CN107944070A - A kind of Diffusion Simulation method and system of urban atmospheres harmful influence leakage - Google Patents

A kind of Diffusion Simulation method and system of urban atmospheres harmful influence leakage Download PDF

Info

Publication number
CN107944070A
CN107944070A CN201710934322.7A CN201710934322A CN107944070A CN 107944070 A CN107944070 A CN 107944070A CN 201710934322 A CN201710934322 A CN 201710934322A CN 107944070 A CN107944070 A CN 107944070A
Authority
CN
China
Prior art keywords
mrow
gas
msub
diffusion
training
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.)
Granted
Application number
CN201710934322.7A
Other languages
Chinese (zh)
Other versions
CN107944070B (en
Inventor
潘维
孙亭
李毅
陈思
叶云
沈自然
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
CETC 28 Research Institute
Original Assignee
CETC 28 Research Institute
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by CETC 28 Research Institute filed Critical CETC 28 Research Institute
Priority to CN201710934322.7A priority Critical patent/CN107944070B/en
Publication of CN107944070A publication Critical patent/CN107944070A/en
Application granted granted Critical
Publication of CN107944070B publication Critical patent/CN107944070B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • G06Q50/265—Personal security, identity or safety
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Tourism & Hospitality (AREA)
  • Economics (AREA)
  • Primary Health Care (AREA)
  • Educational Administration (AREA)
  • Computer Security & Cryptography (AREA)
  • Health & Medical Sciences (AREA)
  • Geometry (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Development Economics (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Medical Informatics (AREA)
  • Computer Hardware Design (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The present invention relates to a kind of Diffusion Simulation method and system of urban atmospheres harmful influence leakage.Corresponding gas diffusion model is selected by field conditions, gas type and weather conditions, obtaining the corresponding learner of gas leakage type by training combines, based on gas diffusion model and learner combination producing gas leakage Diffusion Simulation.This method utilizes the advantage of each discrete gas diffusion model of integrated approach integration in machine learning, can predict the diffusion process of real hazard gas.

Description

A kind of Diffusion Simulation method and system of urban atmospheres harmful influence leakage
Technical field
The present invention relates to gas diffusion to model field, and in particular to a kind of Diffusion Simulation side of urban atmospheres harmful influence leakage Method and system.
Background technology
Currently, security incident caused by dangerous gas leakage, affects numerous people as a kind of potential threat all the time The security of the lives and property, surrounding enviroment, equipment etc. can be caused greatly to damage.Therefore, simulation and prediction simulated hazard gas Diffusion process, takes early warning to eliminate means with specific aim, is of great significance for reducing dangerous gas leakage hazard.
In view of the defects existing in the prior art, urban atmospheres harmful influence leakage Diffusion Simulation method proposed by the present invention, profit With the integrated approach in machine learning, the advantage of each discrete gas diffusion model is effectively integrated, for predicting real hazard gas The diffusion of body has good effect.
The content of the invention
The present invention proposes a kind of Diffusion Simulation method and system of urban atmospheres harmful influence leakage, can be in dangerous material gas During body Release and dispersion, suitable Diffusion Simulation model is selected, simulates the true diffusion process of gas.
It is including following it is an object of the present invention to provide a kind of Diffusion Simulation method of urban atmospheres harmful influence leakage Step：
(1) according to wind direction, wind direction, wind-force, wind speed and atmospheric stability, incident region real-time weather situation is determined；
According to storage tank pressure, leakage area, leak time and leakage shape, incident region field conditions are determined；
Establish gas diffusion model library, including Calm dispersion model, Gauss model, BM models, Sutton models, FEM3 moulds Type；
Determine the type of spillage risk gas, dangerous gas type is divided into three kinds：Light gas, neutral gas, severe gas Body；
(2) incident region real-time weather situation, field conditions and gas leakage type are based on, is selected from gas diffusion model library Select corresponding gas diffusion model；
(3) the corresponding learner of gas leakage type is obtained by training to combine；
(4) the gas diffusion model selected based on step 2, and the learner combination that step 3 obtains, generate last let out Leak simulation of gas dispersion.
Wherein, in step 3 by training obtain the corresponding learner of gas leakage type combine it is specific as follows：
(a) harmful influence gas training set, including light gas training set, neutral gas training set, severe gas training set are established, Specific representation is as follows：
Tname={ (x1, y1), (x2, y2) ..., (xN, yN), name is light gas, neutral gas, wherein severe gas, xi Represent some positions chosen from the true diffusion space of gas, yiRepresent the gas concentration of the position, i=1...N；
(b) according to different gas types, corresponding gas diffusion model, the base as harmful influence gas training set are selected This learner；
(c) according to different gas, the weights distribution of training data is initialized
Wherein, w1iIt is to choose (x in harmful influence gas training seti, yi) probability, N is in harmful influence gas training set Data volume；
(d) the corresponding simplified diffusion model number of every kind of gas is M, and each diffusion model that simplifies is a basic studies device Gm, m=1 ..., M.For example for neutral gas, M 3, diffusion model is respectively basic studies device G1、G2、G3；
(e) G is calculatedm(x) the large error rate on training dataset X=x1..., xN, wherein, large error refers to | Gm(xi)-yi| ＞ α, α yi1 percent, WmiCorresponding to (xi, yi) Probability, P (| Gm(xi)-yi| ＞ α) represent | Gm(xi)-yi| the ratio that ＞ α are concentrated in training data；
(f) basic studies device G is calculatedm(x) coefficient
(g) the weights distribution of training data is updated
Wherein, ZmIt is standardizing factor so that Dm+1As a probability distribution,
As m+1 ＞ M, enter step (h)
(h) different gas is directed to, builds the linear combination of basic studies device
Further, in step b, if m is neutral gas, from gas diffusion model library choose Gauss model, Three kinds of Sutton models, BM models models, M is set to 3 in step d；If m is light gas, Calm dispersion model, Gauss are selected Model, M is set to 2 in step d；The gas if m attaches most importance to, selects BM, FEM3 model, and M is set to 2 in step d.
In addition, in step 4, generating last gas leakage Diffusion Simulation is specially：The β obtained based on the training stagem, and With reference to corresponding gas diffusion model, utilizeThe concentration of current gas each point is obtained, so that For being diffused emulation to leakage gas.
It is a further object to provide a kind of Diffusion Simulation system of urban atmospheres harmful influence leakage, system includes With lower module：
Gas diffusion model library, for storing gas diffusion model, including Calm dispersion model, Gauss model, BM models, Sutton models, FEM3 models；
Gas diffusion model library selecting module, for according to incident region real-time weather situation, field conditions and leakage gas Body type, corresponding gas diffusion model is selected from gas diffusion model library, and selection result is sent to gas leakage diffusion Emulation module；
Learner combined training module, for training the combination of gas leakage type corresponding learner, and by training result It is sent to gas leakage Diffusion Simulation module；
Gas diffusion model generation module, for the gas diffusion model selected according to gas diffusion model library selecting module The learner combination trained with learner combined training module, generates last gas leakage Diffusion Simulation.
Wherein, gas diffusion model library selecting module is including real-time weather situation determines, field conditions determine and leaks gas Body type determines three subelements, wherein,
Real-time weather situation determination subelement is used for according to wind direction, wind direction, wind-force, wind speed and atmospheric stability, determines thing Send out region real-time weather situation；
Field conditions determination subelement is used to, according to storage tank pressure, leakage area, leak time and leakage shape, determine thing Send out region field conditions；
Gas leakage type determination unit is used for the type for determining spillage risk gas, and dangerous gas type is divided into Three kinds：Light gas, neutral gas, severe gas.
Further, learner combined training module obtains the corresponding learner combination tool of gas leakage type by training Body is：
(a) harmful influence gas training set, including light gas training set, neutral gas training set, severe gas training set are established, Specific representation is as follows：
Tname={ (x1, y1), (x2, y2) ..., (xN, yN), name is light gas, neutral gas, wherein severe gas, xi Represent some positions chosen from the true diffusion space of gas, yiRepresent the gas concentration of the position, i=1...N；
(b) according to different gas types, corresponding gas diffusion model, the base as harmful influence gas training set are selected This learner；
(c) according to different gas, the weights for initializing training data are distributed D1=(w11..., w1i..., w1N),I=1,2 ..., N, wherein, w1iIt is to choose (x in harmful influence gas training seti, yi) probability, N is harmful influence gas Data volume in body training set；
(d) the corresponding diffusion model number of every kind of gas is M, and each diffusion model is a basic studies device Gm, m= 1 ..., M；
(e) G is calculatedm(x) the large error rate on training dataset X=x1..., xN, wherein, large error refers to | Gm(xi)-yi| ＞ α, α yi1 percent, WmiCorresponding to (xi, yi) Probability, P (| Gm(xi)-yi| ＞ α) represent | Gm(xi)-yi |The ratio that ＞ α are concentrated in training data；
(f) basic studies device G is calculatedm(x) coefficient
(g) the weights distribution of training data is updated
Wherein, ZmIt is standardizing factor so that Dm+1As a probability distribution,
(h) different gas is directed to, builds the linear combination of basic studies device
Urban atmospheres harmful influence proposed by the present invention reveals Diffusion Simulation method and system, utilizes integrating in machine learning Method, effectively integrates the advantage of each discrete gas diffusion model, selects suitable Diffusion Simulation model, simulates the true expansion of gas The process of dissipating, the diffusion for predicting real hazard gas have good effect.
Brief description of the drawings
Fig. 1 shows the urban atmospheres harmful influence leakage Diffusion Simulation method flow schematic diagram of the present invention.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, with reference to the accompanying drawings and embodiments, it is right The present invention is further elaborated, it will be appreciated that specific embodiment described herein is only to explain the present invention, not For limiting the present invention.
It is shown in the drawings now with detailed reference to the embodiment of the present invention, the example of these embodiments.The suffix of element " module " and " unit " is used herein to conveniently describe, and therefore can convertibly be used, and is distinguished without any Meaning or function.
Although all elements or unit that form the embodiment of the present invention illustrated as being coupled in discrete component or are grasped As discrete component or unit, but the present invention may be not necessarily limited to such a embodiment.According to embodiment, in the purpose of the present invention One or more elements can be selectively bonded to element all in scope and are operating as one or more elements.
Urban atmospheres harmful influence leakage Diffusion Simulation method in one embodiment of the present of invention specifically includes：
(1) according to wind direction, wind direction, wind-force, wind speed and atmospheric stability, incident region real-time weather situation is determined；
According to storage tank pressure, leakage area, leak time and leakage shape, incident region field conditions are determined；
Establish gas diffusion model library, including Calm dispersion model, Gauss model, BM models, Sutton models, FEM3 moulds Type etc., table specific as follows：
Such as：Gauss cigarette rolls into a ball model：
Sutton models：
Determine the type of spillage risk gas, dangerous gas type is divided into three kinds：Light gas, neutral gas, severe gas Body；
(2) incident region real-time weather situation, field conditions and gas leakage type are based on, is selected from gas diffusion model library Select corresponding gas diffusion model；
(3) the corresponding learner of gas leakage type is obtained by training to combine；
(4) the gas diffusion model selected based on step 2, and the learner combination that step 3 obtains, generate last let out Leak simulation of gas dispersion.
Wherein, in step 3 by training obtain the corresponding learner of gas leakage type combine it is specific as follows：
(a) harmful influence gas training set, including light gas training set, neutral gas training set, severe gas training set are established, Specific representation is as follows：
Tname={ (x1, y1), (x2, y2) ..., (xN, yN), name is light gas, neutral gas, wherein severe gas, xi Represent some positions chosen from the true diffusion space of gas, yiRepresent the gas concentration of the position, i=1...N；
(b) according to different gas types, corresponding gas diffusion model, the base as harmful influence gas training set are selected This learner；
(c) according to different gas, the weights distribution of training data is initialized
Wherein, w1iIt is to choose (x in harmful influence gas training seti, yi) probability, N is in harmful influence gas training set Data volume, all gas, it is all identical that it initializes the weights distribution of training data；
(d) the corresponding simplified diffusion model number of every kind of gas is M, and each diffusion model that simplifies is a basic studies device Gm, m=1 ..., M.For example for neutral gas, M 3, diffusion model is respectively basic studies device G1、G2、G3, wherein, Basic studies device refers to a kind of gas diffusion model, that is, a kind of gas concentration function；
(e) G is calculatedm(x) the large error rate on training dataset, x=x1..., xN, wherein, it is larger Error refers to | Gm(xi)-yi| ＞ α, α yi1 percent, i.e. model Gm(x) concentration calculated with it is true Gas concentration difference is more than percent a period of time of actual gas concentration, Gm(x) there is " large error " rate.Specifically it is calculated as：Wherein, emRepresent Gm(x) in training data Large error rate on collection, wmiCorresponding to (xi, yi) probability,
P(|Gm(xi)-yi| ＞ α) represent | Gm(xi)-yi| the ratio that ＞ α are concentrated in training data,
I(|Gm(xi)-yi| ＞ α) what is represented is to work as | Gm(xi)-yi| when ＞ α are set up,
I(|Gm(xi)-yi| ＞ α) value be 1, be otherwise 0,Represent large error rate Specific formula for calculation；
(f) basic studies device G is calculatedm(x) coefficient
(g) the weights distribution of training data is updated
Wherein, ZmIt is standardizing factor so that Dm+1As a probability distribution,
As m+1 ＞ M, enter step (h)
(h) different gas is directed to, builds the linear combination of basic studies device
Further, in step b, if m is neutral gas, from gas diffusion model library choose Gauss model, Three kinds of Sutton models, BM models models, M is set to 3 in step d；If m is light gas, Calm dispersion model, Gauss are selected Model, M is set to 2 in step d；The gas if m attaches most importance to, selects BM, FEM3 model, and M is set to 2 in step d.
In addition, in step 4, generating last gas leakage Diffusion Simulation is specially：For different gas, based on training The β that stage obtainsm, and corresponding gas diffusion model is combined, utilizeObtain current gas The concentration of each point, for being diffused emulation to leakage gas.
Urban atmospheres harmful influence Release and dispersion analogue system in another embodiment of the present invention includes：
Gas diffusion model library, for storing gas diffusion model, including Calm dispersion model, Gauss model, BM models, Sutton models, FEM3 models；
Gas diffusion model library selecting module, for according to incident region real-time weather situation, field conditions and leakage gas Body type, corresponding gas diffusion model is selected from gas diffusion model library, and selection result is sent to gas leakage diffusion Emulation module；
Learner combined training module, for training the combination of gas leakage type corresponding learner, and by training result It is sent to gas leakage Diffusion Simulation module；
Gas diffusion model generation module, for the gas diffusion model selected according to gas diffusion model library selecting module The learner combination trained with learner combined training module, generates last gas leakage Diffusion Simulation.
Wherein, gas diffusion model library selecting module is including real-time weather situation determines, field conditions determine and leaks gas Body type determines three subelements, wherein,
Real-time weather situation determination subelement is used for according to wind direction, wind direction, wind-force, wind speed and atmospheric stability, determines thing Send out region real-time weather situation；
Field conditions determination subelement is used to, according to storage tank pressure, leakage area, leak time and leakage shape, determine thing Send out region field conditions；
Gas leakage type determination unit is used for the type for determining spillage risk gas, and dangerous gas type is divided into Three kinds：Light gas, neutral gas, severe gas.
Further, learner combined training module obtains the corresponding learner combination tool of gas leakage type by training Body is：
(a) harmful influence gas training set, including light gas training set, neutral gas training set, severe gas training set are established, Specific representation is as follows：
Tname={ (x1, y1), (x2, y2) ..., (xN, yN), name is light gas, neutral gas, wherein severe gas, xi Represent some positions chosen from the true diffusion space of gas, yiRepresent the gas concentration of the position, i=1...N；
(b) according to different gas types, corresponding gas diffusion model, the base as harmful influence gas training set are selected This learner；
(c) according to different gas, the weights distribution of training data is initialized
D1=(w11..., w1i..., w1N),I=1,2 ..., N, wherein, w1iIt is to choose harmful influence gas (x in training seti, yi) probability, N is the data volume in harmful influence gas training set；
(d) the corresponding diffusion model number of every kind of gas is M, and each diffusion model is a basic studies device Gm, m= 1 ..., M；
(e) G is calculatedm(x) the large error rate on training dataset, x=x1..., xN, wherein, what large error referred to It is | Gm(xi)-yi| ＞ α, α yi1 percent, i.e. model Gm(x) concentration calculated is more than true with actual gas concentration difference Percent a period of time of gas concentration, Gm(x) there is " large error " rate.Specifically it is calculated as：
Wherein, emRepresent Gm(x) exist Large error rate on training dataset, wmiCorresponding to (xi, yi) probability,
P(|Gm(xi)-yi| ＞ α) represent | Gm(xi)-yi| the ratio that ＞ α are concentrated in training data
I(|Gm(xi)-yi| ＞ α) what is represented is to work as | Gm(xi)-yi| when ＞ α are set up,
I(|Gm(xi)-yi| ＞ α) value be 1, be otherwise 0,Represent large error rate Specific formula for calculation；
(f) basic studies device G is calculatedm(x) coefficient
(g) the weights distribution of training data is updated
Wherein, ZmIt is standardizing factor so that Dm+1As a probability distribution,
(h) different gas is directed to, builds the linear combination of basic studies device
Urban atmospheres harmful influence proposed by the present invention reveals Diffusion Simulation method and system, utilizes integrating in machine learning Method, effectively integrates the advantage of each discrete gas diffusion model, selects suitable Diffusion Simulation model, simulates the true expansion of gas The process of dissipating, the diffusion for predicting real hazard gas have good effect.
It should be appreciated that the functional unit or ability that describe in the present specification be referred to alternatively as or be denoted as component, module or System, more specifically to emphasize their realization independence.For example, component, module or system can be implemented as hardware circuit, its Including customizing ultra-large integrated (VLSI) circuit OR gate array, such as ready-made semiconductor, logic chip, transistor, or its His discrete assembly.Component or module can also realize in programmable hardware device, such as field programmable gate array, programmable array Logic, programmable logic device etc..Component or module can also be real in the software for being performed by various types of processors It is existing.For example, the component or module of the identification of executable code can include one or more computer instructions physically or logically, It can be with for example, be organized as object, program or function.However, the component or module that are identified need not be physically positioned at Together, but the disparate instruction for being stored in diverse location can be included, it includes component or mould when being bonded together in logic Block simultaneously realizes the regulation purpose for component or module.
It should be appreciated that spy is not limited to above by the effect that the present invention can realize by those skilled in the art The content not described, and the further advantage of the present invention will be more clearly understood from detailed description above.
It should be apparent to those skilled in the art that can be without departing from the spirit or scope of the present invention in the present invention In make various modifications and variations.Therefore, if it is contemplated that the present invention modifications and variations fall into subsidiary claim and In the range of their equivalents, then the present invention covers these modifications and variations.

Claims (7)

1. a kind of harmful influence gas Release and dispersion model generating method, it is characterised in that comprise the following steps：
(1) according to wind direction, wind direction, wind-force, wind speed and atmospheric stability, incident region real-time weather situation is determined；
According to storage tank pressure, leakage area, leak time and leakage shape, incident region field conditions are determined；
Establish gas diffusion model library, including Calm dispersion model, Gauss model, BM models, Sutton models, FEM3 models；
Determine the type of spillage risk gas, dangerous gas type is divided into three kinds：Light gas, neutral gas, severe gas；
(2) incident region real-time weather situation, field conditions and gas leakage type are based on, phase is selected from gas diffusion model library The gas diffusion model answered；
(3) the corresponding learner of gas leakage type is obtained by training to combine；
(4) the gas diffusion model selected based on step 2, and the learner combination that step 3 obtains, generate last leakage gas Body Diffusion Simulation.
2. according to the method described in claim 1, corresponded to it is characterized in that, obtaining gas leakage type by training in step 3 Learner combination it is specific as follows：
(a) harmful influence gas training set, including light gas training set, neutral gas training set, severe gas training set are established, specifically Representation is as follows：
Tname={ (x1, y1), (x2, y2) ...,(xN, yN), name is light gas, neutral gas, wherein severe gas, xiRepresent Some positions chosen from the true diffusion space of gas, yiRepresent the gas concentration of the position, i=1...N；
(b) according to different gas types, corresponding gas diffusion model is selected, basic as harmful influence gas training set is learned Practise device；
(c) according to different gas, the weights distribution of training data is initialized
<mrow> <msub> <mi>D</mi> <mn>1</mn> </msub> <mo>=</mo> <mrow> <mo>(</mo> <msub> <mi>w</mi> <mn>11</mn> </msub> <mo>,</mo> <mn>...</mn> <mo>,</mo> <msub> <mi>w</mi> <mrow> <mn>1</mn> <mi>i</mi> </mrow> </msub> <mo>,</mo> <mn>...</mn> <mo>,</mo> <msub> <mi>w</mi> <mrow> <mn>1</mn> <mi>N</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mi>w</mi> <mrow> <mn>1</mn> <mi>i</mi> </mrow> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mi>N</mi> </mfrac> <mo>,</mo> <mi>i</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mn>2</mn> <mo>,</mo> <mn>...</mn> <mo>,</mo> <mi>N</mi> <mo>,</mo> </mrow>
Wherein, w1iIt is to choose (x in harmful influence gas training seti, yi) probability, N is the data in harmful influence gas training set Amount；
(d) the corresponding simplified diffusion model number of every kind of gas is M, and each diffusion model that simplifies is a basic studies device Gm, m= 1 ..., M, for example for neutral gas, M 3, diffusion model is respectively basic studies device G1、G2、G3；
(e) G is calculatedm(x) the large error rate on training datasetWherein, large error refers to | Gm(xi)-yi| ＞ α, α yi1 percent, wmiCorresponding to (xi, yi) probability, P (| Gm(xi)-yi| ＞ α) represent | Gm(xi)-yi| ＞ α exist The ratio that training data is concentrated；
(f) basic studies device G is calculatedm(x) coefficient
<mrow> <msub> <mi>&amp;beta;</mi> <mi>m</mi> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mn>2</mn> </mfrac> <mi>log</mi> <mfrac> <mrow> <mn>1</mn> <mo>-</mo> <msub> <mi>e</mi> <mi>m</mi> </msub> </mrow> <msub> <mi>e</mi> <mi>m</mi> </msub> </mfrac> <mo>;</mo> </mrow>
(g) the weights distribution of training data is updated
<mfenced open = "" close = ""> <mtable> <mtr> <mtd> <mrow> <msub> <mi>D</mi> <mrow> <mi>m</mi> <mo>+</mo> <mn>1</mn> </mrow> </msub> <mo>=</mo> <mrow> <mo>(</mo> <mrow> <msub> <mi>w</mi> <mrow> <mi>m</mi> <mo>+</mo> <mn>1</mn> <mo>,</mo> <mn>1</mn> </mrow> </msub> <mo>,</mo> <mn>...</mn> <mo>,</mo> <msub> <mi>w</mi> <mrow> <mi>m</mi> <mo>+</mo> <mn>1</mn> <mo>,</mo> <mi>i</mi> </mrow> </msub> <mo>,</mo> <mn>...</mn> <mo>,</mo> <msub> <mi>w</mi> <mrow> <mi>m</mi> <mo>+</mo> <mn>1</mn> <mo>,</mo> <mi>N</mi> </mrow> </msub> </mrow> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mi>w</mi> <mrow> <mi>m</mi> <mo>+</mo> <mn>1</mn> <mo>,</mo> <mi>i</mi> </mrow> </msub> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mo>=</mo> <mfrac> <msub> <mi>w</mi> <mrow> <mi>m</mi> <mi>i</mi> </mrow> </msub> <msub> <mi>Z</mi> <mi>m</mi> </msub> </mfrac> <mi>exp</mi> <mrow> <mo>(</mo> <mrow> <mo>-</mo> <msub> <mi>&amp;beta;</mi> <mi>m</mi> </msub> <msub> <mi>y</mi> <mi>i</mi> </msub> <msub> <mi>G</mi> <mi>m</mi> </msub> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> </mrow> <mo>)</mo> </mrow> <mo>,</mo> <mi>i</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mn>2</mn> <mo>,</mo> <mn>...</mn> <mo>,</mo> <mi>N</mi> </mrow> </mtd> </mtr> </mtable> </mfenced>
Wherein, ZmIt is standardizing factor so that Dm+1As a probability distribution,
As m+1 ＞ M, enter step (h)
(h) different gas is directed to, builds the linear combination of basic studies device
<mrow> <mi>f</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>=</mo> <msubsup> <mi>&amp;Sigma;</mi> <mrow> <mi>m</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>M</mi> </msubsup> <msub> <mi>&amp;beta;</mi> <mi>m</mi> </msub> <msub> <mi>G</mi> <mi>m</mi> </msub> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>.</mo> </mrow>
3. according to the method described in claim 2, it is characterized in that, in step b, if m is neutral gas, from gas diffusion Three kinds of Gauss model, Sutton models, BM models models are chosen in model library, M is set to 3 in step d；If m is light gas, Select Calm dispersion model, Gauss model, M is set to 2 in step d；The gas if m attaches most importance to, selects BM, FEM3 model, step d Middle M is set to 2.
4. according to the method described in claim 3, it is characterized in that, in step 4, last gas leakage Diffusion Simulation tool is generated Body is：The β obtained based on the training stagem, and corresponding gas diffusion model is combined, utilize The concentration of current gas each point is obtained, for being diffused emulation to leakage gas.
5. a kind of harmful influence gas Release and dispersion model generates system, it is characterised in that including with lower module：
Gas diffusion model library, for storing gas diffusion model, including Calm dispersion model, Gauss model, BM models, Sutton models, FEM3 models；
Gas diffusion model library selecting module, for according to incident region real-time weather situation, field conditions and gas leakage class Type, corresponding gas diffusion model is selected from gas diffusion model library, and selection result is sent to gas leakage Diffusion Simulation Module；
Learner combined training module, for training the corresponding learner combination of gas leakage type, and training result is sent Give gas leakage Diffusion Simulation module；
Gas diffusion model generation module, for the gas diffusion model selected according to gas diffusion model library selecting module and The learner combination that device combined training module is trained is practised, generates last gas leakage Diffusion Simulation.
6. Diffusion Simulation system according to claim 5, it is characterised in that gas diffusion model library selecting module includes real When weather condition determines, field conditions determine and gas leakage type determines three subelements, wherein,
Real-time weather situation determination subelement is used for according to wind direction, wind direction, wind-force, wind speed and atmospheric stability, determines incident area Domain real-time weather situation；
Field conditions determination subelement is used to, according to storage tank pressure, leakage area, leak time and leakage shape, determine incident area Domain field conditions；
Gas leakage type determination unit is used for the type for determining spillage risk gas, and dangerous gas type is divided into three Kind：Light gas, neutral gas, severe gas.
7. Diffusion Simulation system according to claim 5, it is characterised in that learner combined training module passes through trained It is specially to the corresponding learner combination of gas leakage type：
(a) harmful influence gas training set, including light gas training set, neutral gas training set, severe gas training set are established, specifically Representation is as follows：
Tname={ (x1, y1), (x2, y2) ..., (xN, yN)), name is light gas, neutral gas, wherein severe gas, xiRepresent from Some positions chosen in the true diffusion space of gas, yiRepresent the gas concentration of the position, i=1...N；
(b) according to different gas types, corresponding gas diffusion model is selected, basic as harmful influence gas training set is learned Practise device；
(c) according to different gas, the weights distribution of training data is initializedWherein, w1iIt is to choose in harmful influence gas training set (xi, yi) probability, N is the data volume in harmful influence gas training set；
(d) the corresponding diffusion model number of every kind of gas is M, and each diffusion model is a basic studies device Gm, m=1 ..., M；
(e) G is calculatedm(x) the large error rate on training datasetWherein, large error refers to | Gm (xi)-yi| ＞ α, α yi1 percent, wmiCorresponding to (xi, yi) probability, P (| Gm(xi)-yi| ＞ α) represent | Gm(xi)- yi| the ratio that ＞ α are concentrated in training data；
(f) basic studies device G is calculatedm(x) coefficient
<mrow> <msub> <mi>&amp;beta;</mi> <mi>m</mi> </msub> <mo>=</mo> <mfrac> <mn>1</mn> <mn>2</mn> </mfrac> <mi>log</mi> <mfrac> <mrow> <mn>1</mn> <mo>-</mo> <msub> <mi>e</mi> <mi>m</mi> </msub> </mrow> <msub> <mi>e</mi> <mi>m</mi> </msub> </mfrac> <mo>;</mo> </mrow>
(g) the weights distribution of training data is updated
<mfenced open = '' close = ''> <mtable> <mtr> <mtd> <mrow> <msub> <mi>D</mi> <mrow> <mi>m</mi> <mo>+</mo> <mn>1</mn> </mrow> </msub> <mo>=</mo> <mrow> <mo>(</mo> <mrow> <msub> <mi>w</mi> <mrow> <mi>m</mi> <mo>+</mo> <mn>1</mn> <mo>,</mo> <mn>1</mn> </mrow> </msub> <mo>,</mo> <mn>...</mn> <mo>,</mo> <msub> <mi>w</mi> <mrow> <mi>m</mi> <mo>+</mo> <mn>1</mn> <mo>,</mo> <mi>i</mi> </mrow> </msub> <mo>,</mo> <mn>...</mn> <mo>,</mo> <msub> <mi>w</mi> <mrow> <mi>m</mi> <mo>+</mo> <mn>1</mn> <mo>,</mo> <mi>N</mi> </mrow> </msub> </mrow> <mo>)</mo> </mrow> <mo>,</mo> <msub> <mi>w</mi> <mrow> <mi>m</mi> <mo>+</mo> <mn>1</mn> <mo>,</mo> <mi>i</mi> </mrow> </msub> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mo>=</mo> <mfrac> <msub> <mi>w</mi> <mrow> <mi>m</mi> <mi>i</mi> </mrow> </msub> <msub> <mi>Z</mi> <mi>m</mi> </msub> </mfrac> <mi>exp</mi> <mrow> <mo>(</mo> <mrow> <mo>-</mo> <msub> <mi>&amp;beta;</mi> <mi>m</mi> </msub> <msub> <mi>y</mi> <mi>i</mi> </msub> <msub> <mi>G</mi> <mi>m</mi> </msub> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> </mrow> <mo>)</mo> </mrow> <mo>,</mo> <mi>i</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mn>2</mn> <mo>,</mo> <mn>...</mn> <mo>,</mo> <mi>N</mi> </mrow> </mtd> </mtr> </mtable> </mfenced>
Wherein, ZmIt is standardizing factor so that Dm+1As a probability distribution,
<mrow> <msub> <mi>Z</mi> <mi>m</mi> </msub> <mo>=</mo> <msubsup> <mi>&amp;Sigma;</mi> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>N</mi> </msubsup> <msub> <mi>w</mi> <mrow> <mi>m</mi> <mi>i</mi> </mrow> </msub> <mi>exp</mi> <mrow> <mo>(</mo> <mo>-</mo> <msub> <mi>&amp;beta;</mi> <mi>m</mi> </msub> <msub> <mi>y</mi> <mi>i</mi> </msub> <msub> <mi>G</mi> <mi>m</mi> </msub> <mo>(</mo> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>)</mo> <mo>)</mo> </mrow> <mo>;</mo> </mrow>
(h) different gas is directed to, builds the linear combination of basic studies device
<mrow> <mi>f</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>=</mo> <msubsup> <mi>&amp;Sigma;</mi> <mrow> <mi>m</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>M</mi> </msubsup> <msub> <mi>&amp;beta;</mi> <mi>m</mi> </msub> <msub> <mi>G</mi> <mi>m</mi> </msub> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>.</mo> </mrow>
CN201710934322.7A 2017-10-09 2017-10-09 Diffusion simulation method and system for leakage of urban gas hazardous chemical Active CN107944070B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710934322.7A CN107944070B (en) 2017-10-09 2017-10-09 Diffusion simulation method and system for leakage of urban gas hazardous chemical

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710934322.7A CN107944070B (en) 2017-10-09 2017-10-09 Diffusion simulation method and system for leakage of urban gas hazardous chemical

Publications (2)

Publication Number Publication Date
CN107944070A true CN107944070A (en) 2018-04-20
CN107944070B CN107944070B (en) 2021-06-25

Family

ID=61936108

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710934322.7A Active CN107944070B (en) 2017-10-09 2017-10-09 Diffusion simulation method and system for leakage of urban gas hazardous chemical

Country Status (1)

Country Link
CN (1) CN107944070B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109737309A (en) * 2019-01-09 2019-05-10 慧感(上海)物联网科技有限公司 A kind of risk source based on risk identification leaks source tracing method and its system surely

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030216855A1 (en) * 2002-05-15 2003-11-20 Liang Cho Y. NOx emission-control system using a virtual sensor
KR20050106157A (en) * 2004-05-04 2005-11-09 주식회사 엔버스 Computation smell emssion rate using smell perception sensor and smell monitoring-prediction method from smell dispersion model
CN102608285A (en) * 2012-02-21 2012-07-25 南京工业大学 Organic mixture blasting characteristic prediction method based on support vector machine
CN104598736A (en) * 2015-01-22 2015-05-06 西安交通大学 Roller bearing service life predicting model of self-adaptive multi-kernel combination relevance vector machine
CN104657573A (en) * 2014-01-06 2015-05-27 北京千安哲信息技术有限公司 Leaking gas diffusion forecasting method used for three-dimensional space
US20160298447A1 (en) * 2015-04-07 2016-10-13 West Virginia University Leakage detection using smart field technology
CN106052962A (en) * 2016-05-13 2016-10-26 重庆大学 Gas source positioning technology in wireless sensor network without wind speed and wind direction measuring
CN106197859A (en) * 2016-07-04 2016-12-07 重庆大学 A kind of gas source localization method considering that the confined space retrains
CN106671902A (en) * 2015-11-10 2017-05-17 冯春魁 Measuring, calculating, supervising, monitoring and processing method and system synthesizing vehicle data
CN106961104A (en) * 2017-04-06 2017-07-18 新疆大学 Wind power forecasting method based on data analysis and combination basis function neural network

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030216855A1 (en) * 2002-05-15 2003-11-20 Liang Cho Y. NOx emission-control system using a virtual sensor
KR20050106157A (en) * 2004-05-04 2005-11-09 주식회사 엔버스 Computation smell emssion rate using smell perception sensor and smell monitoring-prediction method from smell dispersion model
CN102608285A (en) * 2012-02-21 2012-07-25 南京工业大学 Organic mixture blasting characteristic prediction method based on support vector machine
CN104657573A (en) * 2014-01-06 2015-05-27 北京千安哲信息技术有限公司 Leaking gas diffusion forecasting method used for three-dimensional space
CN104598736A (en) * 2015-01-22 2015-05-06 西安交通大学 Roller bearing service life predicting model of self-adaptive multi-kernel combination relevance vector machine
US20160298447A1 (en) * 2015-04-07 2016-10-13 West Virginia University Leakage detection using smart field technology
CN106671902A (en) * 2015-11-10 2017-05-17 冯春魁 Measuring, calculating, supervising, monitoring and processing method and system synthesizing vehicle data
CN106052962A (en) * 2016-05-13 2016-10-26 重庆大学 Gas source positioning technology in wireless sensor network without wind speed and wind direction measuring
CN106197859A (en) * 2016-07-04 2016-12-07 重庆大学 A kind of gas source localization method considering that the confined space retrains
CN106961104A (en) * 2017-04-06 2017-07-18 新疆大学 Wind power forecasting method based on data analysis and combination basis function neural network

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
PENG XIUYAN ET AL.: "Ship Motion Prediction of Combination Forecasting Model Based on Adaptive Variable Weight", 《PROCEEDINGS OF THE 34TH CHINESE CONTROL CONFERENCE》 *
邓振文等: "基于模糊控制的气体扩散模型选择", 《化工自动化及仪表》 *
陈彦平: "基于MATLAB的氨泄漏扩散动态模拟研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109737309A (en) * 2019-01-09 2019-05-10 慧感(上海)物联网科技有限公司 A kind of risk source based on risk identification leaks source tracing method and its system surely
CN109737309B (en) * 2019-01-09 2021-05-04 慧感(上海)物联网科技有限公司 Risk source and leakage tracing method and system based on risk identification

Also Published As

Publication number Publication date
CN107944070B (en) 2021-06-25

Similar Documents

Publication Publication Date Title
JP7232610B2 (en) Fluid leak detection system, fluid leak detection device, design support system, design support device, and learning device
CN104217216B (en) Generate method and apparatus, the method and apparatus for detecting target of detection model
Mashford et al. Leak detection in simulated water pipe networks using SVM
WO2019241570A8 (en) Quantum virtual machine for simulation of a quantum processing system
CN106599933A (en) Text emotion classification method based on the joint deep learning model
WO2020155300A1 (en) Model prediction method and device
CN107368613A (en) Short text sentiment analysis method and device
CN106023035A (en) Person evacuation safety evaluation method, device and apparatus in toxic gas leakage accident
CN112001110A (en) Structural damage identification monitoring method based on vibration signal space real-time recursive graph convolutional neural network
CN109143796B (en) Method and device for determining photoetching light source and method and device for training model
CN106022215A (en) Face feature point positioning method and device
Ayob et al. Analysis of pruned neural networks (MobileNetV2-YOLO v2) for underwater object detection
CN107944070A (en) A kind of Diffusion Simulation method and system of urban atmospheres harmful influence leakage
CN108388113B (en) Least square method supporting vector machine soft-measuring modeling method based on distribution estimation local optimum
CN118114575A (en) Combustible gas leakage diffusion range prediction simulation method and system
US20160092782A1 (en) Information processing apparatus, program and information processing method
Ulrich et al. Operator timing of task level primitives for use in computation-based human reliability analysis
CN107958274A (en) A kind of structural fire protection safety index computational methods based on big data sorting algorithm
Marcano et al. A methodology for building a data-enclosing tunnel for automated online-feedback in simulator Training
CN114463613A (en) Fault detection method and system based on residual error network and Faster R-CNN
Kozik et al. Studying the method of adaptive prediction of forest fire evolution on the basis of recurrent neural networks
Bhattacharjee Thermodynamics: an interactive approach
CN115599471A (en) Game course design system and method based on accident analysis
JP2022050973A (en) Information processing apparatus and computer program
CN104464487A (en) Direct combustion type liquid nitrogen pump device simulating training system and method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant