CN115496353A - Intelligent risk assessment method for compressed natural gas filling station - Google Patents
Intelligent risk assessment method for compressed natural gas filling station Download PDFInfo
- Publication number
- CN115496353A CN115496353A CN202211133665.0A CN202211133665A CN115496353A CN 115496353 A CN115496353 A CN 115496353A CN 202211133665 A CN202211133665 A CN 202211133665A CN 115496353 A CN115496353 A CN 115496353A
- Authority
- CN
- China
- Prior art keywords
- natural gas
- filling station
- gas filling
- compressed natural
- risk
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 title claims abstract description 134
- 239000003345 natural gas Substances 0.000 title claims abstract description 67
- 238000012502 risk assessment Methods 0.000 title claims abstract description 38
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012706 support-vector machine Methods 0.000 claims abstract description 44
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 31
- 238000011156 evaluation Methods 0.000 claims abstract description 27
- 238000012549 training Methods 0.000 claims abstract description 18
- 238000005457 optimization Methods 0.000 claims abstract description 17
- 238000013210 evaluation model Methods 0.000 claims abstract description 13
- 238000010276 construction Methods 0.000 claims abstract description 8
- 238000012360 testing method Methods 0.000 claims abstract description 7
- 230000006870 function Effects 0.000 claims description 48
- 230000003247 decreasing effect Effects 0.000 claims description 6
- 239000007789 gas Substances 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 5
- 238000010801 machine learning Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 230000007547 defect Effects 0.000 claims description 4
- 201000004569 Blindness Diseases 0.000 claims description 3
- 150000001875 compounds Chemical class 0.000 claims description 3
- 238000013461 design Methods 0.000 claims description 3
- 238000009826 distribution Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000013519 translation Methods 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000007670 refining Methods 0.000 abstract 1
- 238000004519 manufacturing process Methods 0.000 description 5
- 238000013473 artificial intelligence Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 241000238814 Orthoptera Species 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 231100001267 hazard identification Toxicity 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- General Health & Medical Sciences (AREA)
- Water Supply & Treatment (AREA)
- Computational Linguistics (AREA)
- Primary Health Care (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Public Health (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to an intelligent risk assessment method for a compressed natural gas filling station; the method comprises the following construction steps: step 1, constructing a wedge wave multi-core support vector machine; step 2, determining an optimized wedge wave support vector machine optimized by an improved locust algorithm; step 3, determining a risk assessment index system of the compressed natural gas filling station; and 4, collecting data related to a risk index system of the compressed natural gas filling station and determining input sample data by utilizing an expert evaluation method and a questionnaire method. And 5, dividing the risk level of the compressed natural gas filling station into five grades and taking the risk grade as output sample data of the evaluation model. Step 6, collecting related risk assessment data of the compressed natural gas filling station and determining a training sample and a testing sample; and 7, training the evaluation model by using the training data sample and evaluating the risk level of the compressed natural gas filling station to be evaluated by using the trained evaluation model. The petrochemical enterprise oil refining device energy management system optimization method utilizing the structure is provided. The method and the device can improve the precision and the efficiency of risk assessment of the compressed natural gas filling station, and further can improve the safety management level of the compressed natural gas filling station.
Description
Technical Field
The invention relates to an intelligent risk assessment method for a compressed natural gas filling station, in particular to an intelligent risk assessment method for a compressed natural gas filling station based on a wedge-shaped wave multi-core support vector machine optimized by an improved locust algorithm.
Background
The compressed natural gas has the characteristics of low cost, no pollution, convenient use and the like, and a compressed natural gas filling station usually stores a certain amount of natural gas through a storage tank. In recent years, the number of compressed natural gas filling stations has been increasing. The rapid development of the compressed natural gas industry has led to field dispersion and regulatory difficulties. In addition, many places have relatively weak management work and control means are behind. As a result, many new cng gas stations have many vulnerabilities and problems in service management, equipment monitoring, safety production, etc. Due to corrosion or material defects, the natural gas storage tank may leak, resulting in accidents such as fire and explosion, and accidents of safety production sometimes occur, resulting in casualties and property loss. Therefore, safety issues for cng gas stations are of increasing public concern. In view of the high risk status of the cng, it is necessary to perform hazard identification risk analysis on risk factors that may exist in the production operation of the cng. And combining the accident statistical analysis result, performing preliminary risk analysis on the production device of the compressed natural gas filling station by adopting a qualitative evaluation method, finding out the reasons of common accidents of the compressed natural gas filling station, and providing technical management measures for safe production operation of the compressed natural gas filling station. The risk assessment research of the compressed natural gas filling station ensures the safety of the compressed natural gas filling station and peripheral facilities thereof.
In recent years, with the vigorous development of artificial intelligence technology, machine learning algorithms based on artificial intelligence are becoming more mature. The machine learning algorithm based on artificial intelligence can automatically acquire and release information in real time, and prevent serious or serious damage. Because the risk assessment of the compressed natural gas filling station has strong complexity, nonlinearity, uncertainty and instantaneity, the risk assessment by adopting the traditional mathematical model has certain limitations. The traditional evaluation method has the disadvantages of large subjective randomness and fuzziness, relatively complex operation and lack of self-learning capability. The non-linear processing capability of the support vector machine is achieved by a "kernel mapping" method. For kernel mapping, the kernel function must satisfy the Mercer condition. For example, the gaussian kernel function is a widely used kernel function, and shows good mapping performance when analyzing a non-linearity problem.However, for existing kernel functions, the support vector machine cannot be at some L 2 The (R) subspace approximates any function because existing kernel functions cannot generate a complete set of bases on that subspace by translating.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides an intelligent risk assessment method for a compressed natural gas filling station of a wedge-shaped wave multi-core support vector machine for optimizing an improved locust algorithm. The method and the device can improve the accuracy and efficiency of risk assessment of the compressed natural gas filling station, and further can improve the safety management level of the compressed natural gas filling station.
The technical scheme of the invention comprises the following steps:
step 1: construction of wedge wave multi-core support vector machine
The wedge wave transformation with good robustness to changes in different scales and directions is combined with the support vector machine to establish the wedge wave support vector machine. Aiming at the blindness problem of the single kernel function support vector machine in the evaluation when processing the machine learning task of the multi-feature set, the multi-kernel support vector machine is generated through the weighted linear addition of the local kernel function and the global kernel function to classify the data, so that the classification precision is further improved, and the risk evaluation precision of the compressed natural gas filling station is further improved.
Step 1-1: constructing a combined kernel function, wherein the corresponding formula is as follows:
in the formula, λ i Represents a weight coefficient, k i (x, y) represents a single-core function, and the expression is as follows:
wherein c represents a scale factor, d i And e i Representing a translation factor.
Step 1-2: a regression function is defined as follows:
in the formula, ω represents a weight variable,
representing the mapping function and B the compensation factor.
Step 1-3: the objective function and boundary conditions are defined as follows:
wherein L represents the number of samples and D represents a penalty factor.
Step 1-4: lagrange objective functions and corresponding boundary conditions are constructed by utilizing Lagrange duality, as follows:
in the formula (I), the compound is shown in the specification,
θ i ,
χ i representing the lagrange operator.
According to formula (5):
calculating the ratio of w, B,
ε i and substituting the result back to equation (9) yields the following equation:
the minimization equation can be derived by optimization as follows:
step 1-5: the multi-core support vector machine is obtained by introducing a multi-core function, as shown in the following
1-6: determining a decision function of the wedge wave multi-kernel support vector machine as follows:
step 2: wedge wave support vector machine optimized by determining improved locust algorithm
In order to improve the risk assessment effect of the wedge wave multi-nuclear support vector machine, an improved locust optimization algorithm with better global optimization performance and higher convergence precision is applied to the optimization of penalty factors, kernel function parameters and weights of the wedge wave multi-nuclear support vector machine.
Step 2-1: basic parameters of the improved locust optimization algorithm are initialized, and the basic parameters comprise population size, spatial dimension, maximum iteration number and initial position.
Step 2-2: updating the location of the locust population according to the following formula:
in the formula, T d Representing the target position of the locust population, eta represents the attenuation coefficient, omega l Representing the weight coefficient, t representing the current number of iterations,. Eta max And η min Representing the maximum and minimum attenuation factors. s (-) represents the interaction force function between locust populations as follows:
in the formula, f represents an attraction strength parameter, and r represents an attraction scale parameter.
Step 2-3: the position of a single locust is adjusted by using a Levy flight local search strategy, which is as follows:
X＝X+10×s ts ·L·X (19)
wherein L represents the Levy flight step length, s ts Representing a threshold function that can be used to control the flight method and the probability of change of locusts.
L is calculated using the formula:
L＝μ/|v| 1/β (20)
in the formula, β represents a constant between 0 and 2, and the parameters μ and v represent parameters that follow a normal distribution.
Step 2-4: when the locust individual searches the current optimal position, the original position is replaced; if no optimal solution is found, a linear decreasing parameter random jump-out strategy is used, as follows:
P i ＝(2-2rand(0,1))·P i (21)
in the formula, P i Represents the ith locust. When a new position P is found i And if the new position is better than the current position, replacing the current position by the new position.
Step 2-5: in order to reduce the attenuation coefficient at a faster speed in the early execution stage of the algorithm, locust individuals in the population can be ensured to be quickly close to the optimal target, and the convergence speed of the algorithm is improved; in the later iteration process of the algorithm, the reduction speed of the attenuation coefficient is reduced, so that the locust individual can carefully search the space, and the algorithm is prevented from falling into local optimum. The attenuation factor η is adjusted using a decreasing coefficient update strategy, as follows:
in the formula, N represents a current iteration coefficient, and N represents a maximum iteration number.
And step 3: determining risk assessment index system of compressed natural gas filling station
Risk influencing factors of the compressed natural gas filling station are systematically, comprehensively and qualitatively identified by collecting and analyzing data of the compressed natural gas filling station in detail in design, construction, operation, leakage, defect, personnel, society and economy. And determining a risk index system of the compressed natural gas filling station according to a risk forming mechanism. And determining a first-level index and a second-level index of the natural gas filling station risk assessment system.
And 4, step 4: collecting data related to a risk index system of a compressed natural gas filling station, determining a first-level index value through an expert evaluation method and a questionnaire method, and determining input sample data.
And 5: the risk level of the cng gas station is divided into five grades, I (fraction interval of [0.90,1.00 ]), II (fraction interval of [0.75,0.90)), III (fraction interval of [0.60,0.75)), IV (fraction interval of [0.45,0.60)), and V (fraction interval of [0,0.45)), respectively. And the risk level is used as output sample data of the evaluation model.
Step 6: collecting related data of risk assessment of a compressed natural gas filling station, and dividing the data into two parts, namely a training data sample and a testing data sample.
And 7: training the evaluation model by using the training data sample, evaluating the risk level of the CNG filling station to be evaluated by using the trained evaluation model, and judging the safety condition of the CNG filling station to be evaluated.
The invention has the following advantages and effects:
the wedge wave multi-core support vector machine optimized by the improved locust algorithm is used for evaluating the risk of the compressed natural gas filling station, and the accuracy and the efficiency of the risk evaluation of the compressed natural gas filling station can be effectively improved, so that the effectiveness of the risk evaluation of the compressed natural gas filling station can be improved, a powerful theoretical basis is provided for the risk prevention and control measures formulated by the compressed natural gas filling station, and the wedge wave multi-core support vector machine has a wide development prospect.
Drawings
FIG. 1 training time for different risk assessment methods
Detailed Description
Examples
The technical scheme of the invention comprises the following steps:
step 1: construction of wedge wave multi-core support vector machine
The wedge wave transformation with good robustness to changes in different scales and directions is combined with the support vector machine to establish the wedge wave support vector machine. Aiming at the blindness problem of a single kernel function support vector machine in evaluation when the single kernel function support vector machine processes a machine learning task of a multi-feature set, a multi-kernel support vector machine is generated through the weighted linear addition of local and global kernel functions to classify data, so that the classification precision is further improved, and the risk evaluation precision of a compressed natural gas filling station is further improved.
Step 1-1: constructing a combined kernel function, wherein the corresponding formula is as follows:
in the formula, λ i Represents a weight coefficient, k i (x, y) represents a single-core function, and the expression is as follows:
wherein c represents a scale factor, d i And e i Representing a translation factor.
Step 1-2: a regression function is defined as follows:
in the formula, ω represents a weight variable,
representing the mapping function and B the compensation factor.
Step 1-3: the objective function and boundary conditions are defined as follows:
wherein L represents the number of samples and D represents a penalty factor.
Step 1-4: lagrange objective functions and corresponding boundary conditions are constructed by utilizing Lagrange duality, as follows:
in the formula (I), the compound is shown in the specification,
θ i ,
χ i representing the lagrange operator.
According to formula (5):
calculating the ratio of w, B,
ε i and substituting the result back to equation (9) yields the following equation:
the minimization equation can be derived by optimization as follows:
step 1-5: the multi-core support vector machine is obtained by introducing a multi-core function, as shown in the following
Step 1-6: determining a decision function of the wedge wave multi-kernel support vector machine as follows:
step 2: optimized wedge wave support vector machine for determining improved locust algorithm
In order to improve the risk assessment effect of the wedge wave multi-nuclear support vector machine, an improved locust optimization algorithm with better global optimization performance and higher convergence precision is applied to the optimization of penalty factors, kernel function parameters and weights of the wedge wave multi-nuclear support vector machine.
Step 2-1: basic parameters of the locust optimization algorithm are initialized, wherein the basic parameters comprise population size, spatial dimension, maximum iteration number and initial position.
Step 2-2: updating the location of the locust population according to the following formula:
in the formula, T d Representing the target position of the locust population, eta represents the attenuation coefficient, omega l Representing the weight coefficient, t representing the current iteration number, η max And η min Representing the maximum and minimum attenuation factors. s (-) represents the interaction force function between locust populations as follows:
in the formula, f represents an attraction strength parameter, and r represents an attraction scale parameter.
Step 2-3: the position of a single locust is adjusted by using a Levy flight local search strategy, as follows:
X＝X+10×s ts ·L·X (19)
wherein L represents the Levy flight step length, s ts Representing a threshold function that can be used to control the flight method and the probability of change of locusts.
L is calculated using the formula:
L＝μ/|v| 1/β (20)
in the formula, β represents a constant between 0 and 2, and the parameters μ and v represent parameters that follow a normal distribution.
Step 2-4: when the locust individual searches the current optimal position, the original position is replaced; if no optimal solution is found, a linear decreasing parameter random jump-out strategy is used, as follows:
P i ＝(2-2rand(0,1))·P i (21)
in the formula, P i Represents the ith locust. When a new position P is found i And if the new position is better than the current position, replacing the current position by the new position.
Step 2-5: in order to reduce the attenuation coefficient at a faster speed in the early execution stage of the algorithm, locust individuals in the population can be ensured to be quickly close to the optimal target, and the convergence speed of the algorithm is improved; in the later iteration process of the algorithm, the reduction speed of the attenuation coefficient is reduced, so that the locust individual can carefully search the space, and the algorithm is prevented from falling into local optimum. The attenuation factor η is adjusted using a decreasing coefficient update strategy, as follows:
where N represents the current iteration coefficient and N represents the maximum number of iterations.
And 3, step 3: system for determining risk assessment index of compressed natural gas filling station
Risk influencing factors of the compressed natural gas filling station are systematically, comprehensively and qualitatively identified by collecting and analyzing data of the compressed natural gas filling station in detail in design, construction, operation, leakage, defect, personnel, society and economy. And determining a risk index system of the compressed natural gas filling station according to a risk forming mechanism. And determining a first-level index and a second-level index of the natural gas filling station risk assessment system.
And 4, step 4: collecting data related to a risk index system of a compressed natural gas filling station, determining a first-level index value through an expert evaluation method and a questionnaire method, and determining input sample data.
And 5: the risk level of the cng gas station is divided into five grades, I (fraction interval of [0.90,1.00 ]), II (fraction interval of [0.75,0.90)), III (fraction interval of [0.60,0.75)), IV (fraction interval of [0.45,0.60)), and V (fraction interval of [0,0.45)), respectively. And the risk level is used as output sample data of the evaluation model.
And 6: collecting related data of risk assessment of a compressed natural gas filling station, and dividing the data into two parts, namely a training data sample and a testing data sample.
And 7: training the evaluation model by using the training data sample, evaluating the risk level of the compressed natural gas filling station to be evaluated by using the trained evaluation model, and judging the safety condition of the compressed natural gas filling station to be evaluated.
Specific examples are shown below:
in an embodiment, a compressed natural gas fueling station is selected for risk assessment, the compressed natural gas fueling station being of the following size: the air-entrapping capacity is 1000 standard cubic meters per hour, and the daily air-entrapping capacity is 10000 standard cubic meters. The compressed natural gas filling station mainly supplies gas for small and medium-sized vehicles.
The parameter settings for improving the locust algorithm are as follows: l =2.0, f =1.0, b min ＝0.35,b max ＝0.90,η min ＝0.40,η max =0.95, the size of the population is 350, and the maximum number of iterations is 400.
50 groups of related data of the risk assessment of the CNG filling station are obtained by a questionnaire method and an expert survey method, wherein the first 40 groups of data are used as training samples, and the last 10 groups of data are used as test samples.
In order to verify the effectiveness of the evaluation method provided by the invention, the risk evaluation is also carried out on the training sample by using a B-spline wavelet support vector machine optimized locust algorithm (BSWSVM-LA) and a particle swarm optimization support vector machine (SVM-PSA). The accuracy of the different evaluation models was evaluated using the Mean Absolute Error (MAE) and the Mean Absolute Percentage Error (MAPE). The risk assessment results of the CNG filling station based on different models are shown in Table 1.
Table 1 compressed natural gas risk assessment results based on different models
The training times for the different models are shown in fig. 1. As can be seen from fig. 1, the training time of the evaluation method provided by the present invention is 6.7s, which is the minimum of the three methods, and therefore, the risk evaluation method provided by the present invention can improve the efficiency of risk evaluation of the compressed natural gas filling station.
The risk evaluation is carried out on the test sample by using three evaluation methods trained by the training sample, and the comparison result is shown in table 2.
TABLE 2 Risk assessment results for test samples based on three methods
As can be seen from the results of the calculations in Table 2, the MAE range of the proposed evaluation method is 0.82 to 0.92, the MAE range of BSWSVM-LA is 3.27 to 3.68, and the MEA range of SVM-PSA is 4.17 to 4.57. In addition, the evaluation results of the evaluation method provided by the invention are all correct, the BSWSVM-LA evaluation result has 4 errors, and the SVM-PSA evaluation result has 7 errors, so that the provided WMKSVM-ILA has the highest evaluation correctness in the three models.
Analysis results show that the risk assessment method for the compressed natural gas filling station based on the improved locust algorithm optimized wedge-shaped wave multi-core support vector machine can obtain the best effect, and is a risk assessment method with high practical value.
Claims (3)
1. An intelligent risk assessment method for a compressed natural gas filling station is characterized by comprising the following steps:
step 1: construction of wedge wave multi-core support vector machine
The wedge wave transformation with good robustness to changes of different scales and directions is combined with the support vector machine to establish the wedge wave support vector machine. Aiming at the blindness problem of a single kernel function support vector machine in evaluation when the single kernel function support vector machine processes a machine learning task of a multi-feature set, a multi-kernel support vector machine is generated through the weighted linear addition of local and global kernel functions to classify data, so that the classification precision is further improved, and the risk evaluation precision of a compressed natural gas filling station is further improved.
Step 2: optimized wedge wave support vector machine for determining improved locust algorithm
In order to improve the risk assessment effect of the wedge wave multi-nuclear support vector machine, an improved locust optimization algorithm with better global optimization performance and higher convergence precision is applied to the optimization of penalty factors, kernel function parameters and weights of the wedge wave multi-nuclear support vector machine.
And step 3: determining risk assessment index system of compressed natural gas filling station
Risk influencing factors of the compressed natural gas filling station are systematically, comprehensively and qualitatively identified by collecting and analyzing data of the compressed natural gas filling station in detail in terms of design, construction, operation, leakage, defects, personnel, society and economy. And determining a risk index system of the compressed natural gas filling station according to a risk forming mechanism. And determining a first-level index and a second-level index of the natural gas filling station risk assessment system.
And 4, step 4: collecting data related to a risk index system of a compressed natural gas filling station, determining a first-level index value through an expert evaluation method and a questionnaire method, and determining input sample data.
And 5: the risk level of the cng gas station is divided into five grades, I (fraction interval of [0.90,1.00 ]), II (fraction interval of [0.75,0.90)), III (fraction interval of [0.60,0.75)), IV (fraction interval of [0.45,0.60)), and V (fraction interval of [0,0.45)), respectively. And the risk level is used as output sample data of the evaluation model.
Step 6: collecting related risk assessment data of the CNG filling station, dividing the data into two parts, namely a training data sample and a testing data sample.
And 7: training the evaluation model by using the training data sample, evaluating the risk level of the compressed natural gas filling station to be evaluated by using the trained evaluation model, and judging the safety condition of the compressed natural gas filling station to be evaluated.
2. The intelligent risk assessment method for the compressed natural gas filling station according to claim 1, characterized in that the construction steps of the wedge wave multi-nuclear support vector machine are as follows:
step 1-1: constructing a combined kernel function, wherein the corresponding formula is as follows:
in the formula, λ i Represents a weight coefficient, k i (x, y) represents a single-core function, and the expression is as follows:
wherein c represents a scale factor, d i And e i Representing a translation factor.
Step 1-2: a regression function is defined as follows:
in the formula, ω represents a weight variable,
representing the mapping function and B the compensation factor.
Step 1-3: the objective function and boundary conditions are defined as follows:
wherein L represents the number of samples and D represents a penalty factor.
Step 1-4: lagrange objective functions and corresponding boundary conditions are constructed by utilizing Lagrange duality, as follows:
in the formula (I), the compound is shown in the specification,
θ i ,
χ i representing the lagrangian operator.
According to formula (5):
the calculation of the equation (7) for w, B,
ε i and substituting the result back to equation (9) yields the following equation:
the minimization equation can be derived by optimization as follows:
step 1-5: the multi-core support vector machine is obtained by introducing a multi-core function, as shown in the following
Step 1-6: determining a decision function of the wedge wave multi-kernel support vector machine as follows:
3. the intelligent risk assessment method for the compressed natural gas filling station according to claim 1, characterized in that the analysis of the improved locust algorithm comprises the following specific steps:
step 2-1: basic parameters of the improved locust optimization algorithm are initialized, and the basic parameters comprise population size, spatial dimension, maximum iteration number and initial position.
Step 2-2: updating the location of the locust population according to the following formula:
in the formula, T d Representing the target position of the locust population, eta represents the attenuation coefficient, omega l Representing the weight coefficient, t representing the current number of iterations,. Eta max And η min Representing the maximum and minimum attenuation factors. s (-) represents the interaction force function between locust populations as follows:
in the formula, f represents an attraction strength parameter, and r represents an attraction force scale parameter.
Step 2-3: the position of a single locust is adjusted by using a Levy flight local search strategy, as follows:
X＝X+10×s ts ·L·X (19)
wherein L represents the Levy flight step length, s ts The threshold function is expressed, and can be used for controlling the flight method and the variation probability of the locust.
L is calculated using the formula:
L＝μ/|v| 1/β (20)
in the formula, β represents a constant between 0 and 2, and the parameters μ and v represent parameters that follow a normal distribution.
Step 2-4: when the locust individual searches for the current optimal position, the original position is replaced; if no optimal solution is found, a linear decreasing parameter random jump-out strategy is used, as follows:
P i ＝(2-2rand(0,1))·P i (21)
in the formula, P i Represents the ith locust. When a new position P is found i And if the new position is better than the current position, replacing the current position by the new position.
Step 2-5: in order to reduce the attenuation coefficient at a faster speed in the early execution stage of the algorithm, locust individuals in the population can be ensured to be quickly close to the optimal target, and the convergence speed of the algorithm is improved; in the later iteration process of the algorithm, the reduction speed of the attenuation coefficient is reduced, so that the locust individual can carefully search the space, and the algorithm is prevented from falling into local optimum. The attenuation factor η is adjusted using a decreasing coefficient update strategy, as follows:
in the formula, N represents a current iteration coefficient, and N represents a maximum iteration number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211133665.0A CN115496353A (en) | 2022-09-19 | 2022-09-19 | Intelligent risk assessment method for compressed natural gas filling station |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211133665.0A CN115496353A (en) | 2022-09-19 | 2022-09-19 | Intelligent risk assessment method for compressed natural gas filling station |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115496353A true CN115496353A (en) | 2022-12-20 |
Family
ID=84470773
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211133665.0A Pending CN115496353A (en) | 2022-09-19 | 2022-09-19 | Intelligent risk assessment method for compressed natural gas filling station |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115496353A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116498908A (en) * | 2023-06-26 | 2023-07-28 | 成都秦川物联网科技股份有限公司 | Intelligent gas pipe network monitoring method based on ultrasonic flowmeter and Internet of things system |
| CN117952440A (en) * | 2024-03-26 | 2024-04-30 | 中用科技有限公司 | Chemical industry park production environment supervision method and system |
-
2022
- 2022-09-19 CN CN202211133665.0A patent/CN115496353A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116498908A (en) * | 2023-06-26 | 2023-07-28 | 成都秦川物联网科技股份有限公司 | Intelligent gas pipe network monitoring method based on ultrasonic flowmeter and Internet of things system |
| CN116498908B (en) * | 2023-06-26 | 2023-08-25 | 成都秦川物联网科技股份有限公司 | Intelligent gas pipe network monitoring method based on ultrasonic flowmeter and Internet of things system |
| US11953356B2 (en) | 2023-06-26 | 2024-04-09 | Chengdu Qinchuan Iot Technology Co., Ltd. | Methods and internet of things (IoT) systems for monitoring smart gas pipeline networks based on ultrasonic flowmeters |
| CN117952440A (en) * | 2024-03-26 | 2024-04-30 | 中用科技有限公司 | Chemical industry park production environment supervision method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115496353A (en) | Intelligent risk assessment method for compressed natural gas filling station | |
| CN102496069B (en) | Cable multimode safe operation evaluation method based on fuzzy analytic hierarchy process (FAHP) | |
| CN111950918B (en) | Market risk assessment method based on power transaction data | |
| CN105117602B (en) | A kind of metering device running status method for early warning | |
| CN110705873B (en) | Power distribution network running state portrait analysis method | |
| CN113344471B (en) | Method for representing weather environment adaptability of aircraft system | |
| CN108241901A (en) | A kind of transformer Forewarn evaluation method and device based on prediction data | |
| CN110889587B (en) | Power distribution network line risk assessment method | |
| CN113030633B (en) | GA-BP neural network-based power distribution network fault big data analysis method and system | |
| CN112257914B (en) | Aviation safety causal prediction method based on random forest | |
| CN112541636A (en) | Power transmission line icing risk early warning method and device, medium and electronic equipment | |
| CN110781206A (en) | Method for predicting whether electric energy meter in operation fails or not by learning meter-dismantling and returning failure characteristic rule | |
| CN113988519A (en) | Method for representing risk of cultural relic preservation environment in collection of cultural relics | |
| CN116029160B (en) | Method and system for constructing mapping model of defects and power generation efficiency loss of photovoltaic module | |
| CN117977536A (en) | Smart power grid load identification method | |
| CN116776209A (en) | Method, system, equipment and medium for identifying operation state of gateway metering device | |
| CN111160715A (en) | BP neural network based new and old kinetic energy conversion performance evaluation method and device | |
| CN116562622A (en) | Power transmission and transformation and distribution equipment safety assessment method | |
| CN115864644A (en) | Relay protection device state evaluation method, system, equipment and medium | |
| CN116151799A (en) | BP neural network-based distribution line multi-working-condition fault rate rapid assessment method | |
| CN115759395A (en) | Training of photovoltaic detection model, detection method of photovoltaic power generation and related device | |
| CN115564170A (en) | Reliability assessment method for intelligent power distribution terminal | |
| CN105931133A (en) | Distribution transformer replacement priority evaluation method and device | |
| CN111565121A (en) | Method and device for evaluating IT (information technology) support degree by personnel and technology | |
| Wang et al. | Optimizing the Uncertainty of PPM on Small Batch of Quality Data |
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 |