Air separation unit multi-equipment energy consumption online optimization method based on clustering analysis

Abstract

The invention relates to the technical field of air separation unit energy consumption optimization, in particular to an air separation unit multi-equipment energy consumption online optimization method based on clustering analysis, and the method comprises the steps: monitoring a production process according to production planning data, obtaining production progress data, and predicting to obtain gas demand data; identifying the gas demand data through a clustering algorithm to obtain a gas demand subset, and determining a gas demand mode; obtaining time distribution data of the gas demand subset, and dividing a work cycle of the air separation unit to obtain a regulation and control cycle; by taking the energy consumption, the gas demand subset, the gas demand mode and the internal coupling parameters of the air separation unit in the regulation and control period as constraint conditions, constructing a first energy consumption analysis model; taking second energy consumption between the first energy consumption analysis model and the regulation and control period as a constraint condition, and constructing a second energy consumption analysis model; according to the second energy consumption analysis model, the air separation unit is regulated and controlled in real time.

Description

Technical Field

[0001] The invention relates to the technical field of energy consumption optimization of air separation units, and in particular to an online optimization method for energy consumption of multiple devices of an air separation unit based on cluster analysis. Background Art

[0002] Air separation units have the characteristics of high energy consumption and long cycle, and consume high energy during operation; the issue of energy consumption optimization has received increasing attention, and it is necessary to optimize the energy consumption of air separation units.

[0003] In the prior art, the energy consumption optimization of air separation units mainly adopts static optimization strategies based on physical modeling; by establishing the mechanism models of single equipment such as compressors and heat exchangers, offline simulation calculations are performed based on preset operating parameters. However, this type of solution has three key technical defects: the data processing dimension is single, only considering the steady-state parameters at the equipment level, and not constructing a multidimensional feature vector containing production plan data, user demand fluctuation curves, and environmental parameters; the computing architecture lacks real-time performance, and traditional numerical calculation methods are difficult to complete the solution of multivariable coupling equations within a time scale of seconds; the optimization strategy lacks adaptability, and when encountering changes in production schedules or sudden demand fluctuations, the control strategy cannot be updated through the online learning mechanism.

[0004] In the context of the rapid development of computer technology, data-driven optimization methods provide a new solution to the above problems. In particular, cluster analysis technology can effectively extract working condition feature patterns by reconstructing the feature space of multi-source heterogeneous data.

[0005] Therefore, an online optimization method for multi-equipment energy consumption of air separation units based on cluster analysis is proposed. Summary of the invention

[0006] The purpose of the present invention is to provide an online optimization method for energy consumption of multiple devices of an air separation unit based on cluster analysis, which monitors the production process according to production planning data to obtain production progress data and predicts gas demand data; identifies the gas demand data through a clustering algorithm to obtain a gas demand subset and determine the gas demand pattern; obtains the time distribution data of the gas demand subset, divides the working cycle of the air separation unit to obtain a control cycle; takes the energy consumption of the air separation unit within the control cycle, the gas demand subset, the gas demand pattern, and the internal coupling parameters of the air separation unit as constraints to construct a first energy consumption analysis model; takes the first energy consumption analysis model and the second energy consumption between the control cycle as constraints to construct a second energy consumption analysis model; and performs real-time control of the air separation unit according to the second energy consumption analysis model.

[0007] To achieve the above object, the present invention provides the following technical solution: an online optimization method for energy consumption of multiple devices in an air separation unit based on cluster analysis, comprising: S10. Obtain production planning data, monitor the production process according to the production planning data, and obtain production progress data; S20. Predicting based on production planning data and production progress data to obtain gas demand data; the gas demand data is time series data; S30. Identify the gas demand data by clustering algorithm and divide it into gas demand subsets; identify the gas demand subsets and determine the gas demand pattern; S40. Obtaining the time distribution data of the gas demand subset, dividing the working cycle of the air separation unit according to the time distribution data, and obtaining the control cycle; S50. Taking the energy consumption of the air separation unit within the control period, the gas demand subset, the gas demand pattern, and the internal coupling parameters of the air separation unit as constraints, a first energy consumption analysis model is constructed; S60. Construct a second energy consumption analysis model based on the first energy consumption analysis model and the second energy consumption between the control cycles as constraints; and perform real-time control of the air separation unit based on the second energy consumption analysis model.

[0008] The process of obtaining the gas demand data is as follows: Collecting production planning data, and monitoring the production process with the production planning data; The production process is divided according to the current moment to obtain a first period and a second period; the first period is the completed production process, and the second period is the production process to be completed; Acquire production data for a first period, including a first production parameter and a first gas demand parameter; determine production schedule data according to the first production parameter and production planning data; A second production parameter is predicted based on the production planning data and the production progress data; a second gas demand parameter is determined based on the second production parameter with reference to the first production parameter and the first gas demand parameter; and the second gas demand parameter is used as gas demand data.

[0009] The gas demand data includes gas demand type, gas demand intensity, gas fluctuation data and gas quality data; The process of identifying and obtaining the gas demand subset according to the gas demand data is as follows: The gas demand data is identified through a clustering algorithm, and cluster analysis is performed based on the gas demand type, gas demand intensity, gas fluctuation data, and gas quality data to obtain gas demand subsets.

[0010] The gas demand mode includes a stable demand mode, a transient demand mode and a fluctuating demand mode; The process of determining the gas demand mode is as follows: Identify the data features of gas demand type, gas demand intensity, gas fluctuation data and gas quality data in the gas demand subset to obtain the gas demand fluctuation coefficient; A gas demand fluctuation threshold is set to identify the gas demand fluctuation coefficient and determine the gas demand mode.

[0011] The construction process of the first energy consumption analysis model is as follows: The energy consumption, gas demand subset, gas demand pattern and internal coupling parameters of the air separation unit within the regulation cycle are used as constraints; including: The internal coupling parameters of the air separation unit are taken as the first constraint condition; The gas demand subset and the gas demand pattern are used as the second constraint; The energy consumption of the air separation unit within the regulation period is taken as the third constraint; A deviation threshold is set for the second constraint condition according to the gas demand pattern and the production schedule data.

[0012] The internal coupling parameters of the air separation unit are obtained according to the components and working process of the air separation unit, including: Acquire first coupling data of the air separation unit, where the first coupling data is determined by a coupling relationship between devices included in the air separation unit; Acquire second coupling data of the air separation unit, where the second coupling data is determined by a state change of air during operation of the air separation unit; The internal coupling data of the space separation unit is obtained according to the first coupling data and the second coupling data.

[0013] The construction process of the second energy consumption analysis model is as follows: Acquire first energy consumption data identified by a first energy consumption analysis model within an adjustment period of the air separation unit; Obtaining second energy consumption data generated by the transition between different gas demand modes according to historical data identification; The constraint condition is determined according to the first energy consumption data and the second energy consumption data, so as to determine the equipment parameters when the energy consumption is the minimum during the working cycle of the air separation unit.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention identifies and predicts production planning data and production progress data, divides the production process according to the current moment, and obtains a first period and a second period; the first period is a completed production process, and the second period is a production process to be completed; using the first production parameter and the first gas demand parameter as a reference, determining the second gas demand parameter according to the second production parameter; using the second gas demand parameter as gas demand data; the demand for gas can be accurately determined based on the production planning data and the production progress data.

[0015] 2. The present invention determines gas demand data through gas demand type, gas demand intensity, gas fluctuation data and gas quality data; then identifies the gas demand data according to a clustering algorithm, and determines a gas demand subset through data identification in each data dimension; accurately divides the gas demand data into stages; identifies the gas demand data of each stage to obtain a gas demand fluctuation coefficient, and determines a gas demand pattern; thereby accurately classifying different gas demand stages, providing a basis for energy consumption optimization.

[0016] 3. The present invention constructs a first energy consumption analysis model using the energy consumption of the air separation unit within the regulation cycle, the gas demand subset, the gas demand pattern, and the internal coupling parameters of the air separation unit as constraints; constructs a second energy consumption analysis model based on the first energy consumption analysis model and the second energy consumption between the regulation cycle as constraints; thereby accurately determining the equipment parameters with the lowest energy consumption within the working cycle of the air separation unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic flow chart of an online optimization method for energy consumption of multiple devices in an air separation unit based on cluster analysis according to the present invention; Figure 2 A schematic diagram of obtaining gas demand data according to the present invention; Figure 3 It is a schematic diagram of the structure of the second energy consumption analysis model of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Embodiment 1 The present invention proposes an online optimization method for energy consumption of multiple devices in an air separation unit based on cluster analysis, such as Figure 1 As shown, including: S10. Acquire production planning data, monitor the production process according to the production planning data, and obtain production progress data.

[0020] S20. Predictions are made based on the production planning data and the production progress data to obtain gas demand data; the gas demand data is time series data.

[0021] The process of obtaining the gas demand data is as follows: Collecting production planning data, and monitoring the production process with the production planning data; The production process is divided according to the current moment to obtain a first period and a second period; the first period is the completed production process, and the second period is the production process to be completed; Acquire production data for a first period, including a first production parameter and a first gas demand parameter; determine production schedule data according to the first production parameter and production planning data; The second production parameter is predicted based on the production planning data and the production progress data; the second gas demand parameter is determined based on the second production parameter with reference to the first production parameter and the first gas demand parameter; and the second gas demand parameter is used as the gas demand data. The process of obtaining the gas demand data is as follows: Figure 2 shown.

[0022] The present invention identifies and predicts production planning data and production progress data, divides the production process according to the current moment, and obtains a first period and a second period; the first period is a completed production process, and the second period is a production process to be completed; using the first production parameter and the first gas demand parameter as a reference, determining the second gas demand parameter according to the second production parameter; using the second gas demand parameter as gas demand data; the demand for gas can be accurately determined based on the production planning data and the production progress data.

[0023] S30. Identify the gas demand data by clustering algorithm and divide them into gas demand subsets; identify the gas demand subsets and determine the gas demand pattern.

[0024] The gas demand data includes gas demand type, gas demand intensity, gas fluctuation data and gas quality data; The process of identifying and obtaining the gas demand subset according to the gas demand data is as follows: The gas demand data is identified through a clustering algorithm, and cluster analysis is performed based on the gas demand type, gas demand intensity, gas fluctuation data, and gas quality data to obtain gas demand subsets.

[0025] Gas demand types refer to the types of separable gases in the air, such as oxygen, nitrogen and argon; gas demand intensity describes the average gas demand during the gas demand subset, including average flow rate and average air pressure; gas fluctuation data describes the temporal fluctuation characteristics of gas demand, including fluctuation amplitude and fluctuation frequency, to help optimize the timing and quantity of gas supply; gas quality data describes the composition, purity, humidity and other characteristics of the gas to ensure the smooth progress of the production process.

[0026] The present invention determines gas demand data through gas demand type, gas demand intensity, gas fluctuation data and gas quality data; then identifies the gas demand data according to a clustering algorithm, determines a gas demand subset through data identification in each data dimension; and accurately divides the gas demand data into stages.

[0027] The gas demand mode includes a stable demand mode, a transient demand mode and a fluctuating demand mode; The process of determining the gas demand mode is as follows: Identify the data features of gas demand type, gas demand intensity, gas fluctuation data and gas quality data in the gas demand subset to obtain the gas demand fluctuation coefficient; A gas demand fluctuation threshold is set to identify the gas demand fluctuation coefficient and determine the gas demand mode.

[0028] The calculation process of the gas fluctuation coefficient is: ; in, represents the gas fluctuation coefficient; represents the first gas fluctuation factor; represents the first gas fluctuation weight; represents the second gas fluctuation factor; Represents the second gas fluctuation weight.

[0029] The first gas fluctuation factor reflects the data difference between different gas demand subsets, and is obtained by balancing the time span of the current gas demand subset with the time span of the historical gas demand subset, and then according to the similarity of the data; The second gas fluctuation factor reflects the gas fluctuation within the current gas demand subset, and its calculation formula is: ; in, Indicates gas The demand intensity factor is obtained based on the gas demand intensity identification; Indicates gas The fluctuation characteristic factor is identified based on the gas fluctuation data; Indicates gas The quality fluctuation factor is identified based on the gas quality data; Indicates gas Internal volatility weights; Indicates the gas demand intensity threshold; Indicates the number of gas types.

[0030] The process of determining the gas demand pattern based on the gas demand fluctuation threshold is as follows: ; in, Indicates the gas demand pattern; represents the first gas demand fluctuation threshold; Indicates the second gas demand fluctuation threshold; when When the value is 1, it is a stable demand mode; when When the value is 2, it is a fluctuating demand mode; when When the value is 3, it is transient demand mode.

[0031] The present invention identifies data features of gas demand types, gas demand intensity, gas fluctuation data and gas quality data in a gas demand subset to obtain a gas demand fluctuation coefficient; sets a gas demand fluctuation threshold to identify the gas demand fluctuation coefficient and determines a gas demand pattern, including a stable demand pattern, a transient demand pattern and a fluctuating demand pattern; thereby accurately classifying different gas demand stages and providing a basis for energy consumption optimization.

[0032] S40. Obtain time distribution data of the gas demand subset, divide the working cycle of the air separation unit according to the time distribution data, and obtain a control cycle.

[0033] S50. Taking the energy consumption of the air separation unit within the regulation period, the gas demand subset, the gas demand pattern, and the internal coupling parameters of the air separation unit as constraints, a first energy consumption analysis model is constructed.

[0034] The construction process of the first energy consumption analysis model is as follows: The energy consumption of the air separation unit within the regulation period, the gas demand subset, the gas demand pattern, and the internal coupling parameters of the air separation unit are used as constraint conditions; including: taking the internal coupling parameters of the air separation unit as the first constraint condition; taking the gas demand subset and the gas demand pattern as the second constraint condition; taking the energy consumption of the air separation unit within the regulation period as the third constraint condition; and setting a deviation threshold for the second constraint condition according to the gas demand pattern and production progress data.

[0035] The internal coupling parameters of the air separation unit are obtained according to the constituent equipment and the working process of the air separation unit, including: obtaining first coupling data of the air separation unit, where the first coupling data is determined by the coupling relationship between the equipment contained in the air separation unit; obtaining second coupling data of the air separation unit, where the second coupling data is determined by the state change of air during the working process of the air separation unit; and obtaining the internal coupling data of the air separation unit according to the first coupling data and the second coupling data.

[0036] The present invention determines the first coupling data based on the coupling relationship between the devices contained in the air separation unit, determines the second coupling data based on the state change of the air during the operation of the air separation unit, and obtains the internal coupling parameters of the air separation unit; then uses the internal coupling parameters of the air separation unit as the first constraint condition; uses the gas demand subset and the gas demand pattern as the second constraint condition, and sets a deviation threshold for the second constraint condition according to the gas demand pattern and production progress data; uses the energy consumption of the air separation unit within the control cycle as the third constraint condition; thereby achieving optimal energy consumption within the control cycle in terms of technology for meeting gas demand.

[0037] S60. Construct a second energy consumption analysis model based on the first energy consumption analysis model and the second energy consumption between the control cycles as constraints; and perform real-time control of the air separation unit based on the second energy consumption analysis model.

[0038] The second energy consumption analysis model is composed as follows: Figure 3 As shown, the construction process is: Acquire first energy consumption data identified by a first energy consumption analysis model within an adjustment period of the air separation unit; obtain second energy consumption data resulting from transitions between different gas demand patterns based on historical data identification; determine constraint conditions based on the first energy consumption data and the second energy consumption data, thereby determining equipment parameters that minimize energy consumption within a working period of the air separation unit.

[0039] The present application obtains first energy consumption data identified by a first energy consumption analysis model within an adjustment period of an air separation unit; obtains second energy consumption data resulting from transitions between different gas demand patterns based on historical data identification; determines constraint conditions based on the first energy consumption data and the second energy consumption data, thereby accurately determining equipment parameters with minimal energy consumption within a working period of the air separation unit.

[0040] The present application monitors the production process according to the production planning data, obtains the production progress data, and predicts the gas demand data; identifies the gas demand data through a clustering algorithm, obtains the gas demand subset, and determines the gas demand pattern; obtains the time distribution data of the gas demand subset, divides the working cycle of the air separation unit to obtain the control cycle; constructs a first energy consumption analysis model with the energy consumption of the air separation unit within the control cycle, the gas demand subset, the gas demand pattern, and the internal coupling parameters of the air separation unit as constraints; constructs a second energy consumption analysis model with the first energy consumption analysis model and the second energy consumption between the control cycle as constraints; and performs real-time control of the air separation unit according to the second energy consumption analysis model.

[0041] Embodiment 2 As an indispensable core equipment in the metallurgical industry, the air separation unit mainly separates the main components in the air through cryogenic distillation or pressure swing adsorption. The air separation unit has the characteristics of high energy consumption, strong coupling and high safety risks, large operating fluctuations and long-term operation. The energy efficiency of the air separation unit has a significant impact on the overall production efficiency and energy consumption.

[0042] The present invention proposes an online optimization method for energy consumption of multiple devices of an air separation unit based on cluster analysis, which is used for online optimization of energy consumption of an air separation unit in a steel smelting process; the method comprises: S10. Acquire production planning data, monitor the production process according to the production planning data, and obtain production progress data.

[0043] Obtain the production planning data of steel smelting, monitor the production process according to the production planning data, and obtain the current production progress.

[0044] S20. Predictions are made based on the production planning data and the production progress data to obtain gas demand data; the gas demand data is time series data.

[0045] The process of obtaining the gas demand data is as follows: Collecting production planning data, and monitoring the production process with the production planning data; The production process is divided according to the current moment to obtain a first period and a second period; the first period is the completed production process, and the second period is the production process to be completed; Acquire production data for a first period, including a first production parameter and a first gas demand parameter; determine production schedule data according to the first production parameter and production planning data; The first production parameters include important parameters in the steelmaking process, including: Blast furnace iron production, usually measured by the weight of iron ore processed per hour; furnace temperature, the temperature inside the blast furnace, the amount of oxygen used is closely related to the temperature; furnace gas composition, the gas produced by the blast furnace contains carbon monoxide, carbon dioxide, nitrogen, etc., the composition and pressure of these gases are crucial to operational control, etc.; The first gas demand parameter is the gas demand data during the production process of the first production parameter. During the steelmaking process, the gas supply support provided by the air separation unit usually includes the following aspects: Oxygen supply: During the blast furnace smelting process, oxygen is blown into the furnace to improve combustion efficiency, increase furnace temperature, and promote the reduction reaction of iron ore; nitrogen supply: nitrogen is often used as cooling and protective gas in blast furnace operation to prevent metal oxidation in the furnace; argon supply: argon is used as a protective gas to isolate oxygen and other impurity gases in the air to prevent metal oxidation. At the same time, in the converter or electric furnace, argon can also protect the furnace lining from corrosion by high-temperature metal or slag.

[0046] A second production parameter is predicted based on the production planning data and the production progress data; a second gas demand parameter is determined based on the second production parameter with reference to the first production parameter and the first gas demand parameter; and the second gas demand parameter is used as gas demand data.

[0047] The present invention identifies and predicts production planning data and production progress data, divides the production process according to the current moment, and obtains a first period and a second period; the first period is a completed production process, and the second period is a production process to be completed; using the first production parameter and the first gas demand parameter as a reference, determining the second gas demand parameter according to the second production parameter; using the second gas demand parameter as gas demand data; the demand for gas can be accurately determined based on the production planning data and the production progress data.

[0048] S30. Identify the gas demand data by clustering algorithm and divide them into gas demand subsets; identify the gas demand subsets and determine the gas demand pattern.

[0049] The gas demand data includes gas demand type, gas demand intensity, gas fluctuation data and gas quality data; The process of identifying and obtaining the gas demand subset according to the gas demand data is as follows: The gas demand data is identified through a clustering algorithm, and cluster analysis is performed based on the gas demand type, gas demand intensity, gas fluctuation data, and gas quality data to obtain gas demand subsets.

[0050] Gas demand types refer to the types of separable gases in the air, such as oxygen, nitrogen and argon; gas demand intensity describes the average gas demand during the gas demand subset, including average flow rate and average air pressure; gas fluctuation data describes the temporal fluctuation characteristics of gas demand, including fluctuation amplitude and fluctuation frequency, to help optimize the timing and quantity of gas supply; gas quality data describes the composition, purity, humidity and other characteristics of the gas to ensure the smooth progress of the production process.

[0051] Table 1 is obtained by identifying the gas demand data of the steel smelting process.

[0052] Table 1 Gas demand data table

[0053] The present invention determines gas demand data through gas demand type, gas demand intensity, gas fluctuation data and gas quality data; then identifies the gas demand data according to a clustering algorithm, determines a gas demand subset through data identification in each data dimension; and accurately divides the gas demand data into stages.

[0054] The gas demand mode includes a stable demand mode, a transient demand mode and a fluctuating demand mode; The process of determining the gas demand mode is as follows: Identify the data features of gas demand type, gas demand intensity, gas fluctuation data and gas quality data in the gas demand subset to obtain the gas demand fluctuation coefficient; A gas demand fluctuation threshold is set to identify the gas demand fluctuation coefficient and determine the gas demand mode.

[0055] The calculation process of the gas fluctuation coefficient is: ; in, represents the gas fluctuation coefficient; represents the first gas fluctuation factor; represents the first gas fluctuation weight; represents the second gas fluctuation factor; Represents the second gas fluctuation weight.

[0056] The first gas fluctuation factor reflects the data difference between different gas demand subsets, and is obtained by balancing the time span of the current gas demand subset with the time span of the historical gas demand subset, and then according to the similarity of the data; The second gas fluctuation factor reflects the gas fluctuation within the current gas demand subset, and its calculation formula is: ; in, Indicates gas The demand intensity factor is obtained based on the gas demand intensity identification; Indicates gas The fluctuation characteristic factor is identified based on the gas fluctuation data; Indicates gas The quality fluctuation factor is identified based on the gas quality data; Indicates gas Internal volatility weights; Indicates the gas demand intensity threshold; Indicates the number of gas types.

[0057] The process of determining the gas demand pattern based on the gas demand fluctuation threshold is as follows: ; in, Indicates the gas demand pattern; represents the first gas demand fluctuation threshold; Indicates the second gas demand fluctuation threshold; when When the value is 1, it is a stable demand mode; when When the value is 2, it is a fluctuating demand mode; when When the value is 3, it is transient demand mode.

[0058] The present invention identifies data features of gas demand types, gas demand intensity, gas fluctuation data and gas quality data in a gas demand subset to obtain a gas demand fluctuation coefficient; sets a gas demand fluctuation threshold to identify the gas demand fluctuation coefficient and determines a gas demand pattern, including a stable demand pattern, a transient demand pattern and a fluctuating demand pattern; thereby accurately classifying different gas demand stages and providing a basis for energy consumption optimization.

[0059] S40. Obtain time distribution data of the gas demand subset, divide the working cycle of the air separation unit according to the time distribution data, and obtain a control cycle.

[0060] S50. Taking the energy consumption of the air separation unit within the regulation period, the gas demand subset, the gas demand pattern, and the internal coupling parameters of the air separation unit as constraints, a first energy consumption analysis model is constructed.

[0061] The construction process of the first energy consumption analysis model is as follows: The energy consumption, gas demand subset, gas demand pattern and internal coupling parameters of the air separation unit within the regulation cycle are used as constraints; including: The internal coupling parameters of the air separation unit are used as the first constraint condition; the gas demand subset and the gas demand pattern are used as the second constraint condition; the energy consumption of the air separation unit within the regulation cycle is used as the third constraint condition; according to the gas demand pattern and production progress data, a deviation threshold is set for the second constraint condition. The deviation threshold is used to balance the relationship between the production progress and the gas demand standard.

[0062] The internal coupling parameters of the air separation unit are obtained according to the components and working process of the air separation unit, including: Acquire first coupling data of the air separation unit, where the first coupling data is determined by the coupling relationship between devices contained in the air separation unit; acquire second coupling data of the air separation unit, where the second coupling data is determined by the state change of air during the operation of the air separation unit; and obtain internal coupling data of the air separation unit according to the first coupling data and the second coupling data.

[0063] Through the identification of the first coupling data and the second coupling data, coupling characteristic data are obtained, and the coupling characteristic data include: thermodynamic coupling, fluid network coupling, control logic coupling, safety constraint coupling, and material balance coupling.

[0064] Thermodynamic coupling involves the exchange of energy between different units, such as how the heat generated by a compressor is used by other parts, or how the temperature gradient in a distillation column affects the efficiency of other equipment.

[0065] The coupling of fluid networks may involve changes in pressure, flow and phase state, and requires analysis of the pressure balance and flow distribution of each node; for example, after compression, air enters the precooling system, then enters the molecular sieve, and finally enters the distillation tower.

[0066] Control logic coupling includes the need for coordinated control of the various subsystems of the air separation unit, such as pressure control, temperature control, liquid level control, etc.; these control loops may influence each other, for example, adjusting the reflux ratio of one tower may affect the pressure of another tower, and it is necessary to analyze the interaction and overall stability of the control strategies.

[0067] Safety constraint coupling involves the safe operation of the entire system, such as the safety valve setting of pressure vessels, explosion-proof measures for oxygen concentration, antifreeze protection for cryogenic equipment, etc. These safety measures need to be closely integrated with process parameters and control logic to ensure that the system can be safely shut down or adjusted under abnormal circumstances.

[0068] Material balance coupling refers to the conservation of mass of each component in the entire system; for example, the material balance of a distillation tower will affect product purity and recovery, and it is necessary to analyze the material flow and distribution between towers, as well as how to maintain balance by adjusting operating parameters.

[0069] The present invention determines the first coupling data based on the coupling relationship between the devices contained in the air separation unit, determines the second coupling data based on the state change of the air during the operation of the air separation unit, and obtains the internal coupling parameters of the air separation unit; then uses the internal coupling parameters of the air separation unit as the first constraint condition; uses the gas demand subset and the gas demand pattern as the second constraint condition, and sets a deviation threshold for the second constraint condition according to the gas demand pattern and production progress data; uses the energy consumption of the air separation unit within the control cycle as the third constraint condition; thereby achieving optimal energy consumption within the control cycle in terms of technology for meeting gas demand.

[0070] S60. Construct a second energy consumption analysis model based on the first energy consumption analysis model and the second energy consumption between the control cycles as constraints; and perform real-time control of the air separation unit based on the second energy consumption analysis model.

[0071] The construction process of the second energy consumption analysis model is as follows: Acquire first energy consumption data identified by a first energy consumption analysis model within an adjustment period of the air separation unit; Obtaining second energy consumption data generated by the transition between different gas demand modes according to historical data identification; The constraint condition is determined according to the first energy consumption data and the second energy consumption data, so as to determine the equipment parameters when the energy consumption is the minimum during the working cycle of the air separation unit.

[0072] In order to verify the energy consumption optimization effect of the present invention, a comparative scheme is selected to verify it; the comparative scheme is a static optimization strategy based on a physical model, which is calculated based on working conditions and parameters without considering dynamic changes; after multiple rounds of verification, Table 2 is obtained; wherein the production completion degree refers to the production progress completed within the specified time; the energy consumption comparison is to use the energy consumed by the present application obtained in each verification as a reference standard, and then obtain the ratio of the energy consumption of the comparative scheme for comparison.

[0073] Table 2 Comparison scheme verification data table

[0074] It can be seen from Table 2 that the method of the present application can optimize energy consumption in real time while ensuring the progress of production task completion, and the optimization effect is higher than that of the comparison scheme.

[0075] The present application obtains first energy consumption data identified by a first energy consumption analysis model within an adjustment period of an air separation unit; obtains second energy consumption data resulting from transitions between different gas demand patterns based on historical data identification; determines constraint conditions based on the first energy consumption data and the second energy consumption data, thereby accurately determining equipment parameters with the lowest energy consumption within a working period of the air separation unit.

[0076] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An online optimization method for energy consumption of multiple devices in an air separation unit based on cluster analysis, characterized in that: include: S10. Obtain production planning data, monitor the production process according to the production planning data, and obtain production progress data; S20. Predicting based on production planning data and production progress data to obtain gas demand data; the gas demand data is time series data; S30. Identify the gas demand data by clustering algorithm and divide it into gas demand subsets; identify the gas demand subsets and determine the gas demand pattern; S40. Obtaining the time distribution data of the gas demand subset, dividing the working cycle of the air separation unit according to the time distribution data, and obtaining the control cycle; S50. Taking the energy consumption of the air separation unit within the control period, the gas demand subset, the gas demand pattern, and the internal coupling parameters of the air separation unit as constraints, a first energy consumption analysis model is constructed; S60. Construct a second energy consumption analysis model based on the first energy consumption analysis model and the second energy consumption between the control cycles as constraints; and perform real-time control of the air separation unit based on the second energy consumption analysis model.

2. The method for online optimization of energy consumption of multiple devices in an air separation unit based on cluster analysis according to claim 1 is characterized in that: The process of obtaining the gas demand data is as follows: Collecting production planning data, and monitoring the production process with the production planning data; The production process is divided according to the current moment to obtain a first period and a second period; the first period is the completed production process, and the second period is the production process to be completed; Acquiring production data for a first period, including a first production parameter and a first gas demand parameter; Determine production schedule data according to the first production parameter and the production planning data; A second production parameter is predicted based on the production planning data and the production progress data; a second gas demand parameter is determined based on the second production parameter with reference to the first production parameter and the first gas demand parameter; and the second gas demand parameter is used as gas demand data.

3. The method for online optimization of energy consumption of multiple devices in an air separation unit based on cluster analysis according to claim 1 is characterized in that: The gas demand data includes gas demand type, gas demand intensity, gas fluctuation data and gas quality data; The process of identifying and obtaining the gas demand subset according to the gas demand data is as follows: The gas demand data is identified through a clustering algorithm, and cluster analysis is performed based on the gas demand type, gas demand intensity, gas fluctuation data, and gas quality data to obtain gas demand subsets.

4. The method for online optimization of energy consumption of multiple devices in an air separation unit based on cluster analysis according to claim 1 is characterized in that: The gas demand mode includes a stable demand mode, a transient demand mode and a fluctuating demand mode; The process of determining the gas demand mode is as follows: Identify the data features of gas demand type, gas demand intensity, gas fluctuation data and gas quality data in the gas demand subset to obtain the gas demand fluctuation coefficient; A gas demand fluctuation threshold is set to identify the gas demand fluctuation coefficient and determine the gas demand mode.

5. The method for online optimization of energy consumption of multiple devices in an air separation unit based on cluster analysis according to claim 1 is characterized in that: The construction process of the first energy consumption analysis model is as follows: The energy consumption, gas demand subset, gas demand pattern, and internal coupling parameters of the air separation unit within the regulation cycle are taken as constraints; including: The internal coupling parameters of the air separation unit are taken as the first constraint condition; The gas demand subset and the gas demand pattern are used as the second constraint; The energy consumption of the air separation unit within the regulation period is taken as the third constraint; A deviation threshold is set for the second constraint condition according to the gas demand pattern and the production schedule data.

6. The method for online optimization of energy consumption of multiple devices in an air separation unit based on cluster analysis according to claim 1 is characterized in that: The internal coupling parameters of the air separation unit are obtained according to the components and working process of the air separation unit, including: Acquire first coupling data of the air separation unit, where the first coupling data is determined by a coupling relationship between devices included in the air separation unit; Acquire second coupling data of the air separation unit, where the second coupling data is determined by a state change of air during operation of the air separation unit; The internal coupling data of the space separation unit is obtained according to the first coupling data and the second coupling data.

7. The method for online optimization of energy consumption of multiple devices in an air separation unit based on cluster analysis according to claim 1 is characterized in that: The construction process of the second energy consumption analysis model is as follows: Acquire first energy consumption data identified by a first energy consumption analysis model within an adjustment period of the air separation unit; Obtaining second energy consumption data generated by the transition between different gas demand modes according to historical data identification; The constraint condition is determined according to the first energy consumption data and the second energy consumption data, so as to determine the equipment parameters when the energy consumption is the minimum during the working cycle of the air separation unit.

Images