CN118242564B - Intelligent self-adaptive natural gas odorizing agent concentration control system and control method - Google Patents
Intelligent self-adaptive natural gas odorizing agent concentration control system and control method Download PDFInfo
- Publication number
- CN118242564B CN118242564B CN202410666434.9A CN202410666434A CN118242564B CN 118242564 B CN118242564 B CN 118242564B CN 202410666434 A CN202410666434 A CN 202410666434A CN 118242564 B CN118242564 B CN 118242564B
- Authority
- CN
- China
- Prior art keywords
- odorizing agent
- real
- data
- time
- natural gas
- 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.)
- Active
Links
- 239000003795 chemical substances by application Substances 0.000 title claims abstract description 414
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 title claims abstract description 304
- 239000003345 natural gas Substances 0.000 title claims abstract description 152
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000002347 injection Methods 0.000 claims abstract description 211
- 239000007924 injection Substances 0.000 claims abstract description 211
- 238000001514 detection method Methods 0.000 claims abstract description 140
- 239000007789 gas Substances 0.000 claims abstract description 93
- 238000012545 processing Methods 0.000 claims abstract description 27
- 238000003062 neural network model Methods 0.000 claims abstract description 23
- 238000013135 deep learning Methods 0.000 claims abstract description 22
- 230000007613 environmental effect Effects 0.000 claims description 63
- 238000001179 sorption measurement Methods 0.000 claims description 49
- 239000003205 fragrance Substances 0.000 claims description 42
- 238000009826 distribution Methods 0.000 claims description 31
- 238000004422 calculation algorithm Methods 0.000 claims description 30
- 238000012806 monitoring device Methods 0.000 claims description 27
- 238000003860 storage Methods 0.000 claims description 27
- 230000006870 function Effects 0.000 claims description 26
- 239000000126 substance Substances 0.000 claims description 26
- 210000002569 neuron Anatomy 0.000 claims description 24
- 238000004891 communication Methods 0.000 claims description 20
- 230000003044 adaptive effect Effects 0.000 claims description 15
- 238000013527 convolutional neural network Methods 0.000 claims description 13
- 230000015654 memory Effects 0.000 claims description 13
- 238000004458 analytical method Methods 0.000 claims description 11
- 239000012855 volatile organic compound Substances 0.000 claims description 9
- 238000005033 Fourier transform infrared spectroscopy Methods 0.000 claims description 8
- 230000004913 activation Effects 0.000 claims description 8
- 239000000463 material Substances 0.000 claims description 6
- 239000007787 solid Substances 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 3
- 230000001502 supplementing effect Effects 0.000 claims description 2
- 230000008569 process Effects 0.000 abstract description 26
- 238000005457 optimization Methods 0.000 abstract description 7
- 238000012544 monitoring process Methods 0.000 abstract description 6
- 238000004590 computer program Methods 0.000 description 15
- 230000000694 effects Effects 0.000 description 14
- 238000010586 diagram Methods 0.000 description 12
- 238000012549 training Methods 0.000 description 12
- 238000009827 uniform distribution Methods 0.000 description 11
- 210000004027 cell Anatomy 0.000 description 9
- 239000002699 waste material Substances 0.000 description 7
- 238000004817 gas chromatography Methods 0.000 description 5
- 230000007246 mechanism Effects 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 5
- 238000004611 spectroscopical analysis Methods 0.000 description 5
- 238000001228 spectrum Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 238000012795 verification Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 239000002184 metal Substances 0.000 description 4
- 238000007781 pre-processing Methods 0.000 description 4
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 3
- 238000010521 absorption reaction Methods 0.000 description 3
- 238000004140 cleaning Methods 0.000 description 3
- 238000013136 deep learning model Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 239000004065 semiconductor Substances 0.000 description 3
- 238000000926 separation method Methods 0.000 description 3
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000002790 cross-validation Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000009792 diffusion process Methods 0.000 description 2
- 239000012530 fluid Substances 0.000 description 2
- 239000007943 implant Substances 0.000 description 2
- 230000031700 light absorption Effects 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 239000012071 phase Substances 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000007637 random forest analysis Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 230000008786 sensory perception of smell Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000010183 spectrum analysis Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000012731 temporal analysis Methods 0.000 description 2
- 238000000700 time series analysis Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 241000282412 Homo Species 0.000 description 1
- 238000004566 IR spectroscopy Methods 0.000 description 1
- 238000000862 absorption spectrum Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000012159 carrier gas Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000003795 desorption Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000003487 electrochemical reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 239000012634 fragment Substances 0.000 description 1
- 239000007792 gaseous phase Substances 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 229910021389 graphene Inorganic materials 0.000 description 1
- 239000001307 helium Substances 0.000 description 1
- 229910052734 helium Inorganic materials 0.000 description 1
- SWQJXJOGLNCZEY-UHFFFAOYSA-N helium atom Chemical compound [He] SWQJXJOGLNCZEY-UHFFFAOYSA-N 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000004949 mass spectrometry Methods 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 230000009965 odorless effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Abstract
The invention provides an intelligent self-adaptive natural gas odorizing agent concentration control system and a control method, which belong to the field of natural gas odorizing agent concentration monitoring. The controller comprises a data processing module, an odorizing agent injection strategy determining module and a control module. The data processing module receives and processes real-time data comprising real-time gas flow detection data, real-time odorizing agent concentration detection data, real-time environment parameters and real-time odorizing agent characteristic parameters; the odorizing agent injection strategy determining module is used for outputting an optimal odorizing agent injection strategy based on the deep learning neural network model by combining real-time data; the control module controls a plurality of parameters of the odorizing agent injection pump according to an optimal odorizing agent injection strategy. The invention realizes the accurate control and optimization of the odorizing agent injection process through intelligent and self-adaptive control.
Description
Technical Field
The invention relates to the field of natural gas odorizing agent concentration monitoring and control, in particular to an intelligent self-adaptive natural gas odorizing agent concentration control system and method.
Background
Natural gas is a colorless and odorless gas, and in order to ensure that it can be found in time when leaked, odorizing agents are usually added to natural gas to make it have a distinct smell. However, conventional odorant injection systems often rely on a fixed injection amount or simple control algorithm, and it is difficult to adapt to changes in natural gas flow and environmental conditions, resulting in unstable odorant concentrations, affecting the safety and use of natural gas.
In addition, the conventional system lacks of intelligence and self-adaptation ability, and cannot dynamically adjust the injection strategy of the odorizing agent according to the real-time data, resulting in waste of the odorizing agent and increase of running cost. In addition, the conventional system has difficulty in ensuring uniform distribution and concentration stability of the odorizing agent in the face of complex environmental conditions and variable natural gas flow, and further affects the use safety and user experience of the natural gas.
Therefore, there is a need for an intelligent adaptive natural gas odorant concentration control system to achieve precise control and optimization of the odorant injection process.
Disclosure of Invention
In view of the above, the embodiments of the present invention provide an intelligent self-adaptive natural gas odorizing agent concentration control system and control method, so as to solve the above technical problems.
To achieve the above object, in a first aspect, there is provided an intelligent adaptive natural gas odorizing agent concentration control system, comprising: a gas flow sensor, an odorizing agent injection pump, an odorizing agent concentration monitoring device, an environmental parameter sensor, an odorizing agent characteristic sensor and a controller;
the gas flow sensor is used for acquiring real-time gas flow detection data in the natural gas pipeline;
The odorizing agent concentration monitoring device is used for acquiring real-time odorizing agent concentration detection data in the natural gas;
The environment parameter sensor is arranged around the natural gas pipeline and is electrically connected with the controller and used for acquiring real-time environment parameters including environment temperature, environment humidity and atmospheric pressure;
the odorizing agent characteristic sensor is used for detecting real-time odorizing agent characteristic parameters in the natural gas in real time;
the odorizing agent injection pump is connected with the odorizing agent storage tank and is used for injecting the odorizing agent in the odorizing agent storage tank into the natural gas pipeline under the control of the controller;
the controller includes:
The data processing module is used for receiving and processing real-time data comprising the real-time gas flow detection data, the real-time odorizing agent concentration detection data, the real-time environment parameters and the real-time odorizing agent characteristic parameters;
The odorizing agent injection strategy determining module is used for outputting an optimal odorizing agent injection strategy based on the deep learning neural network model by combining the real-time data;
And the control module is used for controlling a plurality of parameters of the odorizing agent injection pump according to the optimal odorizing agent injection strategy, wherein the parameters comprise injection rate, injection duration, injection frequency and injection quantity distribution.
In some possible embodiments, the odorizing agent concentration monitoring device includes:
The odorizing agent analysis module is used for detecting the concentrations of different types of odorizing agents in the natural gas respectively and obtaining real-time odorizing agent concentration detection data;
The temperature compensation module is used for carrying out temperature compensation on the real-time odorizing agent concentration detection data according to the environmental temperature data obtained in real time to obtain odorizing agent concentration detection data after temperature compensation;
the environment humidity compensation module is used for performing humidity compensation on the temperature-compensated odorizing agent concentration detection data according to the environment humidity data acquired in real time to obtain the humidity-compensated odorizing agent concentration detection data;
and the atmospheric pressure compensation module is used for carrying out pressure compensation on the odorizing agent concentration detection data after the humidity compensation according to the atmospheric pressure data acquired in real time to obtain the odorizing agent concentration detection data after the pressure compensation.
In some possible implementations, the temperature compensation module employs the following temperature compensation algorithm:
Ctemp_comp=craw+ ktemp × (T-Tref); wherein ctemp_comp is odorizing agent concentration detection data after temperature compensation, craw is real-time odorizing agent concentration detection data of preliminary detection, ktemp is temperature compensation coefficient, T is environmental temperature data acquired in real time at present, tref is reference environmental temperature data;
The environmental humidity compensation module adopts the following temperature compensation algorithm:
chum_comp=ctemp_comp+k hum1×H+khum2×H 2; wherein Chum_comp is the odorant concentration detection data after humidity compensation, ctemp_comp is the odorant concentration detection data after temperature compensation, k hum1 and k hum2 are humidity compensation coefficients, and H is the environmental humidity data acquired in real time currently;
the atmospheric pressure compensation module adopts the following pressure compensation algorithm:
Cpress _comp=chum_comp+ kpress × (P-Pref); wherein Cpress _comp is the odorant concentration detection data after pressure compensation, chum_comp is the odorant concentration detection data after humidity compensation, kpress is the pressure compensation coefficient, P is the current atmospheric pressure data acquired in real time, and Pref is the reference atmospheric pressure data.
In some possible embodiments, the odorizing agent characteristic sensor includes:
A density sensor for detecting the density of the odorizing agent;
An odor intensity sensor for detecting the intensity of odor emitted by the odorizing agent;
A volatility sensor for detecting the volatility characteristics of the odorizing agent in the natural gas;
the surface adsorption analyzer is used for measuring the adsorption characteristics of the odorizing agent on the surfaces of different materials;
And the conductivity sensor is used for detecting the conductivity of the odorizing agent.
In some possible embodiments, the density sensor is installed on a natural gas pipeline, and is electrically connected with the controller through a cable or a wireless communication manner, so as to transmit detected density data to the controller;
The odor intensity sensor is arranged on the natural gas pipeline and is electrically connected with the controller through a cable or a wireless communication mode, and detected odor intensity data are transmitted to the controller;
The volatile sensor is arranged on the natural gas pipeline and is electrically connected with the controller through a cable or a wireless communication mode, and detected volatile data are transmitted to the controller;
The surface adsorption analyzer is arranged on a natural gas pipeline and is electrically connected with the controller through a cable or a wireless communication mode, and detected adsorption data are transmitted to the controller;
the conductivity sensor is arranged on the natural gas pipeline and is electrically connected with the controller through a cable or a wireless communication mode, and detected conductivity data are transmitted to the controller.
In some possible embodiments, the volatile sensor comprises a gas chromatograph, a mass spectrometer, or a fourier transform infrared spectrometer for detecting volatile organic compounds of the odorizing agent;
The odor intensity sensor comprises an electronic nose composed of a plurality of chemical sensors, each chemical sensor being sensitive to a different odor component;
The surface adsorption analyzer evaluates adsorption characteristics of odorizing agent molecules by measuring adsorption amounts and adsorption rates thereof on a solid surface.
In some possible embodiments, the deep learning based neural network model includes:
the input layer is used for receiving standardized historical data, including gas flow detection data, odorizing agent concentration detection data, environmental parameters and odorizing agent characteristic parameters;
A plurality of hidden layers, each hidden layer comprising a plurality of neurons, the neurons adopting a linear rectification function as an activation function for extracting and processing characteristics of input history data;
And the output layer is used for outputting the optimal odorizing agent injection strategy according to the characteristics of the historical data, and the output of the output layer comprises injection rate, injection duration, injection frequency and injection quantity distribution.
In some possible embodiments, when the hidden layers use a multi-layer perceptron structure, each hidden layer includes a plurality of neurons, all of which are fully connected, each neuron receives the outputs of all neurons of the previous layer and performs a nonlinear transformation by an activation function;
when the hidden layers use a convolutional neural network structure, each hidden layer comprises a plurality of convolutional kernels, each convolutional kernel slides on input data, local connection and weight sharing are carried out, and local features are extracted;
When the hidden layers use the LSTM structure of the long-short-period memory network, each hidden layer comprises a plurality of LSTM units, each LSTM unit controls the information to flow through an input gate, a forget gate and an output gate, and long-short-period dependency relations in sequence data are captured.
In some possible embodiments, the controller further comprises: and the prediction alarm module is used for calculating the consumption rate of the odorizing agent in the natural gas pipeline according to the real-time gas flow detection data and the real-time odorizing agent concentration detection data, predicting the residual quantity of the odorizing agent storage tank according to the calculated consumption rate of the odorizing agent, and triggering an odorizing agent supplementing alarm when the residual quantity is lower than a preset threshold value.
In a second aspect, the present invention provides a control method of an intelligent adaptive natural gas odorizing agent concentration control system according to any one of the above, the control method comprising:
s1: acquiring real-time gas flow detection data in a natural gas pipeline;
s2: acquiring real-time odorizing agent concentration detection data in the natural gas;
S3: acquiring real-time environmental parameters including ambient temperature, ambient humidity and atmospheric pressure;
s4: acquiring and detecting real-time odorizing agent characteristic parameters in the natural gas;
s5: receiving and processing real-time data including the real-time gas flow detection data, the real-time odorizing agent concentration detection data, the real-time environmental parameters and the real-time odorizing agent characteristic parameters;
s6: based on the neural network model of deep learning, combining the real-time data, and outputting an optimal odorizing agent injection strategy;
s7: and controlling a plurality of parameters of the odorizing agent injection pump according to the optimal odorizing agent injection strategy, wherein the parameters comprise injection rate, injection duration, injection frequency and injection quantity distribution, and injecting the odorizing agent into the natural gas pipeline. .
The technical scheme has the following beneficial technical effects:
Through the collaborative work of a gas flow sensor, an odorizing agent concentration monitoring device, an environment parameter sensor and an odorizing agent characteristic sensor, the system can acquire high-precision detection data in real time, and the accuracy and the reliability of the data are ensured.
Based on the neural network model of deep learning, the system can combine real-time data and historical data to output an optimal odorizing agent injection strategy, so as to realize intelligent control. The strategy is capable of dynamically adjusting injection rate, injection duration, injection frequency and injection quantity profile to accommodate different operating conditions and environmental changes.
The system can dynamically adjust the odorant injection strategy according to the real-time data and the prediction result, and has stronger self-adaptive capacity. The system maintains the stability and uniformity of the odorizing agent concentration, both at high and low flow rates.
By optimizing the odorant injection strategy, the system can effectively reduce the waste of odorants, improve the energy utilization efficiency and reduce the operation cost.
Drawings
The drawings are included to provide a better understanding of the invention and are not to be construed as unduly limiting the invention. Wherein:
FIG. 1 is a logical block diagram of an intelligent adaptive natural gas odorizing agent concentration control system in accordance with an embodiment of the present invention;
FIG. 2 is a logical block diagram of an odorizing agent concentration monitoring device according to an embodiment of the present invention;
FIG. 3 is a logical block diagram of an odorizing agent characteristic sensor, according to an embodiment of the present invention;
FIG. 4 is a logical block diagram of a predictive alert module in accordance with an embodiment of the invention;
FIG. 5 is a flow chart of an intelligent adaptive natural gas odorizing agent concentration control method in accordance with an embodiment of the present invention;
fig. 6 is a logical block diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention will now be described with reference to the accompanying drawings, in which various details of the embodiments of the present invention are included to facilitate understanding, and are to be considered merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
Example 1
As shown in fig. 1, the embodiment provides an intelligent self-adaptive natural gas odorizing agent concentration control system, which comprises a gas flow sensor, an odorizing agent injection pump, an odorizing agent concentration monitoring device, an environmental parameter sensor, an odorizing agent characteristic sensor and a controller.
And the gas flow sensor is arranged on the natural gas pipeline and is used for acquiring real-time gas flow detection data in the natural gas pipeline. For example, using an ultrasonic flowmeter, high-precision flow detection can be achieved.
The odorizing agent concentration monitoring device is arranged on the natural gas pipeline and used for acquiring real-time odorizing agent concentration detection data in the natural gas. For example, using an electrochemical sensor, the respective concentrations of different kinds of odorizing agents can be detected.
And the environment parameter sensor is arranged around the natural gas pipeline and is electrically connected with the controller and used for acquiring real-time environment parameters including the environment temperature, the environment humidity and the atmospheric pressure. For example, using an integrated temperature, humidity and pressure sensor module, multiple environmental parameters can be acquired simultaneously.
And the odorizing agent characteristic sensor is used for detecting the real-time odorizing agent characteristic parameters in the natural gas in real time. For example, the characteristics of the odorizing agent can be detected comprehensively using a density sensor, a refractive index sensor, and a conductivity sensor.
And the density sensor is used for detecting the density of the odorizing agent. The density is a relation index of the quality and the volume of the odorizing agent, and influences the accuracy of the injection quantity. By detecting the density in real time, the injection strategy can be adjusted according to the density change, so that the accurate injection of the odorizing agent is ensured.
And the refractive index sensor is used for detecting the refractive index of the odorizing agent. Refractive index is an index of the purity and quality of the odorizing agent, affecting its effectiveness. The purity of the odorizing agent can be evaluated by detecting the refractive index in real time, so that the quality consistency of the odorizing agent is ensured.
And the conductivity sensor is used for detecting the conductivity of the odorizing agent. Conductivity is an indicator of the concentration and chemical nature of the odorizing agent ion, affecting its chemical stability. By detecting the conductivity in real time, the ion concentration of the odorizing agent can be estimated, and the chemical property of the odorizing agent is ensured to be stable.
And the odorizing agent injection pump is connected with the odorizing agent storage tank and used for injecting the odorizing agent in the odorizing agent storage tank into the natural gas pipeline under the control of the controller. For example, the injection amount of the odorizing agent can be precisely controlled using a precise metering pump.
The controller comprises a data processing module, an odorizing agent injection strategy determining module and a control module. The data processing module is used for receiving and processing real-time data including real-time gas flow detection data, real-time odorizing agent concentration detection data, real-time environment parameters and real-time odorizing agent characteristic parameters. The odorizing agent injection strategy determining module is used for outputting an optimal odorizing agent injection strategy based on the deep learning neural network model and combining real-time data serving as input. The control module is used for controlling a plurality of parameters of the odorizing agent injection pump according to the optimal odorizing agent injection strategy, wherein the plurality of parameters comprise injection rate, injection duration, injection frequency and injection quantity distribution. The injection amount distribution refers to a specific manner and mode in which the odorizing agent injection pump injects odorizing agent into the natural gas pipe over a certain period of time. It includes the amount of each implant and the uniformity of the implant. In particular, uniformity refers to whether the distribution of odorizing agent in the natural gas pipe is uniform. E.g. whether evenly distributed over the length of the pipe or concentrated in certain specific areas.
In some embodiments, determining the uniform distribution of odorizing agent in the natural gas pipeline may be by several methods:
multipoint sampling analysis: a plurality of odorizing agent concentration monitoring devices are arranged at different positions of the natural gas pipeline, and the odorizing agent concentration at each position is detected in real time. The uniformity of the distribution of the odorizing agent in the pipe is evaluated by comparing the odorizing agent concentration data at different locations.
Gas chromatography analysis: gas samples were collected at different locations in the natural gas pipeline and analyzed for odorizing agent concentration using a gas chromatograph. The uniformity of distribution of the odorizing agent in the pipe was evaluated by the analysis result.
Spectral analysis: the concentration of odorizing agent at different locations in the natural gas pipeline is detected using spectroscopic analysis techniques (e.g., fourier transform infrared spectrometer FTIR). The uniformity of the distribution of the odorizing agent in the pipe was evaluated by the spectroscopic data.
Hydrodynamic simulation: computational fluid dynamics (Computational Fluid Dynamics, CFD) was used to simulate the flow of natural gas and odorizing agent in the pipeline. The uniformity of the distribution of the odorizing agent in the pipe was evaluated by the simulation result.
The uniformly distributed odorizing agent can ensure that leakage at any position can be timely detected when natural gas leaks, and the safety of the natural gas is improved. If the distribution of the odorizing agent is uneven, the concentration of the odorizing agent in certain areas may be too low, so that leakage cannot be detected in time, and potential safety hazards are increased. The distribution uniformity of the odorizing agent is monitored in real time, so that the injection strategy of the odorizing agent can be dynamically adjusted, and the uniform distribution of the odorizing agent in the pipeline is ensured. For example, when it is detected that the odorizing agent concentration in some areas is low, the amount of odorizing agent injected in that area may be increased, ensuring uniform distribution of the whole.
The intelligent self-adaptive natural gas odorizing agent concentration control system reflects the intelligent self-adaptive characteristic through real-time data acquisition and processing, intelligent decision-making based on deep learning, dynamic adjustment of injection strategy, prediction and early warning functions, self-learning and optimization. The system can dynamically adjust the injection strategy of the odorizing agent according to the real-time data and the historical data, ensure the uniform distribution of the odorizing agent in the natural gas pipeline, and improve the safety and the use effect of the natural gas.
The working method of the intelligent self-adaptive natural gas odorizing agent concentration control system comprises the following steps:
S11: and (5) data acquisition. The gas flow sensor acquires gas flow detection data in the natural gas pipeline in real time and transmits the data to the controller. The odorizing agent concentration monitoring device acquires odorizing agent concentration detection data in the natural gas in real time, and transmits the data to the controller. The environmental parameter sensor acquires environmental parameters such as environmental temperature, environmental humidity, atmospheric pressure and the like in real time, and transmits data to the controller. The odorizing agent characteristic sensor detects characteristic parameters of the odorizing agent in real time, and transmits data to the controller.
S12: and (5) data processing. The data processing module receives and processes the real-time data, performs data cleaning and preprocessing, and ensures the accuracy and consistency of the data.
S13: and (5) calculating a strategy. The odorizing agent injection strategy determining module is used for outputting an optimal odorizing agent injection strategy based on the deep learning neural network model and combined with the processed real-time data. For example, the neural network model may be trained and optimized based on historical data and real-time data to improve the accuracy and robustness of the strategy. The output layer of the neural network model outputs an optimal odorant injection strategy including injection rate, injection duration, injection frequency and injection quantity distribution.
S14: control is performed. The control module controls a plurality of parameters of the odorizing agent injection pump according to the output optimal odorizing agent injection strategy, including injection rate, injection duration, injection frequency and injection quantity distribution. For example, at high flow rates, the control module may increase the injection rate and injection frequency to ensure stability of the odorizing agent concentration. The injection rate, the injection duration, the injection frequency and the injection quantity distribution are adjusted in real time, so that the stability and the uniformity of the concentration of the odorizing agent are ensured, and the safety of the natural gas is improved. By dynamically adjusting injection parameters, excessive or insufficient injection of the odorizing agent is avoided, waste of the odorizing agent is reduced, and resource utilization efficiency is improved. Through real-time data and a deep learning model, the system can adapt to different flow and environmental conditions, flexibly adjust injection strategies and ensure the stability of the concentration of the odorizing agent. By optimizing the injection strategy, the waste of odorizing agent and the running cost are reduced, and the economy of the system is improved.
The following example is given assuming that in a certain natural gas pipeline system, the gas flow sensor detects a current natural gas flow of 500 cubic meters per hour, the odorant concentration monitoring device detects a current odorant concentration of 10 ppm, the environmental parameter sensor detects an environmental temperature of 25 degrees celsius, an environmental humidity of 60%, an atmospheric pressure of 1013 hPa, the odorant property sensor detects a density of 0.8 g/cm, an odor intensity of 50 ppm, a volatility property of 20 kPa, an adsorption amount on a metal surface of 0.5 mg/m, and an electrical conductivity of 0.1S/m. The data processing module of the controller receives and processes the data, and the odorizing agent injection strategy determining module calculates the optimal odorizing agent injection strategy based on the deep learning neural network model and combining the real-time data. The control module controls the injection rate of the odorizing agent injection pump to be 0.5 liter/hour, the injection duration to be 10 minutes, the injection frequency to be injected once per hour, and the injection quantity to be uniformly distributed according to the strategy.
In some embodiments, the intelligent self-adaptive natural gas odorizing agent concentration control system is provided with a self-learning module, and the odorizing agent injection strategy is continuously optimized and updated according to historical data and real-time data, so that the self-adaptation capability and accuracy of the system are improved. Through self-learning and optimization, the system can adapt to different running conditions and environmental changes, and intelligent and self-adaptive odorizing agent injection control is realized.
The training process of the neural network model based on deep learning comprises the following steps:
S21: data collection and preprocessing.
First, the system needs to collect various data generated during operation, including gas flow rate detection data, odorizing agent concentration detection data, environmental parameters (environmental temperature, environmental humidity, atmospheric pressure), and odorizing agent characteristic parameters (density, odor intensity, volatility characteristics, adsorption characteristics, conductivity). The data are cleaned to remove outliers and noise and fill in missing values. Next, data of different dimensions is normalized to the same range (e.g., all data is normalized to the 0-1 range) to enhance the training effect of the model. Finally, the data set is divided into a training set, a verification set and a test set, and the proportion can be 70% of the training set, 15% of the verification set and 15% of the test set.
S22: and (5) constructing a model.
The structure of the neural network model includes an input layer, a hidden layer, and an output layer. The input layer receives the preprocessed input data, including gas flow, odorizing agent concentration, environmental parameters, and odorizing agent characteristic parameters. The hidden Layer includes a plurality of neuron layers for extracting and processing characteristics of input data, and may use a Multi-Layer Perceptron (MLP), convolutional neural network (Convolutional Neural Network, CNN), or Long Short-Term Memory (LSTM) structure. The output layer outputs the optimal odorizing agent injection strategy, including injection rate, injection duration, injection frequency and injection quantity distribution. In one example, the hidden layer uses 3 layers, each layer containing 64 neurons, with an activation function of ReLU.
In the neural network model, the hidden layer may be configured by a multi-layer perceptron (MLP), a Convolutional Neural Network (CNN), or a long-short-term memory network (LSTM). Each hidden layer is made up of a plurality of neurons having different functions and connections in different network architectures. The following is the logical relationship between these architectures and neurons:
in a multilayer perceptron (MLP), each hidden layer is made up of a plurality of neurons, all of which are fully connected. Each neuron receives the output of all neurons of the previous layer and performs a nonlinear transformation by an activation function (e.g., reLU) and outputs to all neurons of the next layer. Assuming a hidden layer with 64 neurons, each neuron receives the output of all neurons of the previous layer and processes it through the ReLU activation function.
In Convolutional Neural Networks (CNNs), each hidden layer consists of a plurality of convolutional kernels (filters), each of which corresponds to a neuron. The convolution kernel slides on the input data, performs local connection and weight sharing, and extracts local features. The convolutional layer is typically followed by a pooling layer (Pooling Layer) to reduce the data dimension. Assuming a convolutional layer with 32 convolutional kernels, each sliding on the input data, local features are extracted and processed by a ReLU activation function.
In long-term memory networks (LSTM), each hidden layer is made up of a plurality of LSTM cells, each LSTM cell corresponding to a neuron. The LSTM unit can capture long-term and short-term dependency in the sequence data through the flow of input gate, forget gate and output gate control information. Assuming that one LSTM layer has 128 LSTM cells, each cell receives the output of the previous layer and its own state, and processes it through a gating mechanism.
S23: and (5) model training.
During model training, an appropriate loss function (e.g., mean square error (Mean Square Error, MSE)) is selected to measure the difference between the model predicted and real values, and an optimization algorithm (e.g., adam (Adaptive Moment Estimation, adaptive moment estimation), SGD (Stochastic GRADIENT DESCENT, random gradient descent)) is used to minimize the loss function, adjusting the model parameters. The model is trained through multiple iterations (epoch), each iteration using a batch of data for parameter updates. Super parameters such as learning rate, batch size, number of hidden layers and number of neurons also need to be adjusted during training to find the best model structure. The step length of updating the learning rate control parameter can influence the convergence effect of the model when the learning rate is too large or too small. The batch size is the amount of data used per iteration, and affects training speed and model performance.
In one example, model training is performed using the following parameters: the loss function is Mean Square Error (MSE), the optimization algorithm is Adam, the learning rate is 0.001, the batch size is 32, and the training iteration is 5000 times. During the training process, the model learns the features of the data through multiple iterations (epochs) and continuously adjusts the parameters to minimize the loss function. Compared with the model in the prior art, the application of the neural network model based on deep learning in the intelligent self-adaptive natural gas odorizing agent concentration control system has the following advantages: the Mean Square Error (MSE) is used as a loss function, so that the difference between the predicted value and the true value of the model can be effectively measured, and the high-precision prediction of the model is ensured; by adopting an Adam optimization algorithm, the learning rate can be adaptively adjusted, and the convergence speed and stability of the model are improved; the model learns the characteristics of data through multiple iterations (epochs), continuously adjusts parameters to minimize a loss function, can dynamically adapt to real-time data and environmental changes, and provides an optimal odorizing agent injection strategy; the mini batch gradient descent method with the batch size of 32 is used, so that the generalization capability of the model can be improved while the training speed is ensured, and overfitting is avoided; the deep learning model (such as a multi-layer perceptron MLP, a convolutional neural network CNN and a long-short-term memory network LSTM) has strong feature extraction and representation capability, can extract useful features from complex multidimensional data, and improves the prediction and control capability of the model; the deep learning model has high flexibility and expansibility, and can adjust the model structure and parameters according to specific application requirements, so as to adapt to different running conditions and environmental changes.
S24: model verification and testing.
In the model verification process, verification set data are used for evaluating the performance of the model, and the super parameters are adjusted to improve the generalization capability of the model. The stability and reliability of the model was further verified by cross-validation (cross-validation) methods. In the model test process, the test set data is used to evaluate the performance of the final model, ensuring that the model has good predictive power on unseen data.
S25: model deployment and application.
And deploying the trained model into a controller, receiving and processing input data in real time, and outputting an optimal odorizing agent injection strategy. The controller dynamically adjusts a plurality of parameters of the odorizing agent injection pump according to the injection strategy output by the model, and intelligent and self-adaptive odorizing agent injection control is realized. In one example, for example, the injection strategy for the model output is: the injection rate is 0.5 liter/hour, the injection duration is 10 minutes, the injection frequency is once per hour, and the injection quantity is uniformly distributed.
In some embodiments, dynamic adjustment of the injection rate may be achieved. The control module dynamically adjusts the injection rate of the odorizing agent according to the real-time gas flow, the odorizing agent concentration, the ambient temperature and the ambient humidity. Increasing the injection rate at high flow rates and high temperatures; at low flow rates and low temperatures, the injection rate is reduced. For example, as the gas flow rate increases to 600 cubic meters per hour, the injection rate is automatically adjusted to 0.6 liters per hour.
In some embodiments, dynamic adjustment of the injection duration may be achieved. The control module dynamically adjusts the duration of each injection based on fluctuations in natural gas flow, changes in odorizing agent concentration, and ambient humidity. When the flow fluctuation is large and the humidity is high, the injection time is prolonged to ensure the concentration to be stable; and when the flow is stable and the humidity is low, the injection time is shortened. For example, when the flow fluctuation is large, the injection period is automatically adjusted to 15 minutes.
In some embodiments, dynamic adjustment of the injection frequency may be achieved. The control module dynamically adjusts the interval time of the injection operation according to the natural gas flow, the change frequency of the odorizing agent concentration and the ambient pressure. Increasing the injection frequency under conditions of frequent flow rate changes and low pressure; the injection frequency is reduced with a steady flow and a higher pressure. For example, when the flow rate is changed frequently, the injection frequency is automatically adjusted to once every 30 minutes.
In some embodiments, dynamic adjustment of the injection volume distribution may be achieved. The control module dynamically adjusts the distribution mode of the odorizing agent according to the pipeline length, the natural gas flow characteristic and the odorizing agent characteristic parameter. Under the conditions of long-distance pipelines and high-volatility odorizing agents, a multi-point injection mode is adopted to ensure that the odorizing agents are uniformly distributed; in the case of short-distance pipelines and low-volatility odorizing agents, a single-point injection mode is adopted. For example, in long distance pipelines, the injection volume distribution is automatically adjusted to a multipoint uniform distribution.
For example, in a natural gas pipeline system, the gas flow sensor detects that the current natural gas flow is 500 cubic meters per hour, and the odorizing agent concentration monitoring device detects that the current odorizing agent concentration is 10 ppm. The ambient parameter sensor detects an ambient temperature of 25 degrees celsius, an ambient humidity of 60%, and an atmospheric pressure of 1013 hPa. The odorizing agent characteristic sensor detects that the density of the odorizing agent is 0.8 g/cm, the odor intensity is 50 ppm, the volatility characteristic is 20 kPa, the adsorption amount on the metal surface is 0.5 mg/m, and the electrical conductivity is 0.1S/m. The system acquires the data in real time, and performs data cleaning and preprocessing. The odorizing agent injection strategy determining module calculates the optimal odorizing agent injection strategy based on the deep learning neural network model and by combining real-time data. The control module dynamically adjusts a plurality of parameters of the odorizing agent injection pump according to the output optimal odorizing agent injection strategy. For example, the current optimal odorant injection strategy is: the injection rate is 0.5 liter/hour, the injection duration is 10 minutes, the injection frequency is once per hour, and the injection quantity is uniformly distributed. The dynamic adjustment is as follows: when the gas flow rate was increased to 600 cubic meters per hour, the control module automatically adjusted the injection rate to 0.6 liters per hour. When the flow fluctuation is large and the ambient humidity is high, the control module automatically adjusts the injection duration to 15 minutes. When the flow rate is frequently changed and the atmospheric pressure is low, the control module automatically adjusts the injection frequency to once every 30 minutes. Under the conditions of long-distance pipelines and high volatility of the odorizing agent, the control module automatically adjusts the injection quantity distribution to be multi-point uniform distribution.
The intelligent self-adaptive natural gas odorizing agent concentration control system of the embodiment realizes the accurate control of the odorizing agent injection process by integrating various sensors and a control algorithm based on deep learning, and has the following technical effects:
Through the collaborative work of a gas flow sensor, an odorizing agent concentration monitoring device, an environment parameter sensor and an odorizing agent characteristic sensor, the system can acquire high-precision detection data in real time, and the accuracy and the reliability of the data are ensured.
Based on the neural network model of deep learning, the system can combine real-time data and historical data to output an optimal odorizing agent injection strategy, so as to realize intelligent control. The strategy is capable of dynamically adjusting injection rate, injection duration, injection frequency and injection quantity profile to accommodate different operating conditions and environmental changes.
The system can dynamically adjust the odorant injection strategy according to the real-time data and the prediction result, and has stronger self-adaptive capacity. The system maintains the stability and uniformity of the odorizing agent concentration, both at high and low flow rates.
By optimizing the odorant injection strategy, the system can effectively reduce the waste of odorants, improve the energy utilization efficiency and reduce the operation cost.
Example two
As shown in fig. 2, the embodiment provides an odorizing agent concentration monitoring device for an intelligent self-adaptive natural gas odorizing agent concentration control system, which comprises an odorizing agent analysis module, a temperature compensation module, an environmental humidity compensation module and an atmospheric pressure compensation module.
The odorizing agent analysis module is used for detecting the concentrations of different types of odorizing agents in the natural gas, and real-time odorizing agent concentration detection data are obtained. In some embodiments, a spectroscopic analysis module, a gas chromatography module, and a semiconductor gas sensor may be employed. The gas chromatographic module is used for detecting the concentrations of different types of odorizing agents in the natural gas, and real-time odorizing agent concentration detection data are obtained. A gas chromatography module is a technology based on gas separation and detection that is capable of detecting multiple gas components simultaneously. The gas chromatography module separates the components in the mixed gas by a separation column, and then detects the concentration of the components by a detector. The spectrum analysis module is used for detecting the concentrations of different types of odorizing agents in the natural gas respectively and obtaining real-time odorizing agent concentration detection data. Spectroscopic analysis techniques determine the concentration of gas molecules by detecting their light absorption or emission characteristics. In particular, the spectroscopic analysis module includes a light source, a sample cell, a spectrometer, and a detector. The light source emits light in a specific wavelength range, ultraviolet light, visible light or infrared light can be used, and the selection of the light source depends on the light absorption or emission characteristics of the gas molecules to be detected. The gas to be measured passes through the sample cell, and the light emitted by the light source passes through the gas molecules in the sample cell. The design of the sample cell ensures that light is able to uniformly pass through the gas sample. The spectrometer is used to separate and analyze the light after passing through the sample cell, break the light down into spectra of different wavelengths, and measure the light intensity at each wavelength. The gas molecules will absorb or emit light in a specific wavelength range, resulting in characteristic absorption peaks or emission peaks in the spectrum. The detector is used for measuring the light intensity output by the spectrometer and converting the light intensity into an electric signal. The output signal of the detector is proportional to the concentration of the gas molecules. The processor receives the output signal of the detector and performs data processing by the spectrum analysis software, and the concentration of the odorizing agent gas molecules is determined by comparing the intensities and positions of the characteristic absorption peaks or emission peaks in the spectrum.
The temperature compensation module is used for carrying out temperature compensation on the real-time odorizing agent concentration detection data according to the environmental temperature data obtained in real time, and obtaining odorizing agent concentration detection data after temperature compensation.
The environment humidity compensation module is used for performing humidity compensation on the temperature-compensated odorizing agent concentration detection data according to the environment humidity data obtained in real time, and obtaining the humidity-compensated odorizing agent concentration detection data.
The atmospheric pressure compensation module is used for performing pressure compensation on the odorizing agent concentration detection data after the humidity compensation according to the atmospheric pressure data obtained in real time, so as to obtain the odorizing agent concentration detection data after the pressure compensation.
The working process of the odorizing agent concentration monitoring device comprises the following steps:
S31: and (5) data acquisition.
The odorizing agent analyzing module detects the odorizing agent concentration in the natural gas in real time, and preliminary odorizing agent concentration detection data are obtained. The temperature sensor acquires environmental temperature data in real time, the humidity sensor acquires environmental humidity data in real time, and the pressure sensor acquires atmospheric pressure data in real time.
S32: and (5) temperature compensation.
The temperature compensation module performs temperature compensation on the primarily detected odorizing agent concentration detection data based on a temperature compensation algorithm according to the environmental temperature data acquired in real time, and obtains odorizing agent concentration detection data after temperature compensation. The temperature compensation algorithm may use a linear regression model, assuming that the effect of ambient temperature on the odorant concentration is linear:
Ctemp_comp=Craw+ktemp×(T－Tref)。
Wherein ctemp_comp is the temperature compensated odorizing agent concentration detection data, craw is the preliminary detected real-time odorizing agent concentration detection data, ktemp is the temperature compensation coefficient, T is the current real-time acquired environmental temperature data, tref is the reference environmental temperature data.
S33: and (5) humidity compensation.
And the environmental humidity compensation module performs humidity compensation on the temperature-compensated odorizing agent concentration detection data according to the environmental humidity data acquired in real time, so as to obtain the humidity-compensated odorizing agent concentration detection data. The humidity compensation algorithm may use a polynomial regression model assuming that the effect of humidity on concentration is a nonlinear relationship:
Chum_comp=Ctemp_comp +khum1×H+khum2×H 2
Wherein Chum_comp is the odorant concentration detection data after humidity compensation, ctemp_comp is the odorant concentration detection data after temperature compensation, k hum1 and k hum2 are humidity compensation coefficients, and H is the environmental humidity data currently acquired in real time.
S33: and (5) pressure compensation.
The atmospheric pressure compensation module performs pressure compensation on the odorizing agent concentration detection data after humidity compensation according to the atmospheric pressure data acquired in real time, and obtains the odorizing agent concentration detection data after pressure compensation. The pressure compensation algorithm may use a linear regression model, assuming that the effect of pressure on concentration is linear:
Cpress_comp=Chum_comp + kpress×(P－Pref)。
wherein Cpress _comp is the odorant concentration detection data after pressure compensation, chum_comp is the odorant concentration detection data after humidity compensation, kpress is the pressure compensation coefficient, P is the current atmospheric pressure data acquired in real time, and Pref is the reference atmospheric pressure data.
The odorizing agent concentration monitoring device of the embodiment realizes high-precision detection and multiple compensation of the odorizing agent concentration in the natural gas by integrating the odorizing agent analysis module, the temperature compensation module, the environmental humidity compensation module and the atmospheric pressure compensation module, and has the following specific technical effects:
The odorizing agent analysis module adopts an electrochemical sensor, and can detect the concentration of different types of odorizing agents in the natural gas with high sensitivity and high selectivity. The electrochemical sensor can accurately detect the odorizing agent component with low concentration by an electrochemical reaction principle, and ensures the reliability and accuracy of a detection result.
The temperature compensation module performs temperature compensation on the odorizing agent concentration detection data through the environmental temperature data acquired in real time. The temperature compensation algorithm can eliminate the influence of the ambient temperature on the detection result, and ensure the consistency and accuracy of the detection result under different temperature conditions. For example, in the case of a large change in ambient temperature, the temperature compensation module can dynamically adjust the detection data to provide an accurate concentration value after temperature compensation.
The environment humidity compensation module carries out humidity compensation on the temperature compensated odorizing agent concentration detection data through the environment humidity data obtained in real time. The humidity compensation algorithm can eliminate the influence of the ambient humidity on the detection result, and ensure the stability and reliability of the detection result under different humidity conditions. For example, in a high humidity environment, the humidity compensation module can effectively adjust the detection data to provide an accurate concentration value after humidity compensation.
The atmospheric pressure compensation module performs pressure compensation on the odorizing agent concentration detection data after humidity compensation through the atmospheric pressure data acquired in real time. The pressure compensation algorithm can eliminate the influence of atmospheric pressure on the detection result, and ensure the accuracy and consistency of the detection result under different pressure conditions. For example, in high altitude or low altitude environments, the pressure compensation module can dynamically adjust the sensed data to provide an accurate concentration value after pressure compensation.
The odorizing agent concentration monitoring device realizes a multiple compensation mechanism by integrating a temperature compensation module, a humidity compensation module and a pressure compensation module. The multiple compensation mechanism can comprehensively consider the influence of the ambient temperature, the humidity and the pressure on the detection result, and provides high-precision odorizing agent concentration detection data after multiple compensation. The multiple compensation mechanism ensures the stability and reliability of the detection result, and is suitable for detecting the concentration of the odorizing agent under various complex environmental conditions.
The odorizing agent concentration monitoring device can monitor the odorizing agent concentration in the natural gas in real time and dynamically adjust according to environmental parameters acquired in real time. The real-time monitoring and dynamic adjusting functions ensure the real-time performance and accuracy of detection data, can timely reflect the change of environmental conditions, and provide accurate detection results of the concentration of the odorizing agent.
Through the high-precision odorizing agent concentration detection and multiple compensation mechanism, the odorizing agent concentration monitoring device can provide accurate detection data for an intelligent self-adaptive natural gas odorizing agent concentration control system. The accurate detection data can improve the overall performance of the system, ensure the accuracy and the effectiveness of the odorant injection strategy, optimize the use of the odorant, and improve the intelligence and the self-adaptive capacity of the system.
Example III
As shown in fig. 3, the present embodiment provides an odorant characteristic sensor for an intelligent adaptive natural gas odorant concentration control system, including any of a density sensor, an odor intensity sensor, a volatility sensor, a surface adsorption analyzer, and a conductivity sensor.
The density sensor is arranged on the natural gas pipeline and used for detecting the density of the odorizing agent in the gas state in real time. The density sensor is electrically connected with the controller through a cable and transmits detected density data to the controller. The gas density sensor works by measuring the mass and volume of gas to calculate its density.
The odor intensity sensor is arranged on the natural gas pipeline and is used for detecting the odor intensity emitted by the odorizing agent in real time. The odor intensity sensor is electrically connected with the controller through a cable and transmits detected odor intensity data to the controller. The odor intensity sensor works by detecting chemical reactions of gas molecules with the sensor surface to generate an electrical signal proportional to the odor intensity.
The volatility sensor is arranged on the natural gas pipeline and used for detecting the volatility characteristic of the odorizing agent in the natural gas in real time. The volatile sensor is electrically connected with the controller through a cable or a wireless communication mode, and transmits the detected volatile data to the controller. The operation of the volatile sensor is to evaluate the volatility of a gas molecule by measuring its vapor pressure.
The surface adsorption analyzer is arranged on the natural gas pipeline and is used for measuring the adsorption characteristics of the odorizing agent on the surfaces of different materials. The surface adsorption analyzer is electrically connected with the controller through a cable or a wireless communication mode, and transmits detected adsorption data to the controller. The surface adsorption analyzer works by measuring the adsorption of gas molecules on the surfaces of different materials to evaluate the adsorption characteristics.
The conductivity sensor is arranged on the natural gas pipeline and is used for detecting the conductivity of the odorizing agent in a gas state in real time. The conductivity sensor is electrically connected with the controller through a cable or a wireless communication mode, and transmits detected conductivity data to the controller. Gas conductivity sensors operate by measuring the conductivity of a gas to evaluate its ion concentration and chemical properties.
All sensors (density sensor, odor intensity sensor, volatility sensor, surface adsorption analyzer, and conductivity sensor) are mounted on the natural gas pipeline adjacent to each other forming a sensor array. This ensures that all sensors can detect the characteristic parameters of the odorizing agent at the same time as the natural gas flows through the pipe. All the sensors are electrically connected with the controller through cables or wireless communication modes. The output end of each sensor is connected to the input end of the controller through a cable or a wireless communication mode, and the controller receives and processes detection data from each sensor.
In particular, volatility refers to the ability or tendency of the odorizing agent to evaporate in the gas. The more volatile substances are easily converted from liquid or solid to gaseous and are therefore more easily distributed homogeneously in natural gas. The volatility characteristics are very important for odoriferous agents because it directly affects the diffusion rate and effectiveness of the odoriferous agent and thus the perception of the odor of the natural gas during use.
In the detection of volatility, a certain amount of natural gas sample is first extracted from the natural gas pipeline by a sampling device. The sampling device needs to ensure the representativeness of the sample and prevent the sample from being changed during the collection process.
The volatile organic compounds were then detected using the following analytical equipment:
Gas Chromatograph (GC): a gas chromatograph is a device for detecting volatile organic compounds. The gas sample is injected into a gas chromatograph, and the components in the sample are separated in the chromatographic column by pushing with a carrier gas (such as helium or nitrogen). The different components, depending on their chemical nature, move at different speeds in the chromatographic column and are eventually detected at the detector. The detector converts these signals to a chromatogram showing the relative concentrations and volatilities of the components.
Mass Spectrometer (MS): the use of a gas chromatograph in combination with a mass spectrometer (GC-MS) may further improve detection accuracy. After separation in a gas chromatographic column, the sample enters a mass spectrometer which recognizes the chemical structure and concentration of each component by ionization and detecting the mass-to-charge ratio of the molecular fragments. This method can provide very accurate qualitative and quantitative information of volatile organic compounds.
Fourier transform infrared spectrometer (FTIR): FTIR can analyze the composition and concentration of volatile organic compounds by detecting their absorption characteristics in the infrared region of the spectrum. A gas sample is introduced into FTIR and as infrared light passes through the sample, various components absorb infrared light of specific wavelengths, thereby producing a characteristic absorption spectrum. By analyzing these spectra, the kind and content of the volatile organic compounds can be determined.
Electronic nose: electronic nose is a sensor system that simulates the human sense of smell, consisting of multiple chemical sensors, each sensitive to a different volatile organic compound. When sample gas passes through the sensor array, response signals generated by the sensors are subjected to data processing and pattern recognition to obtain characteristic fingerprints of volatile components in the sample. This method allows for rapid detection and differentiation of different volatile organic compounds.
The detection of volatility is critical to ensure uniform distribution of natural gas odorizing agent throughout the pipe system. The high-volatility odorizing agent can be rapidly diffused, so that the odorizing agent can be sensed by smell even under low concentration, and the safety of natural gas is improved. In addition, the detection of the volatility characteristics of the odorizing agent can also help optimize the formulation and the use strategy of the odorizing agent, and improve the use effect and the economy of the odorizing agent.
In particular, the intensity of the odor refers to the extent to which a substance is olfactory perceived in the gas phase. The intensity of the odor is related to the volatility, but not only to the volatility of the substance, but also to its chemical nature and the sensitivity of the human olfactory system. The odor intensity detection device is as follows: an odor sensor (electronic nose) consisting of multiple chemical sensors, each sensitive to a different odor component, simulating the human olfactory system. Volatility concerns the ability and speed of a substance to enter a gaseous state, primarily affecting the distribution of the substance in the gaseous phase. Odor intensity concerns the influence of substances in the gas phase on the smell, mainly affecting the intensity of the odor perceived by humans. The detection of volatility mainly depends on gas chromatography, mass spectrometry, infrared spectrometry, etc. devices that analyze chemical compositions and concentrations of substances. The detection of scent intensity is more dependent on scent sensors (e-noses) focusing on the influence of substances on the sense of smell.
In particular, the adsorption characteristics refer to the adsorption behavior of the odorizing agent on the inner wall of the natural gas pipe or vessel. This behavior describes how the odorizing agent molecules, when contacting a solid surface (e.g., a pipe wall, a tank inner wall, etc.), are captured and retained by the surface matter. The adsorption characteristics have an important impact on the effectiveness and concentration control of the odorizing agent. Excessive adsorption can cause the odorizing agent to be captured by the solid surface during transport, reducing its effective concentration in the natural gas, thereby affecting the function of the odorizing agent. At the same time, the desorption characteristics facilitate the selection of suitable tubing materials and coatings to reduce the loss of odorizing agent. The surface adsorption analyzer evaluates the adsorption characteristics of odorizing agent molecules by measuring the adsorption amount and adsorption rate of the molecules on the solid surface.
The working process of the odorizing agent characteristic sensor comprises the following steps:
S41, density detection: natural gas flows through a density sensor that measures the mass and volume of the gas, calculates its density, and transmits density data to a controller.
S42, odor intensity detection: natural gas flows through an odor intensity sensor that generates an electrical signal proportional to odor intensity by detecting chemical reactions of gas molecules with the sensor surface and transmits odor intensity data to a controller.
S43, detecting volatility: natural gas flows through a volatility sensor that evaluates its volatility by measuring the vapor pressure of the gas molecules and transmits the volatility data to a controller.
S44, surface adsorption detection: natural gas flows through a surface adsorption analyzer, and a sensor evaluates adsorption characteristics of gas molecules by measuring adsorption amounts of the gas molecules on surfaces of different materials and transmits adsorption data to a controller.
S45, conductivity detection: natural gas flows through a conductivity sensor that measures the conductivity of the gas, evaluates its ion concentration and chemical properties, and transmits conductivity data to a controller.
It is assumed that in a certain natural gas pipeline system, the following are detection data of each sensor: the density sensor detects that the current density of the odorizing agent in the natural gas is 0.8 g/cm. The odor intensity sensor detects that the current odorizing agent has an odor intensity of 50 ppm. The volatile sensor detects a current odorizing agent vapor pressure of 20 kPa. The surface adsorption analyzer detects that the adsorption amount of the odorizing agent on the metal surface is 0.5 mg/m < DEG >. The conductivity sensor detects that the current odorizing agent has a conductivity of 0.1S/m. The controller receives and processes the data and calculates the optimal odorizing agent injection strategy by combining the data of other sensors.
In the intelligent self-adaptive natural gas odorizing agent concentration control system, the odorizing agent characteristic sensor is added and various characteristic parameters (density, odor intensity, volatility, surface adsorption and conductivity) of the odorizing agent are detected in real time, so that the accuracy and the intelligent level of the system can be remarkably improved.
By detecting the characteristic parameters of the odorizing agent, such as density, odor intensity, volatility, surface adsorption, conductivity and the like, in real time, the system can comprehensively understand the physical and chemical characteristics of the odorizing agent, so that a more accurate injection strategy is formulated. For example, density detection may help determine the mass and volume relationship of the odorizing agent, ensuring the accuracy of the injection amount; the odor intensity detection can ensure that the odor effect of the odorizing agent reaches the expected value; the volatility detection can evaluate the diffusion capability of the odorizing agent in the natural gas, so as to ensure the uniform distribution of the odorizing agent; the adsorption condition of the odorizing agent on the inner wall of the pipeline can be evaluated by the surface adsorption detection, so that the loss of the odorizing agent is avoided; conductivity detection allows the ionic concentration and chemical nature of the odorant to be assessed, ensuring its chemical stability.
By monitoring various characteristic parameters of the odorizing agent in real time, the system can dynamically adjust the injection strategy according to real-time data, so as to ensure the optimal effect of the odorizing agent under different environmental conditions. For example, in a high-temperature and high-humidity environment, a highly volatile odorizing agent needs to increase the injection frequency and the injection amount; under the low-temperature and low-humidity environment, the odorizing agent with higher density needs to prolong the injection duration and increase the injection rate; in the case of severe adsorption on the inner wall of the pipe, it is necessary to adjust the distribution of the injection amount to ensure uniform distribution of the odorizing agent in the pipe.
By comprehensively analyzing various characteristic parameters of the odorizing agent, the system can adaptively adjust the injection strategy, adapt to different running conditions and environmental changes, and improve the self-adaptive capacity of the system. For example, the system can predict the consumption rate and the residual amount of the odorizing agent according to the real-time data and the historical data, and adjust the injection strategy in advance to avoid the insufficient odorizing agent or the excessive odorizing agent; the system can optimize the injection strategy according to the characteristic parameters of the odorizing agent, and ensure the uniform distribution of the odorizing agent under different pipeline lengths and flow characteristics.
By accurately controlling the injection amount and distribution of the odorizing agent, the system can ensure the odor effect and safety of the natural gas and improve the use experience of users. For example, the uniformly distributed odorizing agent can ensure that the natural gas can be timely detected when the natural gas leaks, so that the safety of the natural gas is improved; the precisely controlled injection amount of the odorizing agent can ensure that the natural gas has consistent odor perception when in use, and the use effect of a user is improved.
By accurately controlling the injection amount and distribution of the odorizing agent, the system can reduce the waste of the odorizing agent, improve the resource utilization efficiency and reduce the operation cost. For example, by monitoring and dynamically adjusting the injection strategy in real time, the system can avoid excessive or insufficient injection of odorizing agent, reducing odorizing agent waste; by optimizing the injection strategy, the system can improve the use efficiency of the odorizing agent and reduce the running cost.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, the specific names of the functional units and modules are only for distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.
Example IV
As shown in fig. 4, the present embodiment provides a predictive alert module for an intelligent adaptive natural gas odorant concentration control system. The prediction alarm module is used for calculating the consumption rate of the odorizing agent in the natural gas pipeline according to the real-time gas flow detection data and the real-time odorizing agent concentration detection data, and predicting the residual quantity of the odorizing agent storage tank according to the calculated consumption rate of the odorizing agent. And triggering an odorizing agent replenishment alarm when the residual amount is lower than a preset threshold value.
The predictive alarm module is arranged in the controller and is electrically connected with the gas flow sensor and the odorizing agent concentration monitoring device through cables or wireless communication modes. The predictive alert module receives and processes real-time data from the gas flow sensor and the odorant concentration monitoring device.
The working process of the prediction alarm module comprises the following steps:
S51, data acquisition: the prediction alarm module receives the gas flow data detected by the gas flow sensor and the odorizing agent concentration data detected by the odorizing agent concentration monitoring device in real time.
S52, calculating the consumption rate: the prediction alarm module calculates the consumption rate of the odorizing agent in the natural gas pipeline according to the real-time gas flow detection data and the real-time odorizing agent concentration detection data. The consumption rate was calculated as follows:
Consumption rate = gas flow x odorizing agent concentration.
S53, predicting the residual quantity: the predictive alert module employs advanced predictive algorithms, such as a time series analysis (ARIMA (Autoregressive Integrated Moving Average, autoregressive integral moving average) model), machine learning algorithms (e.g., random forest, support vector machine), or deep learning algorithms (e.g., LSTM) to predict the remaining amount of the odorizing agent tank. The predictive algorithm not only considers the current consumption rate, but also performs a comprehensive analysis in combination with historical data and environmental parameters.
In the working process of the prediction alarm module, the prediction algorithm not only considers the current consumption rate, but also carries out comprehensive analysis by combining historical data and environmental parameters so as to improve the accuracy and reliability of prediction. Specifically, the current consumption rate is calculated from the real-time gas flow rate detection data and the odorizing agent concentration detection data. A predictive algorithm such as time series analysis (ARIMA model), a machine learning algorithm (e.g., random forest, support vector machine) or a deep learning algorithm (e.g., LSTM) would take the current consumption rate as one of the input variables. Meanwhile, the prediction algorithm also uses historical data, including past parameters such as gas flow, odorizing agent concentration, ambient temperature, ambient humidity, atmospheric pressure and the like, to capture time dependence and trend in the data. In addition, environmental parameters such as temperature, humidity, and pressure may also be input variables, helping the algorithm to better understand and predict the consumption pattern of the odorizing agent. By comprehensively analyzing the current consumption rate, the historical data and the environmental parameters, the prediction algorithm can more accurately predict the residual quantity of the odorizing agent storage tank, trigger an alarm in time and ensure the normal operation of the system and the timely replenishment of the odorizing agent.
S54, alarm triggering: when the predicted remaining amount of the odorizing agent storage tank is below a preset threshold, the prediction alarm module triggers an odorizing agent replenishment alarm, informing an operator of odorizing agent replenishment.
Assume that in a certain natural gas pipeline system, the gas flow sensor detects that the current natural gas flow is 500 cubic meters per hour, and the odorizing agent concentration monitoring device detects that the current odorizing agent concentration is 10 ppm. The initial reserve of the odorizing agent tank is 1000 liters, and the preset threshold value is 100 liters.
In the data acquisition step, the predictive alert module receives gas flow data (500 cubic meters per hour) and odorant concentration data (10 ppm) in real time.
In the consumption rate calculation step, the predictive warning module calculates a consumption rate of the odorizing agent:
consumption rate = 500 cubic meters per hour x 10ppm = 5 liters per hour.
In the residual amount prediction step, the prediction alarm module adopts an LSTM model to predict the residual amount. The LSTM model predicts future odorizing agent consumption by learning historical data and current data. Assuming a run time of 100 hours, the LSTM model predicts a consumption rate of 80 hours into the future as follows:
future consumption rate= [5.1,5.2,5.0,4.9,5.3, … ] l/hr.
Based on the predicted consumption rate, the total consumption for the future 80 hours is calculated:
。
assuming a total consumption of 400 liters, the residual amount is predicted as follows:
the remaining amount=1000 liters-400 liters=600 liters.
In the alarm triggering step, the predictive alarm module triggers an odorizing agent replenishment alarm when the predicted odorizing agent tank remaining amount is below a preset threshold value (100 liters). Assuming that 180 hours of run time passed, the LSTM model predicts a total consumption of 100 liters for 20 hours in the future, the residuals are calculated as follows:
Residual amount=600 liters-100 liters=500 liters.
At this time, the predictive warning module triggers an odorant replenishment warning, informing the operator of the odorant replenishment.
The prediction alarm module accurately predicts the residual quantity of the odorizing agent storage tank through an advanced prediction algorithm and combines real-time monitoring and historical data, and timely triggers an alarm when needed, so that the normal operation of the system and the timely replenishment of odorizing agent are ensured.
Example five
As shown in fig. 5, the embodiment provides an intelligent self-adaptive natural gas odorizing agent concentration control method, which includes the following steps:
s1: real-time gas flow detection data in the natural gas pipeline is acquired.
And acquiring gas flow detection data in the natural gas pipeline in real time through a gas flow sensor arranged on the natural gas pipeline. For example, the gas flow sensor detects a current natural gas flow of 500 cubic meters per hour.
S2: and acquiring real-time odorizing agent concentration detection data in the natural gas.
The odorizing agent concentration monitoring device is arranged on the natural gas pipeline to acquire odorizing agent concentration detection data in the natural gas in real time. For example, the odorant concentration monitoring device detects that the current odorant concentration is 10 ppm.
S3: real-time environmental parameters including ambient temperature, ambient humidity, and barometric pressure are acquired.
Environmental parameters such as ambient temperature, ambient humidity, atmospheric pressure and the like are acquired in real time through an environmental parameter sensor arranged around the natural gas pipeline. For example, the ambient parameter sensor detects an ambient temperature of 25 degrees celsius, an ambient humidity of 60%, and an atmospheric pressure of 1013 hPa.
S4: and acquiring the characteristic parameters of the real-time odorizing agent in the natural gas.
And detecting the characteristic parameters of the odorizing agent in the natural gas in real time through the odorizing agent characteristic sensor. For example, the odorant property sensor detects that the density of the odorant is 0.8 g/cm, the odor intensity is 50 ppm, the volatility property is 20 kPa, the adsorption amount on the metal surface is 0.5 mg/m, and the electrical conductivity is 0.1S/m.
S5: and receiving and processing real-time data comprising the real-time gas flow detection data, the real-time odorizing agent concentration detection data, the real-time environment parameters and the real-time odorizing agent characteristic parameters.
The data processing module of the controller receives and processes the real-time data, and performs data cleaning and preprocessing to ensure the accuracy and consistency of the data.
S6: and based on the neural network model of deep learning, combining the real-time data, and outputting an optimal odorizing agent injection strategy.
The odorizing agent injection strategy determining module calculates the optimal odorizing agent injection strategy based on the deep learning neural network model and by combining real-time data. The neural network model extracts and processes the characteristics of input data through structures such as a multi-layer perceptron (MLP), a Convolutional Neural Network (CNN) or a long-short-term memory network (LSTM), and outputs an optimal odorizing agent injection strategy.
S7: and controlling a plurality of parameters of the odorizing agent injection pump according to the optimal odorizing agent injection strategy, wherein the parameters comprise injection rate, injection duration, injection frequency and injection quantity distribution, and injecting the odorizing agent into the natural gas pipeline.
The control module controls a plurality of parameters of the odorizing agent injection pump, including injection rate, injection duration, injection frequency and injection quantity distribution, according to the calculated optimal odorizing agent injection strategy. For example, the control module controls the odorizing agent injection pump to have an injection rate of 0.5 liters/hour, an injection duration of 10 minutes, an injection frequency of one injection per hour, and an injection amount distribution to be uniform.
The embodiment shows the working principle of the intelligent self-adaptive natural gas odorizing agent concentration control method, all the sensors work cooperatively, various characteristic parameters of odorizing agent in a gas state are detected in real time, data are transmitted to the controller, and the controller calculates the optimal odorizing agent injection strategy based on a deep learning neural network model, so that intelligent and self-adaptive odorizing agent injection control is realized.
Example six
The embodiment of the invention also provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements a control method of any one of the intelligent self-adaptive natural gas odorizing agent concentration control systems described above.
The integrated modules/units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the present invention may implement all or part of the flow of the method of the above embodiment, or may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the computer program may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), an electrical carrier signal, a telecommunications signal, a software distribution medium, and so forth. Of course, there are other ways of readable storage medium, such as quantum memory, graphene memory, etc. It should be noted that the computer readable medium contains content that can be appropriately scaled according to the requirements of jurisdictions in which such content is subject to legislation and patent practice, such as in certain jurisdictions in which such content is subject to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.
The invention also provides electronic equipment. The electronic equipment of the embodiment of the invention comprises: one or more processors; and the storage device is used for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors realize the control method of the intelligent self-adaptive natural gas odorizing agent concentration control system.
Referring now to FIG. 6, there is illustrated a schematic diagram of a computer system 600 suitable for use in implementing an electronic device of an embodiment of the present invention. The electronic device shown in fig. 6 is only an example and should not be construed as limiting the functionality and scope of use of the embodiments of the invention.
As shown in fig. 6, the computer system 600 includes a Central Processing Unit (CPU) 601, which can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 602 or a program loaded from a storage section 608 into a Random Access Memory (RAM) 603. In the RAM603, various programs and data required for the operation of the computer system 600 are also stored. The CPU601, ROM 602, and RAM603 are connected to each other through a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.
The following components are connected to the I/O interface 605: an input portion 606 including a keyboard, mouse, etc.; an output portion 607 including a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, a speaker, and the like; a storage section 608 including a hard disk and the like; and a communication section 609 including a network interface card such as a LAN card, a modem, or the like. The communication section 609 performs communication processing via a network such as the internet. The drive 610 is also connected to the I/O interface 605 as needed. A removable medium 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is installed on the drive 610 as necessary, so that a computer program read out therefrom is installed into the storage section 608 as necessary.
In particular, the processes described in the main step diagrams above may be implemented as computer software programs according to the disclosed embodiments of the invention. For example, embodiments of the present invention include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method shown in the main step diagrams. In the above-described embodiment, the computer program can be downloaded and installed from a network through the communication section 609 and/or installed from the removable medium 611. The above-described functions defined in the system of the present invention are performed when the computer program is executed by the central processing unit 601.
The computer readable medium shown in the present invention may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present invention, a computer readable signal medium may comprise a data signal propagated in baseband or as part of a carrier wave, with computer readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units involved in the embodiments of the present invention may be implemented in software or in hardware. The units described may also be provided in a processor, the names of these units in some cases not constituting a limitation of the unit itself.
The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives can occur depending upon design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.
Claims (6)
1. An intelligent adaptive natural gas odorizing agent concentration control system, comprising: a gas flow sensor, an odorizing agent injection pump, an odorizing agent concentration monitoring device, an environmental parameter sensor, an odorizing agent characteristic sensor and a controller;
the gas flow sensor is used for acquiring real-time gas flow detection data in the natural gas pipeline;
The odorizing agent concentration monitoring device is used for acquiring real-time odorizing agent concentration detection data in the natural gas;
The environment parameter sensor is arranged around the natural gas pipeline and is electrically connected with the controller and used for acquiring real-time environment parameters including environment temperature, environment humidity and atmospheric pressure;
the odorizing agent characteristic sensor is used for detecting real-time odorizing agent characteristic parameters in the natural gas in real time;
the odorizing agent injection pump is connected with the odorizing agent storage tank and is used for injecting the odorizing agent in the odorizing agent storage tank into the natural gas pipeline under the control of the controller;
the controller includes:
The data processing module is used for receiving and processing real-time data comprising the real-time gas flow detection data, the real-time odorizing agent concentration detection data, the real-time environment parameters and the real-time odorizing agent characteristic parameters;
The odorizing agent injection strategy determining module is used for outputting an optimal odorizing agent injection strategy based on the deep learning neural network model by combining the real-time data;
the control module is used for controlling a plurality of parameters of the odorizing agent injection pump according to the optimal odorizing agent injection strategy, wherein the parameters comprise injection rate, injection duration, injection frequency and injection quantity distribution;
the odorizing agent concentration monitoring device includes:
The odorizing agent analysis module is used for detecting the concentrations of different types of odorizing agents in the natural gas respectively and obtaining real-time odorizing agent concentration detection data;
The temperature compensation module is used for carrying out temperature compensation on the real-time odorizing agent concentration detection data according to the environmental temperature data obtained in real time to obtain odorizing agent concentration detection data after temperature compensation;
the environment humidity compensation module is used for performing humidity compensation on the temperature-compensated odorizing agent concentration detection data according to the environment humidity data acquired in real time to obtain the humidity-compensated odorizing agent concentration detection data;
the atmospheric pressure compensation module is used for performing pressure compensation on the humidity-compensated odorizing agent concentration detection data according to the atmospheric pressure data acquired in real time to obtain the pressure-compensated odorizing agent concentration detection data;
the odorizing agent characteristic sensor includes:
A density sensor for detecting the density of the odorizing agent;
An odor intensity sensor for detecting the intensity of odor emitted by the odorizing agent;
A volatility sensor for detecting the volatility characteristics of the odorizing agent in the natural gas;
the surface adsorption analyzer is used for measuring the adsorption characteristics of the odorizing agent on the surfaces of different materials;
a conductivity sensor for detecting the conductivity of the odorizing agent;
The neural network model based on deep learning comprises:
the input layer is used for receiving standardized historical data, including gas flow detection data, odorizing agent concentration detection data, environmental parameters and odorizing agent characteristic parameters;
A plurality of hidden layers, each hidden layer comprising a plurality of neurons, the neurons adopting a linear rectification function as an activation function for extracting and processing characteristics of input history data;
The output layer is used for outputting an optimal odorizing agent injection strategy according to the characteristics of the historical data, and the output of the output layer comprises injection rate, injection duration, injection frequency and injection quantity distribution;
When the hidden layers use a multi-layer perceptron structure, each hidden layer comprises a plurality of neurons which are fully connected, each neuron receives the output of all neurons of the previous layer and carries out nonlinear transformation through an activation function;
when the hidden layers use a convolutional neural network structure, each hidden layer comprises a plurality of convolutional kernels, each convolutional kernel slides on input data, local connection and weight sharing are carried out, and local features are extracted;
When the hidden layers use the LSTM structure of the long-short-period memory network, each hidden layer comprises a plurality of LSTM units, each LSTM unit controls the information to flow through an input gate, a forget gate and an output gate, and long-short-period dependency relations in sequence data are captured.
2. The intelligent adaptive natural gas odorizing agent concentration control system of claim 1, characterized in that,
The temperature compensation module adopts the following temperature compensation algorithm:
Ctemp_comp=craw+ ktemp × (T-Tref); wherein ctemp_comp is odorizing agent concentration detection data after temperature compensation, craw is real-time odorizing agent concentration detection data of preliminary detection, ktemp is temperature compensation coefficient, T is environmental temperature data acquired in real time at present, tref is reference environmental temperature data;
The environmental humidity compensation module adopts the following temperature compensation algorithm:
Chum_comp=ctemp_comp+k hum1×H+khum2×H2; wherein Chum_comp is the odorant concentration detection data after humidity compensation, ctemp_comp is the odorant concentration detection data after temperature compensation, k hum1 and k hum2 are humidity compensation coefficients, and H is the environmental humidity data acquired in real time currently;
the atmospheric pressure compensation module adopts the following pressure compensation algorithm:
Cpress _comp=chum_comp+ kpress × (P-Pref); wherein Cpress _comp is the odorant concentration detection data after pressure compensation, chum_comp is the odorant concentration detection data after humidity compensation, kpress is the pressure compensation coefficient, P is the current atmospheric pressure data acquired in real time, and Pref is the reference atmospheric pressure data.
3. The intelligent adaptive natural gas odorizing agent concentration control system of claim 1, characterized in that,
The density sensor is arranged on the natural gas pipeline and is electrically connected with the controller through a cable or a wireless communication mode, and detected density data are transmitted to the controller;
The odor intensity sensor is arranged on the natural gas pipeline and is electrically connected with the controller through a cable or a wireless communication mode, and detected odor intensity data are transmitted to the controller;
The volatile sensor is arranged on the natural gas pipeline and is electrically connected with the controller through a cable or a wireless communication mode, and detected volatile data are transmitted to the controller;
The surface adsorption analyzer is arranged on a natural gas pipeline and is electrically connected with the controller through a cable or a wireless communication mode, and detected adsorption data are transmitted to the controller;
the conductivity sensor is arranged on the natural gas pipeline and is electrically connected with the controller through a cable or a wireless communication mode, and detected conductivity data are transmitted to the controller.
4. The intelligent adaptive natural gas odorizing agent concentration control system of claim 1, characterized in that,
The volatile sensor comprises a gas chromatograph, a mass spectrometer or a Fourier transform infrared spectrometer and is used for detecting volatile organic compounds of the odorizing agent;
The odor intensity sensor comprises an electronic nose composed of a plurality of chemical sensors, each chemical sensor being sensitive to a different odor component;
The surface adsorption analyzer evaluates adsorption characteristics of odorizing agent molecules by measuring adsorption amounts and adsorption rates thereof on a solid surface.
5. The intelligent adaptive natural gas odorizing agent concentration control system of claim 1, wherein said controller further comprises: and the prediction alarm module is used for calculating the consumption rate of the odorizing agent in the natural gas pipeline according to the real-time gas flow detection data and the real-time odorizing agent concentration detection data, predicting the residual quantity of the odorizing agent storage tank according to the calculated consumption rate of the odorizing agent, and triggering an odorizing agent supplementing alarm when the residual quantity is lower than a preset threshold value.
6. A control method of an intelligent adaptive natural gas odorizing agent concentration control system of any one of claims 1-5, characterized in that the control method comprises:
s1: acquiring real-time gas flow detection data in a natural gas pipeline;
s2: acquiring real-time odorizing agent concentration detection data in the natural gas;
S3: acquiring real-time environmental parameters including ambient temperature, ambient humidity and atmospheric pressure;
s4: acquiring and detecting real-time odorizing agent characteristic parameters in the natural gas;
s5: receiving and processing real-time data including the real-time gas flow detection data, the real-time odorizing agent concentration detection data, the real-time environmental parameters and the real-time odorizing agent characteristic parameters;
s6: based on the neural network model of deep learning, combining the real-time data, and outputting an optimal odorizing agent injection strategy;
S7: and controlling a plurality of parameters of the odorizing agent injection pump according to the optimal odorizing agent injection strategy, wherein the parameters comprise injection rate, injection duration, injection frequency and injection quantity distribution, and injecting the odorizing agent into the natural gas pipeline.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410666434.9A CN118242564B (en) | 2024-05-28 | Intelligent self-adaptive natural gas odorizing agent concentration control system and control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410666434.9A CN118242564B (en) | 2024-05-28 | Intelligent self-adaptive natural gas odorizing agent concentration control system and control method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118242564A CN118242564A (en) | 2024-06-25 |
| CN118242564B true CN118242564B (en) | 2024-07-16 |
Family
ID=
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101663380A (en) * | 2007-04-19 | 2010-03-03 | 丰田自动车株式会社 | Odorant adding device and fuel gas supply system |
| CN109519709A (en) * | 2018-11-30 | 2019-03-26 | 湖北三江航天江北机械工程有限公司 | LNG odorizing agent concentration control system and its method |
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101663380A (en) * | 2007-04-19 | 2010-03-03 | 丰田自动车株式会社 | Odorant adding device and fuel gas supply system |
| CN109519709A (en) * | 2018-11-30 | 2019-03-26 | 湖北三江航天江北机械工程有限公司 | LNG odorizing agent concentration control system and its method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shvarov | A suite of programs, OptimA, OptimB, OptimC, and OptimS compatible with the Unitherm database, for deriving the thermodynamic properties of aqueous species from solubility, potentiometry and spectroscopy measurements | |
| KR101727889B1 (en) | Monitoring, detecting and quantifying chemical compounds in a gas sample stream | |
| Liu et al. | Effects of gas–wall interactions on measurements of semivolatile compounds and small polar molecules | |
| Hutchinson et al. | Use of chamber systems to measure trace gas fluxes | |
| CN104364635B (en) | The collision broadening verified using real-time or near real-time in spectroanalysis instrument is compensated | |
| Xiao et al. | Components of mesophyll resistance and their environmental responses: a theoretical modelling analysis | |
| US7956761B2 (en) | Infrared gas detection and spectral analysis method | |
| Antonanzas-Torres et al. | Generation of daily global solar irradiation with support vector machines for regression | |
| Alizadeh et al. | Electronic nose based on the polymer coated SAW sensors array for the warfare agent simulants classification | |
| CN110793932B (en) | CF4Gas concentration detection method, device and equipment and accuracy verification system | |
| CN106650926B (en) | A kind of steady boosting extreme learning machine integrated modelling approach | |
| Laref et al. | A comparison between SVM and PLS for E-nose based gas concentration monitoring | |
| Torrecilla et al. | Neural networks to estimate the water content of imidazolium-based ionic liquids using their refractive indices | |
| CN118242564B (en) | Intelligent self-adaptive natural gas odorizing agent concentration control system and control method | |
| You et al. | Hidden Markov model based data-driven calibration of non-dispersive infrared gas sensor | |
| Wang et al. | An Accurate and interpretable framework for trustworthy process monitoring | |
| CN118242564A (en) | Intelligent self-adaptive natural gas odorizing agent concentration control system and control method | |
| CN112630180B (en) | Ultraviolet/visible light absorption spectrum model for detecting concentration of organophosphorus pesticide in water body | |
| Wang et al. | Estimation of soil organic matter by in situ Vis-NIR spectroscopy using an automatically optimized hybrid model of convolutional neural network and long short-term memory network | |
| CN116384775A (en) | Management method, system and storage medium of pollution site monitoring equipment | |
| Mei et al. | Study of an adaptable calibration model of near-infrared spectra based on KF-PLS | |
| Ye et al. | Water chemical oxygen demand prediction model based on the CNN and ultraviolet-visible spectroscopy | |
| Li et al. | Back-propagation neural network-based modelling for soil heavy metal | |
| Becnel et al. | Tiny time-series transformers: Realtime multi-target sensor inference at the edge | |
| Tan et al. | Estimation of Flat‐topped Gaussian distribution with application in system identification |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |