley filling function is not interrupted.

[0026] The specific steps in step (7) are as follows: By comparing the actual gas inlet and outlet volume and tank height with the target values ​​of the scheme, as well as the deviation between the actual gas production and consumption and the predicted values, the impact of the deviation on the peak shifting and valley filling effect and the reasons are analyzed. Based on the cause of the deviation, the scheduling model parameters are corrected or the gas holder's inlet and outlet gas scheduling scheme is directly adjusted and sent to the execution control module. The time granularity is consistent with the scheduling plan to ensure timely correction of deviations and maintain the optimal operating state of the gas holder's peak-shifting and valley-filling function.

[0027] In practical applications, the specific steps of implementing this invention are as follows: Step 1: Data Collection The data acquisition module collects various types of data in real time and transmits them to the data preprocessing module. Among these, data acquisition related to peak shaving and valley filling of gas holders is the key focus, as detailed below: Content collected: By using flow sensors, calorific value analyzers, etc., the generation, consumption, and calorific value of gas generated by equipment such as blast furnaces and coke ovens, as well as consuming equipment such as steel rolling heating furnaces, are collected. The sampling frequency is no less than once every minute, which provides a basis for judging the gas supply and demand status and deciding whether the gas holder should store or release gas.

[0028] By using liquid level and pressure sensors, the gas holder height, internal pressure, and gas flow rate are accurately collected to monitor its gas storage and venting capacity in real time. This data forms the basis for peak shaving and valley filling, and the sampling frequency is no less than 1 second per sampling.

[0029] Data on substation power load, purchased electricity, peak-valley electricity pricing time periods, and electricity prices for each time period (which can be obtained from the power sector and dynamically updated) are collected to provide a reference for adjusting the peak-shifting and valley-filling strategies of gas holders in conjunction with the electricity pricing mechanism.

[0030] Data collection on production plans and equipment operating status (such as blast furnace iron production and coking time in coke ovens) provides support for predicting gas production and consumption and planning peak-shifting and valley-filling operations for gas holders in advance.

[0031] The system employs communication methods such as industrial Ethernet and wireless sensor networks, and uses data encryption and verification technologies to ensure secure and reliable data transmission, thereby ensuring the accuracy and timeliness of data related to peak shaving and valley filling in gas holders.

[0032] Step 2: Data Preprocessing The data preprocessing module processes the raw data to provide high-quality data for subsequent analysis and decision-making related to peak shaving and valley filling of gas holders. The steps are as follows: Statistical methods such as the 3σ criterion and box plot method are used to remove outliers from gas holder operation data and gas production and consumption data, and duplicate records are deduplicated to avoid outliers affecting the judgment of the timing of peak shifting and valley filling of gas holders.

[0033] For missing values ​​in gas holder operation data, production and consumption data, linear interpolation, Lagrange interpolation and other methods are used to fill in the missing values ​​based on time series characteristics and correlations, ensuring data continuity and enabling continuous analysis of gas supply and demand trends, thus providing consistent data support for peak shaving and valley filling operations.

[0034] Standardize and convert gas holder data and gas production and consumption data of different units and formats, and normalize or standardize them to the range of [0,1] or [-1,1] to facilitate the unified processing of gas holder peak shifting and valley filling related data in subsequent models.

[0035] By linking coal gas production and consumption data with relevant data from different data sources such as gas holder operation data and electricity data, a complete dataset is constructed, which clearly presents the relationship between coal gas supply and demand and gas storage and release in gas holders, providing comprehensive data support for the formulation of peak shaving and valley filling strategies.

[0036] The processed data is stored in a distributed database, with a focus on ensuring the storage quality and access speed of gas holder operation data and gas production and consumption data, so as to meet the need for rapid access to peak shifting and valley filling data in subsequent real-time scheduling.

[0037] Step 3: Production, Consumption and Load Forecasting Based on preprocessed data, the prediction module uses a combined prediction algorithm to forecast future gas production and consumption, as well as electricity load demand, providing a basis for planning peak-shaving and valley-filling operations for gas holders in advance. By employing a combination algorithm of long short-term memory network and gradient boosting tree, and combining production process and equipment status data, the model parameters are dynamically adjusted to make 1-hour and 4-hour predictions. This accurately captures the time series characteristics and nonlinear features of gas production and consumption, and helps to predict in advance when there will be a gas surplus or a gas shortage.

[0038] Using a similar combined algorithm, combined with information such as production plans, historical load, and peak-to-valley periods, the demand for electricity load in each period can be predicted. This provides a reference for optimizing the peak-shifting and valley-filling strategy of gas holders in conjunction with the electricity pricing mechanism. During peak electricity periods, gas can be released from the gas holders first to meet the demand for gas, thereby reducing the need for external electricity purchases.

[0039] The system calculates indicators such as root mean square error and mean absolute percentage error to evaluate the prediction results. When the accuracy is lower than the threshold, the model is automatically retrained or the algorithm is adjusted to ensure the accuracy of the prediction of gas production and consumption, thus providing a guarantee for the accurate planning of peak shifting and valley filling operations of gas holders.

[0040] Step 4: Constructing the scheduling model The scheduling model construction module builds a dynamic scheduling model based on predicted and real-time data, with the goal of minimizing total energy costs and highlighting the peak-shifting and valley-filling function of the gas holder.

[0041] The objective function is Minimize C = C1 + C2 + C3 + C4 (C is the total energy cost, C1 is the cost of gas venting loss, C2 is the cost of purchased electricity, C3 is the operating cost of the gas holder, and C4 is the operating and maintenance cost of equipment such as gas turbines). Among them, reducing gas venting (reducing C1) and purchased electricity (reducing C2) through the peak shifting and valley filling function of the gas holder is an important way to achieve the objective. The cost of purchased electricity is related to the peak, flat and valley electricity prices. During peak hours, gas venting (peak shifting) from the gas holder increases gas-fired power generation and reduces purchased electricity. During off-peak hours, gas storage (valley filling) from the gas holder stores excess gas for use during peak hours.

[0042] The function constraints include gas flow balance, gas holder operation, equipment operation, power balance, and peak-valley electricity prices.

[0043] The gas flow balance is Q_prod + Q_out = Q_con + Q_in (Q_prod is the total production, Q_out is the gas output from the gas holder, reflecting the peak shifting function; Q_con is the total consumption, Q_in is the gas intake, reflecting the valley filling function). This constraint is the basic balance relationship for the gas holder to achieve peak shifting and valley filling.

[0044] When the gas holder is in operation, H_min≤H≤H_max (H is the height of the holder, H_min ensures that the gas holder has enough space for venting to achieve peak shifting, and H_max ensures that there is enough space for gas storage to achieve valley filling); Q_in≤Q_in_max, Q_out≤Q_out_max (the inlet and outlet flow rates are limited to ensure the safety and stability of valley filling and peak shifting processes).

[0045] The equipment operation must meet the technical requirements such as P_min≤P≤P_max (equipment output range) and the number of start-stop cycles and operating time. The equipment operation must be coordinated with the peak shifting and valley filling operation of the gas holder.

[0046] Power balance, P_gen + P_buy = P_load (P_gen is the total power generation, P_buy is the purchased power, and P_load is the total power load). Gas power generation needs to be adjusted according to the peak shifting and valley filling situation of the gas holder.

[0047] Peak-valley electricity pricing allows for the reasonable allocation of purchased electricity and gas-fired power generation at different times. Combined with the peak-shaving and valley-filling functions of gas holders, a limit is set on purchased electricity during peak hours, prioritizing the use of gas released from gas holders for power generation. During off-peak hours, purchased electricity can be increased appropriately, while allowing gas holders to store as much gas as possible.

[0048] The scheduling model needs to distinguish between normal production conditions and abnormal production conditions.

[0049] Under normal operating conditions, an improved particle swarm optimization algorithm is used to solve the model. This algorithm incorporates a dynamic adjustment mechanism for inertia weights, with an adjustment range of 0.4-0.9. It iteratively searches for the optimal solution, typically converging within 50 iterations. During the solution process, the gas storage and venting variables of the gas holder are given primary consideration; gas storage corresponds to valley filling, and venting corresponds to peak shifting. Simultaneously, a penalty coefficient for equipment start-up and shutdown is introduced into the objective function. When the number of daily start-ups and shutdowns exceeds a set threshold, the penalty coefficient increases accordingly, thereby preventing frequent start-ups and shutdowns from increasing costs and ensuring the efficient operation of the gas holder's peak-shifting and valley-filling functions.

[0050] A three-layer decision tree model is constructed to support rapid response under abnormal operating conditions. The first layer determines the fault type, the second layer assesses the impact on the peak shaving and valley filling function of the gas holder, and the third layer matches the corresponding contingency plan. Simultaneously, a multi-objective optimization model is introduced, adjusting the plan according to the priority of production continuity, energy cost, and equipment safety, with production continuity having a weight of 0.6, energy cost a weight of 0.3, and equipment safety a weight of 0.1. This setup prioritizes ensuring that the gas holder can still perform its basic peak shaving and valley filling functions under abnormal conditions, ensuring stable system operation.

[0051] Step 5: Generate a scheduling scheme Based on the model solution results, the scheduling scheme generation module generates the optimal scheduling scheme, which includes the specific operations of peak shifting and valley filling of the gas holder.

[0052] The gas inlet and outlet volumes of the gas holders at different times should be clearly defined, i.e., the valley filling operation and the peak shifting operation, as well as the gas distribution and supply of each piece of equipment, the output of equipment such as gas turbines, and the purchased electricity. Among these, the gas inlet and outlet volumes of the gas holders are the core content of the plan and need to be accurately determined based on peak, flat, and valley periods and gas production and consumption forecasts to ensure the timeliness and accuracy of peak shifting and valley filling operations.

[0053] It has multi-scenario analysis capabilities, simulating peak-shifting and valley-filling operation schemes for gas holders under different scenarios such as fluctuations in gas production and consumption and changes in electricity prices. It provides decision support and emergency plans for dispatchers, ensuring that peak-shifting and valley-filling can be achieved through the reasonable operation of gas holders under various circumstances.

[0054] The dispatching plan differentiates between normal and abnormal operating conditions. Under normal operating conditions, implementation varies depending on the electricity price period. For abnormal operating conditions, handling takes priority over electricity price period factors.

[0055] During off-peak electricity periods, when electricity prices are low and there may be a surplus of coal gas, the primary operation is filling the gas holders with gas. The gas intake of the gas holders is adjusted every 15 minutes to store as much excess gas as possible; the gas turbine load is reduced, and external power purchases are increased to reserve more gas for peak-shifting operations in the gas holders during peak periods. At the same time, instructions are given to coke ovens to extend coking time and sintering workshops to increase preheating furnace temperatures, in coordination with the filling of the gas holders with gas.

[0056] During periods of flat electricity prices, the peak-shaving and valley-filling operations of the gas holder are flexibly adjusted according to fluctuations in gas production and consumption. When production and consumption fluctuate, the output of the gas turbine and the purchased electricity are adjusted within 5 minutes, while the gas inlet or outlet of the gas holder is adjusted simultaneously; the PID algorithm is used to stabilize the holder position at 65%±3%, reserving sufficient space for peak-shaving and valley-filling operations during subsequent peak and valley periods; the gas turbine is preheated 1 hour in advance to ensure a rapid response to the peak-shaving operations of the gas holder when needed.

[0057] During peak electricity price periods, when electricity prices are high, peak shifting operations using gas holders are the primary method. Gas is allocated according to calorific value; when electricity prices surge, the output of gas turbines is increased within one minute, external power purchases are reduced, and the gas output from gas holders is increased to release previously stored gas to meet demand; real-time monitoring of gas consumption per unit of product and early warning of exceeding limits ensure the high efficiency of peak shifting operations using gas holders.

[0058] An abnormal operating condition adopts a graded response mechanism with built-in contingency plans, generating a solution within 3 seconds, prioritizing the basic peak shifting and valley filling functions of the gas holder.

[0059] Level 1 abnormal operating conditions include: blast furnace gas flow sensor failure, activation of backup sensor, estimation of output using historical data and related equipment data, and stabilization of gas holder inlet and outlet operations to avoid peak shifting and valley filling operation errors due to inaccurate data; gas holder inlet valve jamming, opening bypass valve and adjusting inlet volume calculation model to ensure valley filling operation can continue.

[0060] Level II abnormal operating conditions include a sudden drop in the calorific value of blast furnace gas, which leads to an increase in the proportion of coke oven gas, an increase in the inlet temperature of the gas turbine, and an extension of the heating time for steel rolling. At the same time, the gas output of the gas holder is adjusted to compensate for the impact of the drop in calorific value. In addition, the power grid voltage fluctuates, cutting off some non-critical loads and increasing the emergency output of the gas turbine. At this time, the gas holder prioritizes the supply of gas to critical equipment to ensure production continuity.

[0061] Level 3 abnormal operating conditions include gas holder leakage, main valve closure, nitrogen purging initiation, emergency equipment shutdown, and full external power supply. Although peak shifting and valley filling operations cannot be carried out at this time, it is necessary to quickly push out the leak location and handling plan to restore the gas holder function as soon as possible.

[0062] Step Six: Execution and Control The execution control module converts the peak shifting and valley filling operations of the gas holder in the scheduling plan into control commands and sends them to the execution equipment to achieve real-time control of the gas system.

[0063] The key is to control the inlet and outlet of the gas holder and the opening of the pipeline regulating valves, and to precisely adjust the gas intake and output of the gas holder according to the scheduling plan. During the valley filling stage, the gas intake is increased, and during the peak shifting stage, the gas output is increased to achieve precise control of the gas storage and flow distribution. This is the key to performing peak shifting and valley filling operations.

[0064] Controlling the fuel supply and combustion intensity of equipment such as gas turbines and boilers, and adjusting their output to coordinate with the peak-shifting and valley-filling operations of the gas holder, increases the gas supply to the gas turbine and improves power generation when the gas holder releases gas.

[0065] It communicates with the power sector's dispatch system and automatically adjusts the purchased electricity volume based on the peak-shaving and valley-filling situation of the gas holder, reducing the amount of purchased electricity when the gas holder releases a large amount of gas.

[0066] Employing industrial control technologies such as PLC and DCS, it features manual and automatic switching functions to ensure the accuracy and stability of peak shaving and valley filling operations in the gas holder. In case of abnormalities in automatic control, the gas holder's gas intake and exhaust operations can be performed manually to ensure that the peak shaving and valley filling function is not interrupted.

[0067] Step Seven: Feedback and Adjustment The feedback adjustment module monitors the execution effect and system operation status in real time, focusing on the actual effect of peak shaving and valley filling operations in the gas holder, forming a closed-loop control: By comparing the actual gas inlet and outlet volumes and tank height with the target values ​​of the scheme, as well as the deviations between the actual gas production and consumption and the predicted values, the impact of the deviations on the peak shifting and valley filling effect and the reasons for these deviations are analyzed.

[0068] Based on the cause of the deviation, the scheduling model parameters are corrected or the scheduling scheme, such as the gas holder's gas intake and gas output, is directly adjusted and sent to the execution control module. For example, when the gas production is higher than the forecast, causing the gas holder to approach its upper limit, the gas intake should be reduced or the gas output should be appropriately increased to ensure that the gas holder can always effectively play its role in peak shifting and valley filling.

[0069] The time granularity is consistent with the scheduling plan to ensure timely correction of deviations and maintain the optimal operating state of the gas holder's peak-shifting and valley-filling function.

[0070] This invention's system includes a data acquisition module, a data preprocessing module, a prediction module, a scheduling model construction module, a scheduling scheme generation module, an execution control module, and a feedback adjustment module. These modules work collaboratively to ensure the efficient operation of the gas holder's peak-shaving and valley-filling functions. The system can monitor gas production and consumption data in real time, accurately predict supply and demand changes, and, combined with peak-valley electricity pricing mechanisms, generate optimized schemes through a dynamic scheduling model. The core of this system lies in utilizing the gas holder's storage and release functions to store gas when production is excessive and release it when production is insufficient, thereby achieving peak-shaving and valley-filling, improving gas utilization efficiency, reducing emissions, and lowering energy costs.

Claims

1. A real-time scheduling system for gas using gas holder peak shaving, characterized in that: The application relates to a coal gas storage and consumption scheduling system, which comprises a data collection module, a data preprocessing module, a prediction module, a scheduling model construction module, a scheduling scheme generation module, an execution control module and a feedback adjustment module. The data preprocessing module performs outlier rejection, missing value filling, standardization and correlation processing on the collected data to form a high-quality data set. The prediction module predicts future coal gas production and consumption and power load demand based on the preprocessed data; the scheduling model construction module constructs a dynamic scheduling model aiming at minimizing total energy cost based on the prediction data and real-time data; the execution control module converts the scheduling scheme into control instructions to control the operation of the coal gas storage tank and related equipment; and the feedback adjustment module adjusts the scheduling scheme in real time according to the execution effect of the execution control module to ensure stable operation of the system.

2. A real-time scheduling method for coal gas by using coal gas tank peak shaving, characterized in that The application also discloses a coal gas storage and consumption scheduling method. In step (1), the collected data includes: The flow sensor and the heat value analyzer are used to collect the production amount, consumption amount and heat value of the blast furnace and coke oven gas generation equipment and the steel rolling heating furnace consumption equipment, and the sampling frequency is not less than 1 minute / time, so as to provide a basis for judging the coal gas supply and demand state and deciding whether the coal gas storage tank stores gas or discharges gas; The liquid level and pressure sensor is used to collect the tank position height, tank internal pressure and gas inflow and outflow, so as to realize real-time control of the gas storage capacity and gas discharge capacity of the coal gas storage tank, and the sampling frequency is not less than 1 second / time; The power substation power load and purchased power amount, peak-valley electricity price time period division and each time period electricity price are collected, so as to provide a reference for adjusting the peak-valley regulation strategy of the coal gas storage tank in combination with the electricity price mechanism; The production plan and equipment operation state are collected, so as to provide support for predicting the coal gas production and consumption and planning the peak-valley regulation operation of the coal gas storage tank in advance; The application also discloses a coal gas storage and consumption scheduling system. ​ 3. The real-time scheduling method for coal gas by using the coal gas tank peak load shifting according to claim 2, characterized in that: ​ ​ ​ ​ ​ Industrial Ethernet and wireless sensor network communication methods are adopted, and data encryption and checking technologies are used to ensure safe and reliable data transmission and the accuracy and timeliness of the data related to the gas tank peak load shifting.

4. The coal gas real-time scheduling method for peak load shifting by using the coal gas tank according to claim 2, characterized in that: In the step (2), the specific steps are as follows: The abnormal values in the gas tank operation data and the gas production and consumption data are removed by using the 3σ criterion and the box plot method, and the repeated records are removed, so as to avoid the influence of abnormal data on the judgment of the gas tank peak load shifting opportunity; For the missing values in the gas tank operation data and the gas production and consumption data, linear interpolation and Lagrange interpolation methods are used to fill in the missing values according to the time series characteristics and the correlation, so as to ensure the continuity of the data and provide continuous data support for the analysis of the gas supply and demand trend and the peak load shifting operation; The gas tank data and the gas production and consumption data in different units and formats are standardized and converted, and are normalized or standardized to the interval [0, 1] or [-1, 1], so as to facilitate the subsequent model for the unified processing of the data related to the gas tank peak load shifting; The related data of different data sources, such as the gas production and consumption data, the gas tank operation data and the power data, are associated to construct a complete data set; The processed data are stored in a distributed database, and the storage quality and access speed of the gas tank operation data and the gas production and consumption data are mainly guaranteed to meet the demand for the quick calling of the peak load shifting data in the subsequent real-time scheduling.

5. The coal gas real-time scheduling method for peak load shifting by using the coal gas tank according to claim 2, characterized in that: In the step (3), the specific steps are as follows: A combination algorithm of long short-term memory network and gradient boosting tree is used to dynamically adjust the model parameters in combination with the production process and the equipment state data, to perform 1-hour and 4-hour prediction, accurately capture the time series characteristics and nonlinear features of the gas production and consumption, and predict when the gas surplus and the gas shortage will occur; A similar combination algorithm is used to predict the power load demand in each period in combination with the production plan, the historical load and the peak and valley period information, to provide a reference for optimizing the gas tank peak load shifting strategy in combination with the electricity price mechanism. During the power peak period, the gas demand can be met by preferentially releasing the gas from the gas tank, and the purchased power can be reduced; The root mean square error and the mean absolute percentage error are used to evaluate the prediction results, and when the accuracy is lower than a threshold value, the model is retrained or the algorithm is adjusted, so as to ensure the prediction accuracy of the gas production and consumption and provide protection for the accurate planning of the gas tank peak load shifting operation.

6. The coal gas real-time scheduling method for peak load shifting by using the coal gas tank according to claim 2, characterized in that: In the step (4), the target function is Minimize C=C1+C2+C3+C4, where C is the total energy cost, C1 is the gas release loss cost, C2 is the purchased power cost, C3 is the gas tank operation cost, and C4 is the equipment operation and maintenance cost. Reducing the gas release loss cost and the purchased power cost through the gas tank peak load shifting function is an important way to achieve the target, and the purchased power cost is related to the peak and valley electricity price. During the peak period, the gas is released from the gas tank to shift the peak, the gas power generation is increased, and the purchased power is reduced. During the valley period, the gas is stored in the gas tank to fill the valley, and the excess gas is stored for use during the peak period. The function constraint conditions include the gas flow balance, the gas tank operation, the equipment operation, the power balance and the peak and valley electricity price. The scheduling model distinguishes between normal and abnormal working conditions, the normal working condition adopts an improved particle swarm optimization algorithm to solve the model, and the abnormal working condition constructs a three-layer decision tree model to support the model to respond quickly.

7. The coal gas real-time scheduling method for peak load shifting by using the coal gas tank according to claim 2, characterized in that: In the step (5), The gas intake and discharge of the gas holder, the gas distribution and supply of each device, the output of the gas turbine device, and the amount of purchased electricity are determined, wherein the gas intake and discharge of the gas holder are the core of the scheme and need to be accurately determined according to the peak, flat, and valley periods and the predicted gas production and consumption to ensure the timeliness and accuracy of the peak shifting and valley filling operation; The scheduling scheme has multi-scenario analysis function, simulates the peak shifting and valley filling operation scheme of the gas holder under different gas production and consumption fluctuation and electricity price change scenarios, provides decision support and emergency scheme for the dispatchers, and ensures that the peak shifting and valley filling can be realized through reasonable operation of the gas holder under various conditions; The scheduling scheme distinguishes between normal and abnormal working conditions, the normal working condition distinguishes between different electricity price stages, the abnormal working condition has higher priority than the electricity price period, the abnormal working condition adopts a hierarchical response mechanism, and the preplan is built-in to generate the scheme within 3 seconds to preferentially ensure the basic peak shifting and valley filling function of the gas holder.

8. The coal gas real-time scheduling method for peak load shifting by using the coal gas tank according to claim 2, characterized in that: In the step (6), the specific steps are as follows: The opening degrees of the inlet and outlet regulating valves of the gas holder are controlled, the gas intake and discharge of the gas holder are accurately adjusted according to the scheduling scheme, the gas intake is increased in the valley filling stage, the gas discharge is increased in the peak shifting stage, and the accurate control of the storage capacity and flow distribution is realized; The fuel supply and combustion intensity of the gas turbine and boiler devices are controlled, the output is adjusted, and the peak shifting and valley filling operation of the gas holder is matched, the gas supply of the gas turbine is increased when the gas holder discharges, and the power generation is increased; The purchased electricity is automatically adjusted according to the peak shifting and valley filling of the gas holder, and the purchased electricity is reduced when the gas holder discharges a lot of gas; PLC and DCS industrial control technology are adopted, manual and automatic switching functions are provided, the accuracy and system stability of the peak shifting and valley filling operation control of the gas holder are ensured, the gas intake and discharge of the gas holder can be manually operated when the automatic control is abnormal, and the peak shifting and valley filling function is not interrupted.

9. The coal gas real-time scheduling method for peak load shifting by using the coal gas tank according to claim 2, characterized in that: In the step (7), the specific steps are as follows: The actual gas intake and discharge of the gas holder, the height of the gas holder, and the target values of the scheme, and the deviation between the actual gas production and consumption and the predicted values are compared, the influence and reasons of the deviation on the peak shifting and valley filling effect are analyzed; According to the deviation reasons, the scheduling model parameters are corrected or the gas intake and discharge scheduling scheme content of the gas holder is directly adjusted and sent to the execution control module; The time granularity is consistent with the scheduling scheme to ensure that the deviation is corrected in time and the optimal operation state of the peak shifting and valley filling function of the gas holder is maintained.