CN114429236A - Natural gas purification process control method - Google Patents
Natural gas purification process control method Download PDFInfo
- Publication number
- CN114429236A CN114429236A CN202011182496.0A CN202011182496A CN114429236A CN 114429236 A CN114429236 A CN 114429236A CN 202011182496 A CN202011182496 A CN 202011182496A CN 114429236 A CN114429236 A CN 114429236A
- Authority
- CN
- China
- Prior art keywords
- natural gas
- energy
- gas purification
- energy consumption
- consumption
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 title claims abstract description 288
- 238000000746 purification Methods 0.000 title claims abstract description 224
- 239000003345 natural gas Substances 0.000 title claims abstract description 144
- 238000000034 method Methods 0.000 title claims abstract description 88
- 238000005265 energy consumption Methods 0.000 claims abstract description 123
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 76
- 239000002994 raw material Substances 0.000 claims abstract description 59
- 239000000463 material Substances 0.000 claims abstract description 53
- 238000005457 optimization Methods 0.000 claims abstract description 53
- 230000008569 process Effects 0.000 claims abstract description 48
- 239000002245 particle Substances 0.000 claims abstract description 18
- 238000004519 manufacturing process Methods 0.000 claims abstract description 16
- 230000002068 genetic effect Effects 0.000 claims abstract description 12
- 238000012549 training Methods 0.000 claims abstract description 12
- 238000003062 neural network model Methods 0.000 claims abstract description 6
- 238000010801 machine learning Methods 0.000 claims description 9
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 239000000126 substance Substances 0.000 claims description 3
- 238000011217 control strategy Methods 0.000 abstract 1
- 239000007789 gas Substances 0.000 description 57
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 30
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 24
- 239000010410 layer Substances 0.000 description 23
- 210000002569 neuron Anatomy 0.000 description 17
- 238000013528 artificial neural network Methods 0.000 description 16
- 239000007788 liquid Substances 0.000 description 16
- 229910052717 sulfur Inorganic materials 0.000 description 16
- 239000011593 sulfur Substances 0.000 description 16
- 150000001412 amines Chemical class 0.000 description 15
- 238000004364 calculation method Methods 0.000 description 15
- 239000002737 fuel gas Substances 0.000 description 14
- NINIDFKCEFEMDL-UHFFFAOYSA-N Sulfur Chemical compound [S] NINIDFKCEFEMDL-UHFFFAOYSA-N 0.000 description 13
- 229910052757 nitrogen Inorganic materials 0.000 description 12
- 239000002253 acid Substances 0.000 description 11
- 230000006870 function Effects 0.000 description 11
- 238000010795 Steam Flooding Methods 0.000 description 9
- 238000009826 distribution Methods 0.000 description 7
- 239000008213 purified water Substances 0.000 description 7
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000011084 recovery Methods 0.000 description 6
- 230000009467 reduction Effects 0.000 description 6
- 239000003638 chemical reducing agent Substances 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000008929 regeneration Effects 0.000 description 5
- 238000011069 regeneration method Methods 0.000 description 5
- RWSOTUBLDIXVET-UHFFFAOYSA-N Dihydrogen sulfide Chemical compound S RWSOTUBLDIXVET-UHFFFAOYSA-N 0.000 description 4
- 238000013461 design Methods 0.000 description 4
- 229910000037 hydrogen sulfide Inorganic materials 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 239000002918 waste heat Substances 0.000 description 4
- PVXVWWANJIWJOO-UHFFFAOYSA-N 1-(1,3-benzodioxol-5-yl)-N-ethylpropan-2-amine Chemical compound CCNC(C)CC1=CC=C2OCOC2=C1 PVXVWWANJIWJOO-UHFFFAOYSA-N 0.000 description 3
- QMMZSJPSPRTHGB-UHFFFAOYSA-N MDEA Natural products CC(C)CCCCC=CCC=CC(O)=O QMMZSJPSPRTHGB-UHFFFAOYSA-N 0.000 description 3
- 229910002092 carbon dioxide Inorganic materials 0.000 description 3
- 239000001569 carbon dioxide Substances 0.000 description 3
- 238000005262 decarbonization Methods 0.000 description 3
- 230000018044 dehydration Effects 0.000 description 3
- 238000006297 dehydration reaction Methods 0.000 description 3
- 238000006477 desulfuration reaction Methods 0.000 description 3
- 230000023556 desulfurization Effects 0.000 description 3
- 230000005611 electricity Effects 0.000 description 3
- 238000004134 energy conservation Methods 0.000 description 3
- 239000000446 fuel Substances 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- RAHZWNYVWXNFOC-UHFFFAOYSA-N Sulphur dioxide Chemical compound O=S=O RAHZWNYVWXNFOC-UHFFFAOYSA-N 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 238000005984 hydrogenation reaction Methods 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000010206 sensitivity analysis Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000012897 Levenberg–Marquardt algorithm Methods 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000002745 absorbent Effects 0.000 description 1
- 239000002250 absorbent Substances 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 210000000349 chromosome Anatomy 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- -1 electricity Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000001704 evaporation Methods 0.000 description 1
- 230000008020 evaporation Effects 0.000 description 1
- 239000013505 freshwater Substances 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000013486 operation strategy Methods 0.000 description 1
- 125000001741 organic sulfur group Chemical group 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 239000012495 reaction gas Substances 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 229920006395 saturated elastomer Polymers 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 239000002356 single layer Substances 0.000 description 1
- 239000000779 smoke Substances 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- C—CHEMISTRY; METALLURGY
- C10—PETROLEUM, GAS OR COKE INDUSTRIES; TECHNICAL GASES CONTAINING CARBON MONOXIDE; FUELS; LUBRICANTS; PEAT
- C10L—FUELS NOT OTHERWISE PROVIDED FOR; NATURAL GAS; SYNTHETIC NATURAL GAS OBTAINED BY PROCESSES NOT COVERED BY SUBCLASSES C10G, C10K; LIQUEFIED PETROLEUM GAS; ADDING MATERIALS TO FUELS OR FIRES TO REDUCE SMOKE OR UNDESIRABLE DEPOSITS OR TO FACILITATE SOOT REMOVAL; FIRELIGHTERS
- C10L3/00—Gaseous fuels; Natural gas; Synthetic natural gas obtained by processes not covered by subclass C10G, C10K; Liquefied petroleum gas
- C10L3/06—Natural gas; Synthetic natural gas obtained by processes not covered by C10G, C10K3/02 or C10K3/04
- C10L3/10—Working-up natural gas or synthetic natural gas
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0633—Workflow analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Strategic Management (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Tourism & Hospitality (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Chemical & Material Sciences (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Oil, Petroleum & Natural Gas (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Quality & Reliability (AREA)
- Genetics & Genomics (AREA)
- Physiology (AREA)
- Chemical Kinetics & Catalysis (AREA)
- General Chemical & Material Sciences (AREA)
- Educational Administration (AREA)
Abstract
The invention relates to a natural gas purification process control method, and belongs to the technical field of natural gas purification. The method comprises the following steps: acquiring current raw material natural gas parameters, and optimizing by an intelligent algorithm to obtain the operation parameters of the natural gas purification device with the lowest process comprehensive energy consumption and the energy-carrying working medium demand under the raw material natural gas parameter condition; controlling a public project according to the required quantity of the energy-carrying working medium, and controlling a natural gas purification device according to the operation parameters of the natural gas purification device; the intelligent algorithm can adopt a genetic algorithm, a particle swarm algorithm and the like to carry out iterative optimization; the process comprehensive energy consumption can be obtained by predicting the neural network model after training the historical data. The method can guide the control strategy of the natural gas purification device and the public engineering facilities, reduce the total energy consumption and material consumption in the natural gas purification process and reduce the production cost.
Description
Technical Field
The invention relates to a natural gas purification process control method, and belongs to the technical field of natural gas purification.
Background
Natural gas is used as high-quality clean energy, and the specific gravity of the natural gas in the energy consumption structure of China is steadily increased year by year. However, the natural gas purification process consumes a large amount of energy, and particularly, the purification process of the natural gas with high sulfur content has great energy-saving potential.
The natural gas purification plant comprises a natural gas purification device and a public engineering facility, wherein the natural gas purification device is used for purifying natural gas containing impurities into qualified natural gas and meeting the emission standard of the whole process, and the public engineering facility is used for providing energy-carrying working media and energy sources required in the production process of the natural gas purification device, such as steam, electricity, fuel gas and the like, nitrogen, purified circulating water and the like.
The traditional natural gas purification device adopts a starting device to evenly distribute the processing amount of raw material natural gas (natural gas to be purified, hereinafter also referred to as raw material gas) and establishes mathematical models in ideal states of energy balance, material balance and the like, but the traditional method has larger deviation between calculated energy consumption and actual operation conditions, and has the problems that an efficient device does not fully play a role and the energy utilization rate is low; along with the prolonging of the operation time of natural gas purification devices, natural gas purification devices of public engineering system boilers, pumps and other energy consumption devices, energy-saving measures in the production process are mainly judged according to the experience of dispatchers on the demand and operation parameters of the natural gas purification devices, the amount of energy-carrying working media required by public engineering can be obtained, accurate control can not be achieved, energy utilization maximization can not be achieved, meanwhile, in order to prevent the problem that product gas cannot reach the standard due to insufficient energy supply, production is often carried out in an excessive supply mode, and the problem of energy waste exists. Especially for a purification series with a plurality of parallel connections, when the flow rate of the raw natural gas is lower than the working condition of full load design, the production scheduling is more difficult to master. The difficulty of the work is that the purification process comprises a plurality of process units such as acid gas removal, dehydration, sulfur recovery, tail gas treatment, acid water stripping and the like, the key parameters influencing energy consumption are numerous, the coupling relevance is strong, the influence on the system energy consumption is nonlinear, and the function relationship between the system energy consumption and the operation parameters is difficult to obtain by a classical mathematical regression analysis method. The operation parameters of the purification device firstly meet the treatment requirements of raw material gas, which are related to the flow and components of fuel gas, and the requirements of the purification device on energy-carrying working media, energy sources, circulating water and nitrogen are different under different operation parameters; different energy-carrying working medium requirements further require different public work operation parameters, such as operation parameters of a boiler and a power pump, and the change of the public work operation parameters further causes the change of the public work energy consumption. The energy saving problem is that if the energy consumption of the purification device (the energy and material consumption directly supplied by energy, circulating water, nitrogen and the like) is minimized by controlling only considering the flow rate and components of the raw gas, the energy consumption of the public engineering may be increased by the demand of the purification device on the energy-carrying working medium, so that the energy consumption of the whole natural gas purification process is difficult to be minimized.
Disclosure of Invention
The invention aims to provide a natural gas purification process control method, which is used for solving the problem of high energy consumption of the whole natural gas purification process.
In order to achieve the above object, the scheme of the invention comprises:
the invention discloses a natural gas purification process control method, which comprises the following steps:
1) acquiring current raw material natural gas parameters, and obtaining natural gas purification device operation parameters and energy-carrying working medium demand with lowest process comprehensive energy consumption under the raw material natural gas parameter condition through optimization of an intelligent algorithm B;
2) controlling a public project according to the required quantity of the energy-carrying working medium, and controlling a natural gas purification device according to the operation parameters of the natural gas purification device;
the process of searching the optimal by the intelligent algorithm B in the step 1) comprises the following steps: firstly, initializing a population, wherein the population comprises operation parameters of a natural gas purification device; then calculating population fitness, wherein the fitness is process comprehensive energy consumption which is calculated through a prediction model A; finally, changing the population, iterating for multiple times, and finally selecting the natural gas purification device operation parameters corresponding to the population with the lowest comprehensive energy consumption in the process and the corresponding energy-carrying working medium demand as an optimization result;
the prediction model A is a machine learning model and is obtained by training historical data of raw material natural gas parameters, natural gas purification device operation parameters, energy-carrying working medium demand, energy consumption and material consumption;
The comprehensive energy consumption of the process is the sum of the demand of energy-carrying working media, the energy consumption and the material consumption in the natural gas purification process.
The public works of the natural gas purification plant need to provide necessary energy-carrying working media for the purification process, and the energy-carrying working media are key points and difficulties in energy conservation and consumption reduction. The demand of the energy-carrying working medium in the purification process is related to parameters such as flow and pressure of the raw material natural gas to be treated and a plurality of operation parameters of the purification equipment, and the operation parameters of the purification equipment determine the power consumption, the fuel gas consumption and the consumption of materials such as water, nitrogen and the like of the purification equipment. The invention regards the consumption of the purifying equipment and the consumption of the public works as the total consumption based on the historical data, finds the lowest total consumption of the purifying equipment and the public works through an optimization algorithm, adjusts and controls the purifying device according to the state when the total consumption is the lowest, controls the public works to correspondingly provide the energy-carrying working medium required when the total consumption is the lowest, and effectively reduces the total consumption in the natural gas purifying process flow. The method breaks through the traditional natural gas purification plant, each purification series is controlled independently, the public works are also controlled independently, and the mode of over-supply according to experience is adopted according to the sum of the requirements of the energy-carrying working media of each series of purification devices. In the traditional mode, the energy consumption and the material consumption of each purification series can be controlled to a lower level, but the energy consumption of the whole plant is difficult to achieve the optimal level in combination with the energy consumption of public works. The method provided by the invention overcomes the traditional natural gas purification control idea, and combines the demand of energy-carrying working media to realize modeling optimization with the lowest total consumption as a target, thereby realizing reduction of energy consumption in the whole process of natural gas purification.
Further, in the step 2), the method for controlling the public works according to the energy carrying working medium demand comprises the steps of obtaining an optimized public work equipment operation parameter with the lowest energy consumption of the public work equipment when the energy carrying working medium demand is met through optimizing by an intelligent algorithm D, and controlling the corresponding equipment of the public works according to the optimized public work equipment operation parameter;
the optimizing process of the intelligent algorithm D comprises the following steps: firstly, initializing a population, wherein the population comprises public engineering equipment operation parameters; then calculating population fitness, wherein the fitness is the energy consumption of public engineering equipment, and the energy consumption of the public engineering equipment is calculated through a prediction model C; finally, changing the population and iterating for multiple times to finally select the public engineering equipment operation parameter corresponding to the population with the lowest public engineering equipment energy consumption as the optimized public engineering equipment operation parameter;
the prediction model C is a machine learning model and is obtained by training historical data of energy-carrying working medium supply quantity and public engineering equipment operation parameters corresponding to public engineering equipment energy consumption.
The invention controls the public works to provide energy-carrying working medium according to the energy-carrying working medium demand, for the public works, the output of the energy-carrying working medium is in direct proportion to the energy consumption and the material consumption of the public works, and the lower the energy-carrying working medium demand is, the lower the energy consumption and the material consumption of the public works are, but the invention further optimizes and finds out the public work control operation parameters which meet the current energy-carrying working medium demand and the lowest energy consumption of the public works (generally speaking, the lowest material consumption of the public works is when the energy consumption of the public works is lowest) under the condition of the lowest current energy-carrying working medium demand based on historical data, thereby further reducing the total energy consumption of the natural gas purification process.
Further, the prediction model C is a neural network model.
Further, the intelligent algorithm D is a particle swarm algorithm.
The prediction model C can be any machine learning model, the machine learning model is trained through a large amount of historical data, and the energy-carrying working medium supply quantity and the corresponding public engineering energy consumption under different public engineering operation parameters can be predicted based on the model.
The intelligent algorithm D can adopt an iterative optimization algorithm such as a genetic algorithm or a particle swarm algorithm.
Furthermore, the operation parameters of the natural gas purification device are parameters influencing the production of energy-consuming working media by the purification device.
Through the research on the natural gas purification process and the purification equipment, the operation parameters closely related to the demand of the energy-carrying working medium of the purification equipment in the purification process are found out, and the optimization with the lowest process comprehensive energy consumption is carried out only aiming at the parameters, so that the interference of the parameters without influence on the demand of the energy-carrying working medium is eliminated, the model is simplified, and the optimization operation efficiency is increased.
Furthermore, the sum of the demand of the energy-carrying working medium, the energy consumption and the material consumption is obtained by weighted summation of various consumed energy sources and substances according to respective energy conversion coefficients.
Further, the prediction model a is a neural network model.
Further, the intelligent algorithm B is a genetic algorithm.
The prediction model A can be any machine learning model, the machine learning model is trained through a large amount of historical data, and the demand of energy-carrying working media and other energy consumption and material consumption under different feed gas parameters and operation parameters of the purification device can be predicted based on the model.
The intelligent algorithm B can adopt an iterative optimization algorithm such as a genetic algorithm or a particle swarm algorithm.
Furthermore, when a plurality of purification series are operated in parallel under non-full load, the purification series with lower unit comprehensive energy consumption under the same raw material natural gas parameters undertake more purification tasks; the unit comprehensive energy consumption is the ratio of the process comprehensive energy consumption to the corresponding raw material natural gas treatment capacity.
The purification equipment with complex process design in the natural gas purification series is numerous, and the operation time of the purification equipment is different, so that certain individual difference exists between the purification series, and the energy consumption performance of each purification series is different (for example, the purification series with the same feed gas parameter in the purification process have higher energy-carrying working medium or other energy consumption and material consumption and lower energy consumption and lower material consumption), so that the purification series with better energy consumption performance can undertake more purification tasks after modeling and optimizing are respectively carried out on the basis of each purification series, and the energy consumption of the whole plant is further reduced after load optimization.
Drawings
FIG. 1 is a flow chart of an optimization algorithm B for process-oriented optimization of comprehensive energy consumption under certain feed gas parameters;
FIG. 2 is a flow chart of an optimization algorithm D for optimizing energy consumption of a utility under a certain energy-carrying working medium demand;
FIG. 3 is a flow chart of the method of the present invention;
FIG. 4 is a flow diagram of a purification process of a high sulfur natural gas purification plant complex;
FIG. 5 is a schematic diagram of a model structure of a typical artificial neural network;
FIG. 6 is a diagram showing the comparison of the predicted result of the neural network prediction model A with the actual value of the historical data;
FIG. 7 is a diagram showing the relative error distribution of all sample prediction results of the neural network prediction model A;
FIG. 8 is a schematic flow chart of an iterative optimization algorithm B based on a genetic algorithm and oriented to the lowest comprehensive energy consumption;
FIG. 9 shows a flow rate of 100kNm3A result schematic diagram of iterative calculation of a raw material natural gas with a pressure of 7.80Mpa by utilizing an optimization algorithm B;
FIG. 10 is a schematic diagram of the results of the integrated energy consumption calculations for a unit of integrated plant;
FIG. 11 is a graphical representation of the results of a pressure steam production calculation in a combined plant;
FIG. 12 is a graphical representation of the results of a combination low pressure steam consumption calculation.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
The method for controlling the natural gas purification process provided by the embodiment includes, as shown in fig. 1-2, 1.1) establishing prediction models A of energy-carrying working medium consumption such as steam generation and steam consumption and electric power, fuel gas, purified water, nitrogen consumption and the like of different purification device operation parameters in different fuel gas parameters based on historical data, 1.2) obtaining the purification device operation parameters and the energy-carrying working medium consumption when the total energy consumption is the lowest in the purification process based on prediction model A based on raw material gas parameters such as the treatment capacity of raw material natural gas to be treated, and adopting an algorithm B in the optimization process. 2.1) establishing a prediction model C of the energy-carrying working medium supply quantity of the public works corresponding to the operation parameters of the public works according to historical data, 2.2) optimizing to obtain the operation parameters of the public works with the lowest energy consumption based on the prediction model C according to the energy-carrying working medium consumption quantity (as the supply quantity of the public works) obtained in 1.2), and adopting an algorithm D in the optimizing process. The utility project generally comprises a plurality of devices which are operated in parallel and generate energy-carrying working medium, wherein the operation parameters of the utility project comprise the operation combination of the utility project devices (not all the devices are operated and put into use to generate the energy-carrying working medium) and the operation parameters of the corresponding devices (or the device is stopped and stops operating is also an operation parameter). Therefore, the historical data should include the energy-carrying working medium supply quantity, the corresponding operation combination of the public engineering equipment and the operation parameters of the corresponding equipment.
In the natural gas purification process, the flow rate (treatment capacity) of raw material natural gas, the pressure of the raw material gas, component data and the like are firstly obtained as raw material gas parameters, the operation parameters of the natural gas purification device are determined according to the working principle of the natural gas purification process, energy-carrying working media (such as low-pressure steam), energy sources (electric power and fuel gas) and materials (nitrogen and purified water) required by the natural gas purification device are judged according to the operation parameters and the operation state of the natural gas purification device, and then the energy-carrying working media, the energy sources and the materials required by public works are controlled and controlled to be supplied, the energy sources and the materials can be directly supplied according to the required quantity, but the steam in the energy-carrying working media needs to control and adjust the parameters of a boiler in the public works to obtain and supply the required steam.
In the method, as shown in fig. 3, 3.1-3.2) current raw material gas parameters are obtained and optimized by adopting an algorithm B, so that the operation parameters of the purification device, the energy-carrying working medium demand, the energy demand and the material demand are obtained when the total energy consumption of the whole purification process under the current raw material gas purification treatment is the lowest, wherein the lowest total process energy consumption means that the sum of the consumed energy-carrying working medium, the energy and the material is the lowest. And 3.3) optimizing the demand of the energy-carrying working medium of the purification device, which is obtained by optimizing the algorithm B, as the output by adopting an algorithm D, so as to obtain the operation parameters of the corresponding facilities (such as a boiler) of the public engineering when the energy consumption of the public engineering is the lowest (the material consumption is also the lowest) under the condition of meeting the demand of the energy-carrying working medium. 3.4) controlling and adjusting the purification device and related facilities of the public engineering by taking the operation parameters of the purification device and the operation parameters of the public engineering respectively obtained by the two times of optimization as control targets so that the corresponding parameters reach the control targets or are close to the control targets; and meanwhile, the requirements of the purification device obtained by the optimization algorithm B on energy and materials are met, and the energy material requirements of the energy-carrying working medium output by public works are met. And finally, the integrated optimization of the high-sulfur natural gas purification device and the energy consumption of public works is realized, and the prediction and control of the lowest energy consumption of the natural gas purification device are realized.
In the natural gas purification process, energy-carrying working media, electricity, fuel gas, nitrogen, purified water and the like required by a purification device are provided by public works, energy sources and materials such as electricity, fuel gas, nitrogen, purified water and the like can be accurately provided according to the quantity, and the steam and the like used as the energy-carrying working media need to regulate the yield through controlling a boiler, so that the public work operation parameters in the prediction model C and the energy-carrying working media supply quantity optimization algorithm D in the invention refer to corresponding operation parameters (such as the temperature of the boiler, the rotating speed of a water pump and the like) of equipment relevant to the generation of the energy-carrying working media; fuel gas, nitrogen, purified water and other energy sources and materials are directly supplied by public works according to the amount.
At the same time, it should be clear to a person skilled in the art that the regulation of the operating parameters towards the control target involves problems of hysteresis and oscillations, as well as possible reasonable errors in the model, and therefore the supply of energy-carrying working medium, energy source and material included in the control should have a certain redundancy, which is different from the overfeeding.
In particular, the method of the invention comprises 7 sections. The selected natural gas purification plant process is a typical high-sulfur natural gas purification process, and the purification process mainly comprises five main units and public works, namely MDEA deacidification gas, TEG dehydration, conventional Claus (Claus) sulfur recovery, hydrogenation reduction tail gas treatment, low-pressure acid water stripping and the like.
1. According to the raw gas parameters, the historical big data between the operating parameters of the natural gas purification device and the consumption of the energy-carrying working medium corresponding to the operating parameters, an energy-carrying working medium consumption prediction model A of the natural gas purification device is established, and the consumption of the energy-carrying working medium under different working conditions is predicted.
Taking a certain high-sulfur natural gas purification plant as an example, the main process flow of the purification process of the purification plant is shown in fig. 4, which shows a process flow of a combined device with two purification series (I and II), and the prediction model a and the optimization algorithm B are both directed at a complete purification series. The purification process of the natural gas with high sulfur content comprises the following steps: removing hydrogen sulfide, partial organic sulfur, carbon dioxide and the like from raw natural gas by using MDEA lean amine liquid through a deacidification unit; the product gas meets the requirement after being dehydrated by a dehydration unit and is output by a long-distance pipeline network; acid gas generated by the deacidification gas unit enters a sulfur recovery unit, is mixed with air and enters a reaction furnace to react to recover sulfur element in the acid gas into liquid sulfur, industrial sulfur is produced by a sulfur forming unit, and reaction gas in the process generates medium-pressure saturated steam by a waste heat boiler; tail gas of the sulfur recovery unit is treated by a tail gas treatment unit, meets the environmental protection requirement and then is conveyed to a tail gas incinerator for incineration, and smoke is discharged through a chimney; acid water generated in the purification process is sent to an acid water stripping unit, the generated purified water is recycled, and acid gas generated by stripping is transmitted to a tail gas treatment unit for treatment.
(1) Items of key influence parameters which have influence on the production and consumption of steam and energy and material consumption in the purification process are determined.
In the purification process of the embodiment, devices capable of generating steam are also arranged in the purification device, for example, some waste heat boilers can generate medium-pressure steam, and the medium-pressure steam can be converted into low-pressure steam, so that only a public project is needed to provide the residual required low-pressure steam. According to the natural gas purification process flow and the working principle, the established process flow simulation model is combined to carry out parameter sensitivity analysis, and key equipment and corresponding influence parameters related to energy consumption, and output and consumption of the concerned medium-pressure and low-pressure steam can be determined. Firstly, it can be confirmed that the parameters (flow, components, etc.) of the raw material natural gas to be processed can have important influence on the energy consumption and the demand of the energy-carrying working medium, so that all or part of the key influence parameters except the raw material gas parameters can be used as the operating parameters of the natural gas purification device to realize the control of the natural gas purification device.
Specifically, in the case of a deacidification unit, MDEA solution is widely used as an absorbent for acid gases. The process flow is summarized, the raw material natural gas is in countercurrent contact with lean amine liquid in an absorption tower to remove hydrogen sulfide and carbon dioxide, the rich amine liquid at the bottom of the tower enters an amine liquid regeneration tower after expansion, pressure reduction, flash evaporation and heat exchange with the lean amine liquid, and a reboiler consumes a large amount of low-pressure steam to desorb the hydrogen sulfide and the carbon dioxide absorbed in the rich amine liquid to regenerate the lean amine liquid. And the semi-lean amine liquid generated by the acid gas generated by the acid water stripping treatment of the tail gas treatment unit is also recycled to the amine liquid regeneration tower for regeneration. In view of steam consumption, the reboiler of the unit is a main consumption device of low-pressure steam, and after parameter sensitivity analysis is carried out on the basis of a simulation model, key parameters influencing low-pressure steam consumption are found to be lean amine liquid flow, a low-pressure steam flow/lean amine liquid ratio and the outlet temperature of an air cooler at the top of an amine liquid regeneration tower. The main energy consumption equipment and key parameter determination of other units are similar, and the detailed description is omitted here.
In summary, according to the basic principle of the purification process and the field operation parameters, the key influence parameters are determined as follows:
the method comprises the following steps of raw material natural gas flow, raw material natural gas pressure, raw material natural gas component proportion (if raw material natural gas components are the same and basically unchanged for a long time, the raw material natural gas components do not need to be taken as a key influence parameter), lean amine liquid flow of a desulfurization and decarbonization unit, low steam consumption of unit lean amine liquid flow of the desulfurization and decarbonization unit, outlet temperature of an air cooler of an amine liquid regeneration tower of the desulfurization and decarbonization unit, steam drum pressure of a Claus waste heat boiler of a sulfur recovery unit, molar ratio of tail gas hydrogen sulfide to sulfur dioxide of the sulfur recovery unit, air distribution ratio of a hydrogenation furnace of a tail gas treatment unit, tail gas treatment unit tail furnace hearth temperature, tail gas treatment unit tail furnace outlet oxygen content and tail gas treatment unit waste heat boiler outlet steam temperature. The raw material natural gas flow, the raw material natural gas pressure and the raw material natural gas component ratio are used as raw material natural gas parameters.
In the natural gas purification process of the embodiment, the purification unit, the equipment related to steam production and consumption, the medium and low pressure steam production and consumption conditions, and the key influence parameter conditions influencing the low pressure steam demand are shown in the following table.
Table 1 table for the steam production and consumption of the main unit of natural gas purification process
(2) And calculating the process comprehensive energy consumption and the process unit comprehensive energy consumption.
Calculating the unit integrated energy consumption of the purification device, which is the ratio of the total process energy consumption (integrated process energy consumption) to the raw natural gas treatment capacity, and taking the ratio asIdentification in MJ/Nm3The calculation expression is:
wherein M isRaw material gasIs the processing amount of natural gas as the raw material of the process, and the unit is 104Nm3/t;EProcedureThe energy consumption is the process comprehensive energy consumption, the unit is MJ/t, the comprehensive total consumption of energy-carrying working media, various energy sources and materials in the purification process is included, the comprehensive total consumption mainly comprises total power consumption, fuel gas consumption, nitrogen consumption, fresh water consumption, energy-carrying working media consumption and the like, the energy-carrying working media are obtained by weighting summation according to the national standard and respective energy conversion coefficients, and the energy-carrying working media are calculated by a formula (2):
wherein E is1、E2、E3…ENAnd (4) unifying the dimension weighted and converted values for the various types of consumption.
(3) A prediction model A of the energy-carrying working medium consumption and the energy consumption and material consumption of the natural gas purification device is established based on intelligent algorithm machine learning.
Selecting a historical data acquisition interval, such as 1 month to 3 months in 2019, and acquiring key influence parameters, energy-carrying working medium consumption of the purification device at corresponding time and production and consumption data of energy consumption and material consumption. Specifically, a set of corresponding data may be collected at set time intervals.
And (3) eliminating shutdown and abnormal data (if any) in the historical data, and obtaining a prediction model A which takes key influence parameters as independent variables and energy-carrying working medium demand and energy and material demand as dependent variables by adopting an artificial intelligence algorithm according to the effective historical data (the historical data of qualified natural gas produced after purification) of the single purification series. In the embodiment, the prediction model A is established by adopting an artificial neural network, which is a certain simplification, abstraction and simulation of a biological neural structure and is an empirical modeling tool. The method can learn complex input and output relations among data collected from specific problem domains, has good performance in accurate prediction and classification, and is widely used in multiple fields of engineering application at present. The multilayer perceptron artificial neural network is one of the most widely used neural networks in various fields of engineering problems, and comprises an input layer, a hidden layer and an output layer, wherein each layer is composed of neurons. The number of neurons in the input layer is equal to the number of input parameters, namely key influence parameters; the number of neurons in the output layer is equal to the number of prediction targets, wherein the prediction targets are energy-carrying working medium demand, energy demand and material demand. The hidden layer can be composed of one or more layers, and the number of the layers and the number of the neurons of the hidden layer can be obtained through trial and error or by combining intelligent algorithm optimization.
FIG. 5 illustrates a typical fully-connected network architecture, where the model receives data input through input layer nodes/neurons, passes it to hidden layer nodes, and finally passes the information to output nodes. The neuron connects to any neuron in the next layer through a communication link associated with a connection Weight (synthetic Weight). Each neuron receives the output of the neurons in the previous layer and performs weighted summation with the connection weight, and the deviation is added on the basis to obtain the single output of the neuron through calculation of an activation function. To adapt to a particular data/problem, a corresponding artificial neural network model needs to be configured and trained, and the training process can be viewed as minimizing the error between the desired output and the actual output of the model. The training of the artificial neural network model is a process of introducing randomly selected samples with input data and expected output into the artificial neural network configuration model, determining the error between the expected output value and the actual output of the model, and minimizing the error by modifying/optimizing the connection weight and deviation of the neurons.
The method comprises the steps that an artificial neural network prediction model A is adopted, the input of the model is a key influence parameter, and the output of the model is predicted energy-carrying working medium demand, energy demand and material demand. The method is used for establishing model samples from historical operating data of 2019 in 1-3 months, and a group of effective historical data is extracted every hour. Randomly dividing the obtained sample data into three types: the percentage of the test set is 70%, 15% and 15% respectively as a training set, a validation set and a test set.
Before sample data is input into the neural network model for training, input data is normalized to the range of [ -1,1] using the minmax function in Matlab. The transfer function of the hidden layer is an S-shaped function 'tandig', the transfer function of the output layer is a linear function 'purelin', the training function of the network adopts a 'trainlm' function, and the function updates the weight and the deviation value based on a Levenberg-Marquardt algorithm. A model is established by adopting a single-layer hidden layer network, the number of neurons in the optimal hidden layer is found by using a trial-and-error method, the Mean Square Error (MSE) is adopted as a network performance function, and the correlation between expected data and actual network output is measured by a regression R value. Because the initial values of the samples used for training the network, the network weight and the deviation are randomly selected and generated, and the training results of the neural networks with the same structure are different, each structure of the neural networks is trained for 5 times, the mean square error value is taken, and the mean square error values calculated by the neural networks with different structures are shown in table 2. It can be seen that when the number of hidden layer neurons is 18, both the mean squared error value and the correlation coefficient value of the test set are ideal and therefore selected for subsequent studies.
Finally constructing a neural network prediction model A with an 11-18-5 structure; 11 is the number of neurons in the input layer, and corresponds to the input quantity of 11 key influence parameters; 18 is the number of hidden layer neurons; and 5 is the number of neurons in an output layer, and corresponds to five output quantities, namely energy-carrying working medium demand, power consumption, fuel gas demand, purified water demand and nitrogen demand which are used as energy consumption and material consumption.
TABLE 2 neural network output calculation results under different neuron number design of hidden layer
After verification is carried out by utilizing the test set and the verification set, in order to quantify the difference between the energy-carrying working medium consumption and the energy consumption and material consumption prediction model A prediction result and the true value, the average relative deviation (AAD%) is defined and calculated by the following formula, wherein yi,xiAnd n represents the true value, the calculated value of the network model, and the number of samples, respectively.
And calculating the Mean Square Error (MSE) of all samples according to the model A to be 0.005, wherein the average relative deviation is 1.9%, the relative deviation is 33.7% of the number of samples in the interval of [ -1%, 1% ], 79.1% of the number of samples in the interval of [ -3%, 3% ], 94.8% of the number of samples in the interval of [ -5%, 5% ], and the model precision is reliable and can be used as engineering application. FIG. 6 is a comparison graph of the predicted value of the integrated energy consumption of the process unit versus the actual historical value under different operating conditions, i.e., different key operating parameters. FIG. 7 shows the error rate of the predicted value relative to the error distribution within. + -. 5%, and the overall prediction accuracy is good.
2. And (2) establishing an optimization algorithm B based on the prediction model A obtained in the step (1), and obtaining operation parameters of the purification device, energy-carrying working medium consumption (energy-carrying working medium demand) of the purification device, energy demand and material demand in the raw material gas purification process with optimal energy consumption by using the iterative optimization algorithm B under the given raw material gas parameters.
The embodiment uses a genetic algorithm to perform iterative optimization, and the algorithm flow is shown in fig. 8. The basic elements of the genetic algorithm mainly comprise: chromosome coding, population initialization, individual fitness calculation, selection operation, crossover operation and mutation operation aiming at a specific problem to be solved. The genetic algorithm starts from an initial population, and finally converges to one or more individuals with the best fitness through multi-generation evolution, so that the optimal solution or the satisfactory solution of the problem is obtained.
In this embodiment, individuals in the initial population correspond to different key operating parameters (the key operating parameters include a feed gas parameter and a purification device operating parameter), the fitness is the process comprehensive energy consumption defined in step 1(2), the prediction model a and the formula 1 are fitness functions, and the feed gas parameter in the key influence parameters does not participate in variation and serves as a constraint condition; calculating the individual adaptability, comparing the individual adaptability values, and changing the operation parameters of the purification devices in the individuals through cross variation until the minimum adaptability value is found, namely the operation parameters of the purification devices corresponding to the lowest unit comprehensive energy consumption under different raw material gas parameters, the energy-carrying working medium demand of the purification devices, the energy demand and the material demand.
The initial population number, the iteration number, the cross probability and the mutation probability of the genetic algorithm designed in this embodiment are set to 300, 200, 0.5 and 0.05, respectively. With a raw natural gas throughput of 100kNm3For example, the variation curve of the optimal fitness value in the iterative calculation process of the genetic algorithm is shown in fig. 9, and it can be seen that when the iteration frequency reaches 130 generations, the calculation result tends to converge and remain stable, and the output result after the iteration is finished is the feed gas parameter of 100kNm3The optimal energy-carrying working medium consumption, energy consumption and material consumption under the pressure of 7.80MPa, and the variation result of the corresponding key operating parameters (the operating parameters of the purification device) is the operating parameters of the purification device reaching the optimal unit comprehensive energy consumption.
3. And (3) utilizing a genetic algorithm-based optimization algorithm B to obtain the optimal natural gas purification device operation parameters, and the corresponding energy-carrying working medium consumption, energy consumption and material consumption under the raw material gas parameter conditions such as the raw material natural gas treatment capacity in the actual operation of the natural gas purification plant.
The actual operation raw material natural gas treatment capacity of the natural gas purification plant at present is distributed to each purification series of the combination device, each purification series obtains the optimal process comprehensive energy consumption and process unit comprehensive energy consumption under the corresponding treatment capacity of each purification series through an optimization algorithm B facing the lowest comprehensive energy consumption of the natural gas purification device according to the respective treatment capacity and other raw material gas parameters, and FIG. 10 shows that under different raw material natural gas flow rates, the total optimal process unit comprehensive energy consumption of the combination device is obtained through calculation by the optimization algorithm B, and the total optimal process unit comprehensive energy consumption is shown in black data points. In field engineering application, according to the calculation result of the optimization algorithm, the relationship between the raw natural gas flow and the unit energy consumption of the process can be fitted, and the fitted curve of the embodiment is shown as a curve in fig. 10. The curve shows that the unit comprehensive energy consumption of the combined device is reduced along with the increase of the treatment capacity of the raw material natural gas, and the trend is very consistent with the actual operation on site.
The essence of the energy-carrying working medium consumption or demand of the purification device, namely the demand of the steam, is that the total steam demand of the purification device in the purification process subtracts the steam generation amount of the purification device in the purification process. Fig. 11 and 12 are optimized values of the medium pressure steam amount output and the low pressure steam amount consumed in the corresponding purification process obtained based on the optimization algorithm B and fitted curves under different raw material gas flow rates.
The optimization algorithm B for the lowest comprehensive energy consumption of the natural gas purification device constructed by the invention aims at one purification series, I or II in fig. 4. When a plurality of purification series operate in parallel, the optimal purification device operation parameters, energy-carrying working medium consumption and energy material consumption under the raw material gas parameters such as the raw material gas flow and the like processed by the purification series are calculated respectively for each purification series. Because of the difference between the purification series, the result of optimizing according to the optimizing algorithm B according to the historical operation data of each purification series has larger difference, namely the operation parameters, the energy-carrying working medium consumption and the energy material consumption of the optimal purification device of different purification series are different under the same raw material gas parameter. Therefore, when a natural gas purification plant with a plurality of purification series is operated at a non-full load, for the purpose of saving energy to the maximum extent, the load optimization distribution principle of different series is as follows: more natural gas purification tasks should be undertaken by a purification series with low unit integrated energy consumption at the same feed gas flow rate, but not exceeding its maximum processing capacity.
According to the method, the load optimization distribution scheme of three sets of 6 series of combined devices of a certain purification plant under different raw material natural gas loads is calculated. The specific calculation results are shown in table 3. Compared with actual operation data, the energy-saving potential under different series of corresponding treatment capacities under the condition of Table 3 is 4-10% according to the optimized distribution.
TABLE 3 calculation results of optimization of the load distribution of the combination units under different total processing amounts
Total amount of treatment | 111 columns | 112 columns | 121 column (c) | 122 rows | 131 rows of | 132 columns |
kNm3/h | kNm3/h | kNm3/h | kNm3/h | kNm3/h | kNm3/h | kNm3/h |
400 | 109.5 | 104.8 | 0 | 0 | 66.8 | 119 |
500 | 129.5 | 125.3 | 0 | 117.7 | 0 | 127.5 |
600 | 125.7 | 121.7 | 0 | 109.6 | 114.9 | 128.0 |
700 | 123.1 | 122.3 | 95.9 | 112.9 | 121 | 124.7 |
After the load (the processing capacity, namely the raw material gas flow) of each purification series of the combination device is distributed, the raw material gas parameters are brought into the optimization algorithm B of each purification series, the optimal purification device operation parameters, the energy-carrying working medium demand, the energy source demand and the material demand of each purification series are calculated, each purification series are controlled according to the optimal purification device operation parameters, a natural gas purification plant is usually provided with a public engineering system, therefore, the public engineering is controlled according to the sum of the energy-carrying working medium demands of each purification series to generate energy-carrying working media, the energy-carrying working media are distributed to each purification series according to the calculated demand, and meanwhile, the energy source demand and the material demand are provided and distributed according to the calculated energy source demand and the calculated material demand.
Table 4 shows the low pressure steam production in a combined plant (two purification trains).
TABLE 4 Low pressure steam production and consumption in the combined plant
4. And establishing a prediction model C of the operation parameters of the corresponding public engineering equipment under different energy-carrying working medium supply quantities by using the operation parameters of the public engineering equipment and historical big data of the flow of the supplied energy-carrying working medium.
Firstly, determining key public engineering equipment and operation parameters related to the steam generation amount of the energy-carrying working medium. A typical boiler energy balance relationship is shown in equation (3). Wherein, FBu,c,tAnd SBu,c,tRespectively representing boiler fuel gas consumption and steam production, HcIndicating fuel enthalpy of fuel c, i.e. low heating value of natural gas, etau,cRepresents the energy efficiency, H, of the boiler u at the time of consuming the fuel cstmAnd HwatIndicating the enthalpy of the boiler outlet steam and the inlet boiler water.
FBu,c,t·Hc·ηu,c=(Hstm-Hwat)SBu,c,t (3)
The steam drive equipment of the public engineering comprises a water feed pump steam drive, a circulating water field steam drive and an air separation air compression steam drive, and is matched with matched electric drive equipment to provide driving force required by boiler feed water, circulating water and compressed air together, and energy consumption models of a boiler feed water pump unit, a circulating water pump unit and an air compressor unit are expressed as follows:
∑u∈UBu,c,t·FTu,c,t(histm-hostm)ηu,c+∑v∈VBv,c,t·ETv,c,t·ηv,c=XETc,t (4)
wherein FTu,c,tRepresenting the steam consumption, hi, of the turbine u stmAnd hostmTo represent the enthalpy values of the inlet and outlet steam, respectivelyu,cIndicating the steam-drive efficiency, ETv,c,tIndicating indicated power, η, of the motorv,cIndicating the efficiency of the motor, XETc,tRepresenting the shaft power required for driving the boiler feed water, circulating water or compressed air, Bu,c,tAnd Bv,c,tIs a variable from 0 to 1.
The material balance and energy conservation equations of the temperature and pressure reducer are shown as (5) and (6):
FOu,c,t=FIu,c,t+FI′u,c,t (5)
FOu,c,t·hostm=FIu,c,t·histm+FI′u,c,t·hiwat (6)
in the formula, FIu,c,tFor a period t of the flow rate of steam (and high-pressure steam) at the inlet of the temperature and pressure reducer, histmIs the enthalpy of the inlet steam, FOu,c,tFor the flow of the steam (i.e. low-pressure steam) at the outlet of the temperature and pressure reducer hostmIs the enthalpy of the outlet steam, FI'u,c,tAnd hiwatRespectively the flow rate and enthalpy of water needed by the temperature and pressure reducing device.
In the equipment, the efficiency of the boiler, the steam drive and the pump is given by fitting historical big data. The modeling method adopts a neural network algorithm, and the process is similar to the purification process and is not described again.
The operation parameters of the public engineering equipment related to the generation of the energy-carrying working medium, namely the steam, found in the embodiment comprise the consumption of the boiler fuel gas of the power station, the operation state of the pump, the efficiency of the pump, the steam flow of the temperature and pressure reducer, the steam consumption of the steam turbine, the power consumption of the power consumption equipment and the like. Historical big data of the flow of the energy-carrying working medium supplied and discharged by the public works comprise the flow of low-pressure steam.
Selecting a historical data acquisition interval, such as 1-3 months in 2019, and acquiring the key equipment of the public engineering, the operation parameters of the key equipment and the amount of the energy-carrying working medium output in the public engineering at the corresponding time.
And establishing a prediction model C of the energy-carrying working medium supply quantity corresponding to the key equipment and the operation parameters of the public engineering according to the historical big data.
5. And establishing an optimization algorithm D based on the prediction model C of the energy-carrying working medium consumption of the public engineering equipment, and obtaining the public engineering key equipment with optimal energy consumption and operation parameters by utilizing the optimization algorithm D when the energy-carrying working medium demand is given.
The present embodiment adopts a Particle Swarm Optimization (PSO) algorithm. The particle swarm algorithm is an evolutionary computing technology and is derived from the behavior research of bird swarm predation. The basic idea is as follows: the optimal solution is found through cooperation and information sharing among individuals in a group. The basic principle is as follows: the PSO is initialized to a population of random particles (random solution) and then an optimal solution is found by iteration. In each iteration, the particle updates itself by tracking two "extrema" (pbest). After finding these two optimal values, the particle updates its velocity by equation (7) below. The first part of the equation represents the effect of the last speed magnitude and direction; the second part is a vector pointing to the best point of the particle from the current point and represents the part of the motion of the particle from own experience; the third part is a vector pointing from the current point to the best point of the population, reflecting the cooperative cooperation and knowledge sharing among the particles. c1 and c2 are self learning factor and group learning factor, respectively, and rand is a random number between 0 and 1.
vi=vi+c1×rand()×(pbesti-xi)+c2×rand()×(gbesti-xi) (7)
On this basis, the position update of the particles in the continuous particle swarm is realized by the following equation.
xi=xi+vi
And the position update of the particles in the discrete particle swarm is realized by the following formula.
In the design of the embodiment, the output conditions of three gas boilers adopt continuous variables, a medium-pressure boiler feed water pump (total 2 steam drives and 3 electric drives), an air compression station compressor (total 3 steam drives and 1 electric drive), and a circulating water pump (total 7 steam drives and 7 electric drives) adopt binary variables, the energy consumption of the public engineering can be calculated according to key equipment and equipment operation parameters which participate in the operation to generate the energy-carrying working medium, or can be obtained according to energy consumption data in historical data, if the energy consumption is obtained according to the historical data, the historical statistical data of the energy consumption of the public engineering corresponding to the control parameters of the public engineering is added into the prediction model C in the step 4, namely the operation parameters of the public engineering equipment corresponding to different energy-carrying working medium supply quantities and the prediction model C of the energy consumption of the public engineering are established in the step 4. And (4) optimally solving by using the lowest comprehensive energy consumption of the public works as a fitness objective function through a particle swarm optimization algorithm, namely an optimization algorithm D with the lowest energy consumption of the public works.
In the step 3, the method calculates the optimal operation parameters of the purification device, the energy-carrying working medium demand, the energy demand and the material demand of the natural gas purification plant under the corresponding working conditions (namely, the feed gas parameters) of each purification series, substitutes the total energy-carrying working medium demand into the optimization algorithm D with the lowest energy consumption of the public engineering, and obtains the operation combination of the public engineering equipment with the optimal energy consumption, the operation state of key equipment, the flow control and other operation parameters. And taking the obtained public engineering operation parameters as a control target, adjusting and controlling the corresponding equipment according to the optimal public engineering equipment combination to approach and reach the control target, namely meeting the energy-carrying working medium requirements of the purification series under the optimal energy consumption condition.
The operation combination of the key equipment of the public engineering under different low-pressure steam supply amounts (energy-carrying working medium supply amounts) and the corresponding raw material gas amount capable of being processed are obtained by calculation and are shown in table 5.
TABLE 5 Combined results of key plant operations in utilities at different natural gas throughputs
In general, the method of the invention needs to establish an optimization algorithm model B with lowest unit comprehensive energy consumption of the purification device and an optimization algorithm model D with lowest energy consumption of the public engineering according to the steps 1, 2, 4 and 5, firstly, the load of each purification series is optimized and distributed according to the fuel gas to be processed of the whole plant, the raw material gas parameters (raw material gas flow and the like) of the corresponding purification series are obtained according to the load, the raw material gas parameters are correspondingly introduced into the optimization algorithm model B of the purification series, the optimal purification device operation parameters, energy carrying working medium demand, energy source demand and material demand of each purification series are obtained, then the sum of the energy carrying working medium demands of each purification series is introduced into the optimization algorithm model D of the public engineering of the plant, the optimal equipment combination and operation parameters of the public engineering are obtained, then, the purification device of the corresponding purification series is controlled according to the optimal purification device operation parameters, and controlling the corresponding equipment of the public works to generate enough energy-carrying working medium according to the optimal equipment combination and the operation parameters of the public works and distributing the required quantity of each purification series, and simultaneously controlling the public works to supply required quantity of energy and materials and distributing the required quantity of the energy and the materials according to the required quantity of each purification series.
The public works of natural gas purification plants need to provide necessary energy and substances for the purification process, such as low-pressure steam, purified circulating water, necessary nitrogen and the like, are main units for energy-carrying working medium consumption, and are important and difficult points for energy conservation and consumption reduction. The method is characterized in that the operation characteristics of key equipment in the public engineering, such as a boiler, a steam-driven turbine, a temperature reduction pressure reducer and the like, are obtained on site according to historical operation data, an optimized operation strategy of the public engineering is established according to dynamic requirements of a series of purification units or combined devices on energy consumption and material consumption, the start and stop of the equipment are scheduled, and the cooperative optimization and accurate allocation with the purification process are achieved. The difficulty of optimizing operation is that a plurality of devices are operated, and the operation performance and the starting of the devices are simultaneously included in an optimization strategy, which is difficult to realize by a classical mathematical analysis method.
To sum up, the deep digging energy-saving potential of the natural gas purification plant needs to solve two problems, namely the scheduling optimization problem of the series of purification units and the accurate allocation problem of the public works. The invention provides a natural gas purification process control method, which determines key influence parameters according to the working principle of a purification process, and establishes a prediction and optimization algorithm model for the consumption of energy-carrying working media such as steam generated and consumed in a purification device boundary region under different treatment capacities of natural gas based on field historical data; meanwhile, according to historical data, a public engineering key equipment energy-carrying working medium consumption prediction model and an equipment operation combination optimization algorithm model are established; the energy consumption and the material consumption of the purification device and the public engineering are considered, the integrated optimization of the energy consumption of the natural gas purification device and the public engineering is realized, and the minimum energy consumption control of the natural gas purification process is realized.
The scale of the natural gas purification plant is only one purification series or a plurality of purification devices connected in parallel, and the method of the invention can be used for analyzing and mining the key operation parameters of the purification equipment and the operation parameters of the public works under different raw material natural gas flow rates according to historical operation data to obtain how to adjust the key operation parameters of the purification equipment so as to ensure that the total energy consumption of the purification series and the public works is the lowest.
Claims (9)
1. A natural gas purification process control method is characterized by comprising the following steps:
1) acquiring current raw material natural gas parameters, and obtaining natural gas purification device operation parameters and energy-carrying working medium demand with lowest process comprehensive energy consumption under the raw material natural gas parameter condition through optimization of an intelligent algorithm B;
2) controlling a public project according to the required quantity of the energy-carrying working medium, and controlling a natural gas purification device according to the operation parameters of the natural gas purification device;
the process of searching the optimal by the intelligent algorithm B in the step 1) comprises the following steps: firstly, initializing a population, wherein the population comprises operation parameters of a natural gas purification device; then calculating population fitness, wherein the fitness is process comprehensive energy consumption which is calculated through a prediction model A; finally, changing the population, iterating for multiple times, and finally selecting the natural gas purification device operation parameters corresponding to the population with the lowest comprehensive energy consumption in the process and the corresponding energy-carrying working medium demand as an optimization result;
The prediction model A is a machine learning model and is obtained by training historical data of raw material natural gas parameters, natural gas purification device operation parameters, energy-carrying working medium demand, energy consumption and material consumption;
the process comprehensive energy consumption is the sum of the demand of energy-carrying working media, energy consumption and material consumption in the natural gas purification process.
2. The natural gas purification process control method according to claim 1, wherein in step 2), the method for controlling the utility according to the energy-carrying working medium demand amount is to obtain an optimized utility equipment operation parameter with the lowest energy consumption of the utility equipment when meeting the energy-carrying working medium demand amount by optimizing through an intelligent algorithm D, and control the corresponding equipment of the utility according to the optimized utility equipment operation parameter;
the intelligent algorithm D optimizing process comprises the following steps: firstly, initializing a population, wherein the population comprises public engineering equipment operation parameters; then calculating population fitness, wherein the fitness is the energy consumption of public engineering equipment, and the energy consumption of the public engineering equipment is calculated through a prediction model C; finally, changing the population and iterating for multiple times to finally select the public engineering equipment operation parameter corresponding to the population with the lowest public engineering equipment energy consumption as the optimized public engineering equipment operation parameter;
The prediction model C is a machine learning model and is obtained by training historical data of the energy-carrying working medium supply quantity and the energy consumption of the public engineering equipment corresponding to the public engineering equipment operation parameters.
3. The natural gas purification process control method according to claim 2, wherein the prediction model C is a neural network model.
4. The natural gas purification process control method according to claim 3, wherein the intelligent algorithm D is a particle swarm algorithm.
5. The natural gas purification process control method according to claim 1 or 4, wherein the natural gas purification device operating parameter is a parameter that affects the production of energy-consuming working medium by the purification device.
6. The natural gas purification process control method according to claim 5, wherein the sum of the demand of the energy-carrying working medium, the energy consumption and the material consumption is obtained by weighting and summing various consumed energy sources and substances according to respective energy conversion coefficients.
7. The natural gas purification process control method according to claim 1, wherein the prediction model a is a neural network model.
8. The natural gas purification process control method according to claim 7, wherein the intelligent algorithm B is a genetic algorithm.
9. The natural gas purification process control method according to claim 1 or 2, wherein when a plurality of purification series are operated in parallel under a non-full load, the purification series with lower unit comprehensive energy consumption under the same raw material natural gas parameter undertakes more purification tasks; the unit comprehensive energy consumption is the ratio of the process comprehensive energy consumption to the corresponding raw material natural gas treatment capacity.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011182496.0A CN114429236A (en) | 2020-10-29 | 2020-10-29 | Natural gas purification process control method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011182496.0A CN114429236A (en) | 2020-10-29 | 2020-10-29 | Natural gas purification process control method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114429236A true CN114429236A (en) | 2022-05-03 |
Family
ID=81310152
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011182496.0A Pending CN114429236A (en) | 2020-10-29 | 2020-10-29 | Natural gas purification process control method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114429236A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115019907A (en) * | 2022-06-28 | 2022-09-06 | 重庆大学 | Digital twin system of natural gas triethylene glycol dewatering device |
US11893518B2 (en) * | 2022-10-20 | 2024-02-06 | Chengdu Qinchuan Iot Technology Co., Ltd. | Methods and systems of optimizing pressure regulation at intelligent gas gate stations based on internet of things |
-
2020
- 2020-10-29 CN CN202011182496.0A patent/CN114429236A/en active Pending
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115019907A (en) * | 2022-06-28 | 2022-09-06 | 重庆大学 | Digital twin system of natural gas triethylene glycol dewatering device |
CN115019907B (en) * | 2022-06-28 | 2024-03-01 | 重庆大学 | Digital twin system of natural gas triethylene glycol dehydration device |
US11893518B2 (en) * | 2022-10-20 | 2024-02-06 | Chengdu Qinchuan Iot Technology Co., Ltd. | Methods and systems of optimizing pressure regulation at intelligent gas gate stations based on internet of things |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114429236A (en) | Natural gas purification process control method | |
Zhao et al. | The green behavioral effect of clean coal technology on China's power generation industry | |
CN111598308B (en) | Method for solving combination optimization of slurry circulating pump based on regression and double PSO algorithm | |
CN101286044A (en) | Coal-burning boiler system mixing modeling method | |
CN114427684A (en) | Control method and control device for combustion furnace in natural gas purification process | |
CN114357782A (en) | Comprehensive energy system optimization scheduling method considering carbon source sink effect | |
Xie et al. | Dynamic multi-objective intelligent optimal control toward wastewater treatment processes | |
CN111552175A (en) | Overall optimization scheduling and rapid variable load control method for supercritical coal-fired power plant-carbon capture system after chemical adsorption combustion | |
US20060121616A1 (en) | Method and system for increasing efficiency of FGD operation in fossil fuel boilers | |
CN114676878A (en) | Multi-region virtual power plant optimal scheduling method oriented to multi-energy complementation and low carbonization | |
Shao et al. | Predictive scheduling of wet flue gas desulfurization system based on reinforcement learning | |
CN115860205A (en) | Two-stage distribution robust hydrogen storage equipment optimal configuration method considering cross-season scheduling | |
CN112783115A (en) | Online real-time optimization method and device for steam power system | |
Ma et al. | Double recovery strategy of carbon for coal-to-power based on a multi-energy system with tradable green certificates | |
CN118117668A (en) | Cold region comprehensive energy optimal scheduling method based on wind energy and solar energy | |
CN112686447A (en) | Multi-energy flow coupling load prediction method for offshore oil and gas field development | |
CN114429258A (en) | Natural gas purification control method based on energy efficiency evaluation and online optimization platform | |
CN114741966A (en) | Grading optimization control method for organic pollutant production and treatment synergy of reactive distillation | |
CN113191083B (en) | Optimization design method of flue gas waste heat recovery system considering all-working-condition external parameter change | |
Qian et al. | Artificial intelligence for smart energy systems in process industries | |
CN117847537B (en) | Combustion and smoke emission integrated control method and system for waste incineration power plant | |
Yan et al. | Model Prediction and Optimal Control of Gas Oxygen Content for A Municipal Solid Waste Incineration Process | |
Zhang et al. | Application of neural network in modeling of activated sludge wastewater treatment process | |
Liu et al. | Model‐free adaptive control based on switching mechanism for multi‐input multi‐output wet flue gas desulfurization system | |
Borujeni et al. | Fuel cell voltage control using neural network based on model predictive control |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |