WO2022126544A1 - 确定污染源的方法、装置和计算机可读存储介质 - Google Patents
确定污染源的方法、装置和计算机可读存储介质 Download PDFInfo
- Publication number
- WO2022126544A1 WO2022126544A1 PCT/CN2020/137343 CN2020137343W WO2022126544A1 WO 2022126544 A1 WO2022126544 A1 WO 2022126544A1 CN 2020137343 W CN2020137343 W CN 2020137343W WO 2022126544 A1 WO2022126544 A1 WO 2022126544A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- wind direction
- sensor
- pollution source
- bayesian network
- target
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000012544 monitoring process Methods 0.000 claims abstract description 37
- 230000001364 causal effect Effects 0.000 claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 21
- 238000009826 distribution Methods 0.000 claims description 61
- 238000004422 calculation algorithm Methods 0.000 claims description 28
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000004088 simulation Methods 0.000 claims description 7
- 230000000694 effects Effects 0.000 abstract description 6
- 239000003344 environmental pollutant Substances 0.000 description 34
- 231100000719 pollutant Toxicity 0.000 description 34
- 238000010586 diagram Methods 0.000 description 13
- 230000006870 function Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 238000011109 contamination Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- RAHZWNYVWXNFOC-UHFFFAOYSA-N Sulphur dioxide Chemical compound O=S=O RAHZWNYVWXNFOC-UHFFFAOYSA-N 0.000 description 2
- 238000003491 array Methods 0.000 description 2
- 238000009827 uniform distribution Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000010297 mechanical methods and process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Definitions
- the present invention relates to the technical field of pollution monitoring, and in particular, to a method, a device and a computer-readable storage medium for determining a pollution source.
- An industrial park is an area of land that is delineated within a certain range and planned for the setting and use of industrial facilities.
- Industrial parks usually have strict regulations on the discharge of pollutants (liquid or gas). More and more organizations are realizing the importance of Eco-Industrial Park (EIP) for sustainable development.
- EIP Eco-Industrial Park
- monitoring points are usually arranged at fixed locations in industrial parks to monitor pollutant emissions, and based on the monitoring results, pollution sources are identified.
- Embodiments of the present invention provide a method, an apparatus, and a computer-readable storage medium for determining a pollution source.
- methods for identifying pollution sources including:
- a pollution source is determined from the potential pollution sources based on current sensor data when the wind direction is the target wind direction and the trained Bayesian network.
- means for determining the source of pollution including:
- the relationship determination module is used to determine the causal relationship between the sensor and the potential pollution source in the downward direction of the target wind based on the positional relationship between the sensor and the potential pollution source;
- a network establishment module configured to establish a Bayesian network corresponding to the target wind direction and including the causal relationship
- a training module for training the Bayesian network based on the historical sensor data when the wind direction is the target wind direction;
- a determination module configured to determine a pollution source from the potential pollution sources based on the current sensor data when the wind direction is the target wind direction and the trained Bayesian network.
- an apparatus for determining a source of contamination including a processor and a memory;
- An application program executable by the processor is stored in the memory for causing the processor to execute the method for determining a pollution source as described in any one of the above.
- a computer-readable storage medium in which computer-readable instructions are stored, the computer-readable instructions for performing the method of determining a pollution source as described in any of the above.
- the embodiment of the present invention establishes respective Bayesian networks corresponding to different target wind directions, and then uses the Bayesian network corresponding to the current target wind direction to accurately determine the pollution source. Therefore, the embodiment of the present invention takes into account the influence of the wind direction on the monitoring effect of the sensor, The contamination source is determined with high accuracy.
- it further includes: determining the prior probability distribution of the Bayesian network; based on the prior probability distribution of the Bayesian network and the historical sensor data when the wind direction is the target wind direction, take the confidence
- a propagation algorithm computes the posterior probability distribution of the Bayesian network.
- the posterior probability distribution of the Bayesian network is calculated by the belief propagation algorithm, and the Bayesian network can be rapidly trained.
- calculating the posterior probability distribution of the Bayesian network with the belief propagation algorithm includes: in each step of the iterative calculation of the belief propagation algorithm, set the wind direction as the target wind direction The wind speed is used as a parameter affecting the monitoring value of the sensor, wherein the larger the wind speed is, the smaller the monitoring value of the sensor is.
- the influence of wind speed on the monitoring effect of the sensor is also considered, and the wind speed is used as a parameter to participate in the training process of the Bayesian network, so that the trained Bayesian network is more accurate.
- determining the pollution source from the potential pollution sources includes:
- the potential pollution source corresponding to the maximum emission value is determined as the pollution source.
- the embodiment of the present invention calculates the emission probability distribution of each potential pollution source, and determines the potential pollution source corresponding to the maximum emission value as the pollution source, thereby realizing accurate identification of the pollution source.
- the training of the Bayesian network based on the historical sensor data when the wind direction is the target wind direction includes:
- the Bayesian network is trained based on the sensor simulation historical data when the wind direction is the target wind direction.
- the embodiments of the present invention can not only train the Bayesian network by using the actual historical data of the sensor, but also train the Bayesian network by using the simulated historical data of the sensor, thereby enriching the training data.
- FIG. 1 is an exemplary flowchart of a method for determining a pollution source according to an embodiment of the present invention.
- FIG. 2 is an exemplary schematic diagram of a topology diagram of an industrial park according to an embodiment of the present invention.
- FIG. 3 is an exemplary schematic diagram of a Bayesian network corresponding to a wind direction according to an embodiment of the present invention.
- FIG. 4 is an exemplary schematic diagram of determining a pollution source according to an embodiment of the present invention.
- FIG. 5 is an exemplary flowchart of a method for determining pollution sources in an industrial park according to an embodiment of the present invention.
- FIG. 6 is an exemplary structural diagram of an apparatus for determining a pollution source according to an embodiment of the present invention.
- FIG. 7 is an exemplary structural diagram of an apparatus for determining a pollution source according to an embodiment of the present invention.
- 602 network determination module 603 training module 604 Determine the module 700 A device to determine the source of contamination 701 processor 702 memory
- the embodiment of the present invention takes into account the influence of wind direction on the monitoring effect of the sensor, establishes in advance a Bayesian network corresponding to different wind directions for determining pollution sources, and uses the training data of each wind direction to train each wind direction. Then use the current sensor data at the current wind direction and the trained Bayesian network corresponding to the current wind direction to accurately determine the pollution source.
- the embodiments of the present invention also take into account the influence of wind speed on the monitoring effect of the sensor (for example, when the wind speed is strong, the monitored pollutant concentration is low), and the wind speed is used as a parameter to participate in the training process of the Bayesian network, The trained Bayesian network is more accurate.
- FIG. 1 is an exemplary flowchart of a method for determining a pollution source according to an embodiment of the present invention.
- the method 100 includes:
- Step 101 Based on the positional relationship between the sensor and the potential pollution source, determine the causal relationship between the sensor and the potential pollution source with the target wind down.
- the number of sensors may be one or more, and the number of potential pollution sources may also be one or more.
- the sensors are adapted to monitor physical quantities (eg, concentrations) of pollutants emitted by potential pollution sources.
- the deployment position of the sensor is fixed, and the deployment position of the potential pollution source is also fixed.
- the sensors are multiple sensors adapted to detect the respective types of pollutants.
- the applicant has found that when the wind direction is consistent with the detection direction of the sensor, the sensor is more likely to detect pollutants, and when the wind direction is opposite to the detection direction of the sensor, it is difficult for the sensor to detect pollutants. Therefore, several predetermined target wind directions can be determined first, and then based on the positional relationship between the sensor and the potential pollution source, the causal relationship between the sensor and the potential pollution source in the respective target wind directions can be determined.
- FIG. 2 is an exemplary schematic diagram of a topology diagram of an industrial park according to an embodiment of the present invention.
- a building 11 , a building 12 , and a building 13 are present in the industrial park 10 .
- a potential pollution source B in the building 11 there is a potential pollution source B in the building 11 , a potential pollution source D in the building 12 , and a potential pollution source A and a potential pollution source C in the building 13 .
- the potential pollution source A, the potential pollution source B, the potential pollution source C, and the potential pollution source D are adapted to emit the same type of pollutants, such as sulfur dioxide gas.
- the direction indicated by the arrow N is the north direction.
- the sensor X is arranged at the west edge of the industrial park 10
- the sensor Y is arranged at the south edge of the industrial park 10
- the sensor Z is arranged at the east edge of the industrial park 10 .
- the sensor X, the sensor Y and the sensor Z are adapted to monitor the pollutants emitted by the potential pollution source A, the potential pollution source B, the potential pollution source C and the potential pollution source D, respectively.
- sensor X, sensor Y and sensor Z are respectively set with a concentration threshold value, when the monitored pollutant exceeds the concentration threshold value, it is determined that the pollutant is monitored and the concentration threshold value is recorded.
- the causal relationship between the target wind down sensor and potential pollution sources can be determined.
- the number of target wind directions may be multiple. For example, when the wind direction angle is 90 degrees, it can be divided into four target wind directions: northwest wind, northeast wind, southwest wind and southeast wind. The smaller the wind direction angle, the greater the number of divided target wind directions.
- the causal relationship between sensor X, sensor Y and sensor Z and potential pollution source A, potential pollution source B, potential pollution source C and potential pollution source D includes:
- the target wind direction is the northwest wind (that is, the wind blows from the northwest)
- the pollutants emitted by the potential pollution source A, the potential pollution source B and the potential pollution source C will be monitored by the sensor Y.
- the pollutants emitted by the potential pollution source A will be monitored by the sensor X; the pollutants emitted by the potential pollution source B will be detected by the sensor X and the sensor X. Y is monitored; the pollutants emitted by potential pollution source D will be monitored by sensor Y.
- the pollutants emitted by the potential pollution source A will be monitored by the sensor Z; the pollutants emitted by the potential pollution source B will be monitored by the sensor Z. ; The pollutants emitted by the potential pollution source C will be monitored by the sensor Z; the pollutants emitted by the potential pollution source D will be monitored by the sensor Z.
- the pollutants emitted by the potential pollution source A will be monitored by the sensor X; the pollutants emitted by the potential pollution source C will be monitored by the sensor X. ; The pollutants emitted by the potential pollution source D will be monitored by the sensor X and the sensor Z.
- the causal relationship between the sensor and the potential pollution source is exemplarily described. Those skilled in the art can realize that this description is only exemplary, and is not intended to limit the implementation of the present invention. protected range. In fact, the causal relationship between the sensor and the potential pollution source can also have various implementations based on the different divisions of the wind direction and the positional relationship between the sensor and the potential pollution source.
- Step 102 Establish a Bayesian network corresponding to the target wind direction and including causality.
- the number of target wind directions is plural, the number of Bayesian networks corresponding to the target wind directions is also plural.
- each Bayesian network corresponding to the respective target wind direction is a Directed Acyclic Graph (DAG), which consists of nodes and directed edges connecting these nodes.
- Nodes represent random variables, and the directed edges between nodes represent the mutual relationship between nodes (from the parent node to its child nodes), in which the conditional probability is used to express the strength of the relationship, and the prior probability is used to express the absence of a parent node.
- the random variables of the Bayesian network for each target wind direction include sensors and potential pollution sources. Each arrow indicates that the pollutants emitted by the potential pollution source can be monitored by the sensor when the target wind direction occurs, which is a causal relationship modeling.
- FIG. 3 is an exemplary schematic diagram of a Bayesian network corresponding to a target wind direction according to an embodiment of the present invention.
- the N arrow points in the north direction; the E arrow points in the east direction.
- the Bayesian network 31 corresponding to the northeasterly wind includes node A, node B, node D, node X, and node Y.
- node A represents potential pollution source A
- node B represents potential pollution source B
- node C represents potential pollution source C
- node D represents potential pollution source D
- node X represents sensor X
- node Y represents sensor Y.
- Node A and Node B point to Sensor X, respectively, and Node B and Node D point to Sensor Y, respectively.
- the Bayesian network 32 corresponding to the northwesterly wind includes node A, node B, node C, and node Y.
- node A represents potential pollution source A
- node B represents potential pollution source B
- node C represents potential pollution source C
- node Y represents sensor Y.
- Node A, Node B, and Node C point to Sensor Y, respectively.
- the Bayesian network 33 corresponding to the southwest wind includes node A, node B, node C, node D, and node Z.
- node A represents potential pollution source A
- node B represents potential pollution source B
- node C represents potential pollution source C
- node D represents potential pollution source D
- node Z represents sensor Z.
- Node A, Node B, Node C, and Node D point to sensor Z, respectively.
- the Bayesian network 34 corresponding to the southeasterly wind includes Node A, Node C, Node D, Node X, and Node Z.
- node A represents potential pollution source A
- node C represents potential pollution source C
- node D represents potential pollution source D
- node X represents sensor X
- node Z represents sensor Z.
- Node A, Node C, and Node D point to Sensor X, respectively, and Node D also points to Sensor Z.
- Step 103 Train the Bayesian network based on the sensor historical data when the wind direction is the target wind direction.
- the historical data of the sensor when the wind direction is the target wind direction can be used to train the Bayesian network corresponding to the target wind direction.
- its historical data is used to train the sensor of the Bayesian network corresponding to the target wind direction, and belongs to the Bayesian network corresponding to the target wind direction.
- a Bayesian network 31 corresponding to a northeasterly wind is trained using historical sensor data during a northeasterly wind.
- the Bayesian network 31 corresponding to the northeasterly wind contains sensor X and sensor Y. Therefore, the Bayesian network 31 corresponding to the northeast wind is trained using the historical data of the sensor X when the wind direction is the northeast wind and the historical data of the sensor Y when the wind direction is the northeast wind.
- a Bayesian network 32 for northwesterly winds is trained using historical sensor data during northwesterly winds.
- the Bayesian network 32 corresponding to the northwesterly wind includes sensor Y. Therefore, the Bayesian network 32 corresponding to the northwesterly wind is trained using the historical data of the sensor Y when the wind direction is northwesterly.
- a Bayesian network 33 for southwesterly winds is trained using historical sensor data during southwesterly winds.
- the Bayesian network 33 corresponding to the southwesterly wind includes sensor Z. Therefore, the Bayesian network 33 corresponding to the southwest wind is trained using the historical data of the sensor Z when the wind direction is southwesterly.
- a Bayesian network 34 for southeasterly winds is trained using historical sensor data during southeasterly winds.
- the Bayesian network 34 corresponding to the southeasterly wind includes sensor X and sensor Z. Therefore, the Bayesian network 34 corresponding to the southeast wind is trained using the historical data of sensor X when the wind direction is southeasterly and the historical data of sensor Z when the wind direction is southeasterly.
- a Bayesian network corresponding to the target wind direction may be trained based on the actual historical data of the sensor when the wind direction is the target wind direction.
- the actual historical data (monitoring value) of the sensor in the history when the wind direction is the target wind direction can be obtained through the historical database of the sensor, and the actual historical data of the sensor when the wind direction is the target wind direction can be used to train corresponding to the target wind direction.
- Bayesian network for target wind direction For example, the Bayesian network 31 corresponding to the northeast wind is trained using the actual historical data of sensor X and sensor Y contained in the Bayesian network 31 corresponding to the northeast wind.
- a Bayesian network corresponding to the target wind direction may be trained based on the sensor simulation historical data at the target wind direction.
- a sensor simulation model is established in advance, and the sensor simulation history data of the target wind direction is obtained by using the sensor simulation model.
- a Bayesian network corresponding to the target wind direction is trained using the sensor simulation historical data of the target wind direction.
- the Bayesian network 31 corresponding to the northeast wind is trained using the simulated historical data of the sensor X and the sensor Y included in the Bayesian network 31 corresponding to the northeast wind.
- training the Bayesian network based on the sensor historical data when the wind direction is the target wind direction includes: determining the prior probability distribution of each Bayesian network corresponding to the target wind direction respectively; based on the Bayesian network corresponding to the target wind direction The prior probability distribution of the Bayesian network and the historical sensor data when the wind direction is the target wind direction are used to calculate the posterior probability distribution of the Bayesian network corresponding to the target wind direction by the belief propagation algorithm.
- the belief propagation algorithm also known as the sum-product information passing algorithm, is a message passing algorithm for inference on graphical models, which can be used in Bayesian networks and Markov random fields.
- the following describes the process of calculating the prior probability distribution of the Bayesian network by taking the topology shown in FIG. 2 as an example.
- potential pollution source A, potential pollution source B, potential pollution source C and potential pollution source D obey the uniform distribution U(0,m) respectively, where m can be set as the historical wind direction of sensor X, sensor Y and sensor Z the detected maximum value.
- m max ⁇ y 1 , y 2 , y 3 , ... y T ⁇ ; wherein y 1 y 2 , y 3 , ... y T is the detection value of sensor Y at each time point;
- the potential pollution source A obeys a uniform distribution, then the probability density function f(a) of the potential pollution source A is:
- f(a) 0, a ⁇ m or a ⁇ 0;
- the posterior probability distribution of the Bayesian network can be calculated using a belief propagation algorithm using historical sensor data corresponding to the wind direction.
- calculating the posterior probability distribution of the Bayesian network with the belief propagation algorithm includes: in each step of the iterative calculation of the belief propagation algorithm, taking the wind speed when the wind direction is the target wind direction as the influence corresponding to the target wind direction.
- the parameters of the Bayesian network in which the sensors monitor the values include: The larger the wind speed, the smaller the sensor monitoring value. Correspondingly, the smaller the wind speed, the larger the sensor monitoring value.
- the belief propagation algorithm is as follows:
- Step 104 Determine the pollution source from the potential pollution sources based on the current sensor data and the trained Bayesian network when the wind direction is the target wind direction.
- the current sensor data when the wind direction is the target wind direction are input into the trained Bayesian network corresponding to the target wind direction, so that the Bayesian network determines the pollution source from the potential pollution sources.
- the step 104 determines the pollution source from the potential pollution sources based on the current sensor data when the wind direction is the target wind direction and the trained Bayesian network, specifically including: inputting the current sensor data when the wind direction is the target wind direction The trained Bayesian network; based on the posterior probability of the trained Bayesian network and the current data of the sensor when the wind direction is the target wind direction, calculate the emission probability distribution of each potential pollution source; the emission probability distribution for each potential pollution source The emission value corresponding to the highest probability point is sorted; the potential pollution source corresponding to the maximum emission value is determined as the pollution source.
- the current data of the sensor in the northwest wind (for example, the current data of the sensor Y in Fig. 3 in the northwest wind) is input into the trained shell corresponding to the northwest wind.
- Yessian network calculates the emission probability distribution for each potential pollution source based on the posterior probability of the trained Bayesian network corresponding to the northwesterly wind and the sensor current data in the northwesterly wind; the highest probability for each emission probability distribution
- the emission values corresponding to the points are sorted; the potential pollution source corresponding to the maximum emission value is determined as the pollution source.
- the current data of the sensor during the northeast wind (for example, the current data of the sensor X and the current data of the sensor Y in FIG. Wind, trained Bayesian network; based on the posterior probabilities of the trained Bayesian network corresponding to northeasterly winds and sensor current data when northeasterly winds, calculate the emission probability distribution for each potential pollution source; The emission values corresponding to the highest probability points of each emission probability distribution are sorted; the potential pollution source corresponding to the maximum emission value is determined as the pollution source.
- FIG. 4 is an exemplary schematic diagram of determining a pollution source according to an embodiment of the present invention.
- the abscissa is the emission value, the ordinate is the probability, and the current wind direction is the northwesterly wind.
- the curve 41 is the emission probability distribution curve of the potential pollution source A; the curve 42 is the emission probability distribution curve of the potential pollution source B; the curve 43 is the emission probability distribution curve of the potential pollution source C.
- the emission value corresponding to the highest probability point of curve 41 (that is, the emission value of potential pollution source A) is MA
- the emission value corresponding to the highest probability point of curve 42 (that is, the emission value of potential pollution source B) is MB
- the probability of curve 43 The emission value corresponding to the highest point (ie, the emission value of potential pollution source C) is MC. It can be seen that MC is greater than MB, and MB is greater than MA, so the pollution source is determined to be the potential pollution source C.
- FIG. 4 a specific example of determining the pollution source is described by taking the current wind direction as the northwest wind as an example. Those skilled in the art may realize that, for other wind directions, the pollution source may also be determined in a similar manner, which will not be repeated in the embodiment of the present invention.
- FIG. 5 is an exemplary flowchart of a method for determining a pollution source in an industrial park according to an embodiment of the present invention.
- the method includes:
- Step 501 Based on the positional relationship between the sensor and the potential pollution source in the industrial park, determine the causal relationship between the sensor and the potential pollution source in the predetermined target wind direction.
- the determined causal relationships include: (1) When the wind direction is northwesterly, the pollution emitted by potential pollution source A, potential pollution source B and potential pollution source C object, which will be monitored by sensor Y. (2) When the wind direction is northeasterly, the pollutants emitted by potential pollution source A will be monitored by sensor X; the pollutants emitted by potential pollution source B will be monitored by sensor X and sensor Y; the pollutants emitted by potential pollution source D will be monitored by sensor X and sensor Y. object, which will be monitored by sensor Y.
- Step 502 Establish a Bayesian network corresponding to the target wind direction and including the causal relationship.
- a Bayesian network 31 corresponding to the northeast wind, a Bayesian network 32 corresponding to the northwest wind, a Bayesian network 33 corresponding to the southwest wind, and a Bayesian network 34 corresponding to the southeast wind are established. in:
- the Bayesian network 32 corresponding to the northwesterly wind includes node A, node B, node C, and node Y.
- Node A represents potential pollution source A
- node B represents potential pollution source B
- node C represents potential pollution source C
- node Y represents sensor Y.
- Node A, Node B, and Node C point to Sensor Y, respectively.
- the Bayesian network 31 corresponding to the northeasterly wind includes node A, node B, node D, node X, and node Y.
- Node A represents potential pollution source A
- node B represents potential pollution source B
- node C represents potential pollution source C
- node D represents potential pollution source D
- node X represents sensor X
- node Y represents sensor Y.
- Node A and Node B point to Sensor X, respectively, and Node B and Node D point to Sensor Y, respectively.
- the Bayesian network 33 corresponding to the southwest wind includes node A, node B, node C, node D, and node Z.
- Node A represents potential pollution source A
- node B represents potential pollution source B
- node C represents potential pollution source C
- node D represents potential pollution source D
- node Z represents sensor Z.
- Node A, Node B, Node C, and Node D point to sensor Z, respectively.
- the Bayesian network 34 corresponding to the southeasterly wind includes Node A, Node C, Node D, Node X, and Node Z.
- Node A represents potential pollution source A
- node C represents potential pollution source C
- node D represents potential pollution source D
- node X represents sensor X
- node Z represents sensor Z.
- Node A, Node C, and Node D point to Sensor X, respectively, and Node D also points to Sensor Z.
- Step 503 Acquire the actual historical data of the sensors in the respective target wind directions.
- the actual historical data of the sensor in the northeast wind the actual historical data of the sensor in the northwest wind, the actual historical data of the sensor in the southeast wind, and the actual historical data of the sensor in the southwest wind are respectively obtained.
- Step 504 Train a Bayesian network corresponding to each wind direction by using the actual historical data of the sensor in each wind direction.
- the Bayesian network 31 corresponding to the northeasterly wind is trained using the actual historical data of the sensor during the northeasterly wind.
- the prior probability distribution of the Bayesian network 31 corresponding to the northeast wind is first determined; then based on the prior probability distribution of the Bayesian network 31 corresponding to the northeast wind and the actual historical data of the sensor when the northeast wind
- the propagation algorithm calculates the posterior probability distribution of the Bayesian network 31 corresponding to the northeasterly wind. More preferably, in the iterative calculation of each step of the belief propagation algorithm, the wind speed of the northeasterly wind is used as a parameter that affects the monitoring values of the sensor X and the sensor Y in the Bayesian network 31 . Among them: the greater the wind speed, the smaller the monitoring value of sensor X and the monitoring value of sensor Y; the smaller the wind speed, the greater the monitoring value of sensor X and the monitoring value of sensor Y.
- the Bayesian network 32 corresponding to the northwesterly wind is trained using the actual historical data of the sensor during the northwesterly wind.
- the prior probability distribution of the Bayesian network 32 corresponding to the northwest wind is first determined; then based on the prior probability distribution of the Bayesian network 32 corresponding to the northwest wind and the actual historical data of the sensor when the northwest wind
- the propagation algorithm computes the posterior probability distribution of the Bayesian network 32 corresponding to the northwesterly wind. More preferably, in each step of the iterative calculation of the belief propagation algorithm, the wind speed of the northwesterly wind is used as a parameter that affects the monitoring value of the sensor Y in the Bayesian network 32 . Among them: the greater the wind speed, the smaller the monitoring value of the sensor Y; the smaller the wind speed, the greater the monitoring value of the sensor Y.
- a Bayesian network 33 for southwesterly winds is trained using actual historical data from sensors during southwesterly winds.
- the prior probability distribution of the Bayesian network 33 corresponding to the southwest wind is first determined; then based on the prior probability distribution of the Bayesian network 33 corresponding to the southwest wind and the actual historical data of the sensor when the southwest wind
- the propagation algorithm computes the posterior probability distribution of the Bayesian network 33 corresponding to the southwesterly wind. More preferably, in each step of the iterative calculation of the belief propagation algorithm, the wind speed of the southwest wind is used as a parameter that affects the monitoring value of the sensor Z in the Bayesian network 33 . Among them: the greater the wind speed, the smaller the monitoring value of sensor Z; the smaller the wind speed, the greater the monitoring value of sensor Z.
- a Bayesian network 34 for southeasterly winds is trained using actual historical data from sensors during southeasterly winds.
- the prior probability distribution of the Bayesian network 34 corresponding to the southeast wind is first determined; then based on the prior probability distribution of the Bayesian network 34 corresponding to the southeast wind and the actual historical data of the sensor when the southeast wind
- the propagation algorithm computes the posterior probability distribution of the Bayesian network 34 corresponding to the southeasterly wind. More preferably, in each step of the iterative calculation of the belief propagation algorithm, the wind speed of the southeasterly wind is used as a parameter that affects the monitoring values of the sensor X and the sensor Z in the Bayesian network 34 . Among them: the greater the wind speed, the smaller the monitoring values of sensor X and sensor Z; the smaller the wind speed, the greater the monitoring value of sensor X and sensor Z.
- Step 505 Determine the pollution source from the potential pollution sources based on the current sensor data at the current wind direction and the trained Bayesian network corresponding to the current wind direction.
- the current wind direction is the northwesterly wind
- select the Bayesian network 32 corresponding to the northwesterly wind and then input the current sensor data in the current wind direction (ie, the northwesterly wind) into the Bayesian network 32 corresponding to the northwesterly wind (preferably, The wind speed of the current northwesterly wind is also input to the Bayesian network corresponding to the northwesterly wind 32).
- the Bayesian network 32 corresponding to the northwesterly wind outputs the emission probability distributions of potential pollution source A, potential pollution source B, and potential pollution source C.
- the emission values corresponding to the respective highest probability points of potential pollution source A, potential pollution source B and potential pollution source C are sorted; the potential pollution source corresponding to the maximum emission value is determined as the pollution source.
- the Bayesian network 31 corresponding to the northeast wind selects the Bayesian network 31 corresponding to the northeast wind, and then input the current data of the sensor in the current wind direction (ie, the northeast wind) into the Bayesian network 31 corresponding to the northeast wind (preferably, also input the wind speed of the current northeast wind. to a Bayesian network corresponding to northeasterly winds31).
- the Bayesian network 31 corresponding to the northeasterly wind outputs the emission probability distributions of potential pollution source A, potential pollution source B, and potential pollution source D.
- the emission values corresponding to the highest probability points of each emission probability distribution are sorted; the potential pollution source corresponding to the maximum emission value is determined as the pollution source.
- an embodiment of the present invention further provides an apparatus for determining a pollution source.
- FIG. 6 is an exemplary structural diagram of an apparatus for determining a pollution source according to an embodiment of the present invention.
- the apparatus 600 for determining the pollution source includes:
- a relationship determination module 601 configured to determine a causal relationship between the sensor and the potential pollution source in the downward direction of the target wind based on the positional relationship between the sensor and the potential pollution source;
- a network establishment module 602 configured to establish a Bayesian network corresponding to the target wind direction and including the causal relationship;
- a training module 603, configured to train the Bayesian network based on the historical sensor data when the wind direction is the target wind direction;
- a determination module 604 configured to determine a pollution source from the potential pollution sources based on current sensor data when the wind direction is the target wind direction and the trained Bayesian network.
- the training module 603 is configured to determine the prior probability distribution of the Bayesian network; based on the prior probability distribution of the Bayesian network and the sensor historical data when the wind direction is the target wind direction, The posterior probability distribution of the Bayesian network is calculated with a belief propagation algorithm.
- the training module 603 is configured to use the wind speed when the wind direction is the target wind direction as the parameter affecting the monitoring value of the sensor during each step of the iterative calculation of the belief propagation algorithm, wherein the higher the wind speed, the higher the monitoring value of the sensor. Small.
- the determining module 604 is configured to input the current sensor data when the wind direction is the target wind direction into the trained Bayesian network; based on the posterior probability and wind direction of the Bayesian network Calculate the emission probability distribution of each potential pollution source for the current data of the sensor in the target wind direction; sort the emission value corresponding to the highest probability point of the emission probability distribution of each potential pollution source; The pollution source is determined as the pollution source.
- the training module 603 trains the Bayesian network based on the actual historical data of the sensor when the wind direction is the target wind direction; or trains the Bayesian network based on the simulated historical data of the sensor when the wind direction is the target wind direction s network.
- FIG. 7 is an exemplary structural diagram of an apparatus for determining a pollution source according to an embodiment of the present invention.
- an apparatus 700 for determining a pollution source includes a memory 702 and a processor 701; the memory 702 stores an application program executable by the processor 701, for causing the processor 701 to perform the determination as described in any of the above methods of pollution sources.
- the memory 702 can be specifically implemented as various storage media such as Electrically Erasable Programmable Read-Only Memory (EEPROM), Flash Memory (Flash memory), Programmable Program Read-Only Memory (PROM).
- the processor 701 may be implemented to include one or more central processing units or one or more field programmable gate arrays, wherein the field programmable gate arrays integrate one or more central processing unit cores.
- the central processing unit or the central processing unit core may be implemented as a CPU or an MCU or a DSP or the like.
- the hardware modules in various embodiments may be implemented mechanically or electronically.
- a hardware module may include specially designed permanent circuits or logic devices (eg, special purpose processors, such as FPGAs or ASICs) for performing specific operations.
- Hardware modules may also include programmable logic devices or circuits (eg, including general-purpose processors or other programmable processors) temporarily configured by software for performing particular operations.
- programmable logic devices or circuits eg, including general-purpose processors or other programmable processors
- the present invention also provides a machine-readable storage medium storing instructions for causing a machine to perform a method as described herein.
- a system or device equipped with a storage medium on which software program codes for realizing the functions of any one of the above-described embodiments are stored, and make the computer (or CPU or MPU of the system or device) ) to read and execute the program code stored in the storage medium.
- a part or all of the actual operation can also be completed by an operating system or the like operating on the computer based on the instructions of the program code.
- the program code read out from the storage medium can also be written into the memory provided in the expansion board inserted into the computer or into the memory provided in the expansion unit connected to the computer, and then the instructions based on the program code cause the device to be installed in the computer.
- the CPU on the expansion board or the expansion unit or the like performs part and all of the actual operations, so as to realize the functions of any one of the above-mentioned embodiments.
- Embodiments of storage media for providing program code include floppy disks, hard disks, magneto-optical disks, optical disks (eg, CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), Magnetic tapes, non-volatile memory cards and ROMs.
- the program code may be downloaded from a server computer or cloud over a communications network.
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Indicating Or Recording The Presence, Absence, Or Direction Of Movement (AREA)
Abstract
一种确定污染源的方法(100)、装置(600,700)和计算机可读存储介质,涉及污染监测技术领域。其中的方法(100)包括:基于传感器与潜在污染源的位置关系,确定目标风向下的、传感器与潜在污染源之间的因果关系(101);建立对应于目标风向的、包含因果关系的贝叶斯网络(102);基于风向为目标风向时的传感器历史数据训练贝叶斯网络(103);基于风向为目标风向时的传感器当前数据和已训练的贝叶斯网络,从潜在污染源中确定出污染源(104)。由于考虑到风向对传感器监测效果的影响,因此确定出的污染源的准确度高。
Description
本发明涉及污染监测技术领域,尤其涉及确定污染源的方法、装置和计算机可读存储介质。
工业园区为划定一定范围的土地,并予以规划以供工业设施设置使用的地区。工业园区通常对污染物(液体或气体)的排放有严格的限制规定。越来越多的组织意识到生态友好型工业园区(Eco-Industrial Park,EIP)对于可持续发展的重要性。
目前,通常在工业园区的固定位置处布置监控点以监测污染物排放情况,并基于监测结果排查污染源。
然而,工业园区内的风向可能导致监控点的监测结果产生偏差,导致污染源排查结果不准确。
发明内容
本发明实施方式提出确定污染源的方法、装置和计算机可读存储介质。
第一方面,提供确定污染源的方法,包括:
基于传感器与潜在污染源的位置关系,确定目标风向下的、传感器与潜在污染源之间的因果关系;
建立对应于所述目标风向的、包含所述因果关系的贝叶斯网络;
基于风向为所述目标风向时的传感器历史数据训练所述贝叶斯网络;
基于风向为所述目标风向时的传感器当前数据和已训练的所述贝叶斯网络,从所述潜在污染源中确定出污染源。
第二方面,提供确定污染源的装置,包括:
关系确定模块,用于基于传感器与潜在污染源的位置关系,确定目标风向下的、传感器与潜在污染源之间的因果关系;
网络建立模块,用于建立对应于所述目标风向的、包含所述因果关系的贝叶斯网络;
训练模块,用于基于风向为所述目标风向时的传感器历史数据训练所述贝叶斯网络;
确定模块,用于基于风向为所述目标风向时的传感器当前数据和已训练的所述贝叶斯网 络,从所述潜在污染源中确定出污染源。
第三方面,提供确定污染源的装置,包括处理器和存储器;
所述存储器中存储有可被所述处理器执行的应用程序,用于使得所述处理器执行如上任一项所述的确定污染源的方法。
第四方面,提供计算机可读存储介质,其中存储有计算机可读指令,该计算机可读指令用于执行如上任一项所述的确定污染源的方法。
可见,本发明实施方式建立对应于不同目标风向的各自贝叶斯网络,然后利用对应于当前目标风向的贝叶斯网络准确确定污染源,因此本发明实施方式考虑到风向对传感器监测效果的影响,确定出的污染源的准确度高。
对于上述任一方面,优选地,还包括:确定所述贝叶斯网络的先验概率分布;基于所述贝叶斯网络的先验概率分布和风向为目标风向时的传感器历史数据,以置信传播算法计算所述贝叶斯网络的后验概率分布。
因此,本发明实施方式通过置信传播算法计算贝叶斯网络的后验概率分布,可以快速训练贝叶斯网络。
对于上述任一方面,优选地,所述以置信传播算法计算所述贝叶斯网络的后验概率分布包括:在所述置信传播算法的每步迭代计算时,将风向为所述目标风向时的风速作为影响传感器监测值的参数,其中所述风速越大,所述传感器监测值越小。
可见,本发明实施方式还考虑到风速对传感器监测效果的影响,将风速作为参数参与到贝叶斯网络的训练过程中,从而训练出的贝叶斯网络更加准确。
对于上述任一方面,优选地,所述基于风向为目标风向时的传感器当前数据和已训练的所述贝叶斯网络,从所述潜在污染源中确定出污染源包括:
将风向为目标风向时的传感器当前数据输入所述已训练的贝叶斯网络;
基于所述已训练的贝叶斯网络的后验概率和所述风向为目标风向时的传感器当前数据,计算每个潜在污染源的排放概率分布;
对每个潜在污染源的排放概率分布的概率最高点所对应的排放值进行排序;
将最大排放值所对应的潜在污染源确定为所述污染源。
因此,本发明实施方式计算每个潜在污染源的排放概率分布,将最大排放值所对应的潜在污染源确定为污染源,从而实现了准确标识污染源。
对于上述任一方面,优选地,所述基于所述风向为目标风向时的传感器历史数据训练所述贝叶斯网络包括:
基于风向为目标风向时的传感器实际历史数据训练所述贝叶斯网络;或
基于风向为目标风向时的传感器仿真历史数据训练所述贝叶斯网络。
可见,本发明实施方式既可以利用传感器的实际历史数据训练贝叶斯网络,还可以利用传感器的仿真历史数据训练贝叶斯网络,从而丰富了训练数据。
图1为本发明实施方式的确定污染源的方法的示范性流程图。
图2为本发明实施方式的工业园区拓扑图的示范性示意图。
图3为本发明实施方式对应于风向的贝叶斯网络的示范性示意图。
图4为本发明实施方式确定污染源的示范性示意图。
图5为本发明实施方式在工业园区中确定污染源的方法的示范性流程图。
图6为本发明实施方式的确定污染源的装置的示范性结构图。
图7为本发明实施方式的确定污染源的装置的示范性结构图。
其中,附图标记如下:
标号 | 含义 |
100 | 确定污染源的方法 |
101~104 | 步骤 |
10 | 工业园区 |
11、12、13 | 建筑物 |
31 | 对应于东北风的贝叶斯网络 |
32 | 对应于西北风的贝叶斯网络 |
33 | 对应于西南风的贝叶斯网络 |
34 | 对应于东南风的贝叶斯网络 |
41 | 潜在污染源A的排放概率分布曲线 |
42 | 潜在污染源B的排放概率分布曲线 |
43 | 潜在污染源C的排放概率分布曲线 |
500 | 在工业园区中确定污染源的方法 |
501~505 | 步骤 |
600 | 确定污染源的装置 |
601 | 关系确定模块 |
602 | 网络确定模块 |
603 | 训练模块 |
604 | 确定模块 |
700 | 确定污染源的装置 |
701 | 处理器 |
702 | 存储器 |
为了使本发明的技术方案及优点更加清楚明白,以下结合附图及实施方式,对本发明进行进一步详细说明。应当理解,此处所描述的具体实施方式仅仅用以阐述性说明本发明,并不被配置为用于限定本发明的保护范围。
为了描述上的简洁和直观,下文通过描述若干代表性的实施方式来对本发明的方案进行阐述。实施方式中大量的细节仅被配置为用于帮助理解本发明的方案。但是很明显,本发明的技术方案实现时可以不局限于这些细节。为了避免不必要地模糊了本发明的方案,一些实施方式没有进行细致地描述,而是仅给出了框架。下文中,“包括”是指“包括但不限于”,“根据……”是指“至少根据……,但不限于仅根据……”。由于汉语的语言习惯,下文中没有特别指出一个成分的数量时,意味着该成分可以是一个也可以是多个,或可理解为至少一个。
申请人发现:现有技术忽视了风向对传感器监测效果的影响,导致确定出的污染源的准确度不高。
为解决该技术问题,本发明实施方式考虑到风向对传感器监测效果的影响,预先建立对应于不同风向的、用于确定出污染源的贝叶斯网络,并利用各自风向下的训练数据训练各自风向的贝叶斯网络,再利用当前风向时的传感器当前数据以及对应于当前风向的、已训练的贝叶斯网络准确确定污染源。
进一步地,本发明实施方式还考虑到风速对传感器监测效果的影响(比如,风速较强时,监测到的污染物浓度较低),将风速作为参数参与到贝叶斯网络的训练过程中,训练出的贝叶斯网络更加准确。
图1为本发明实施方式的确定污染源的方法的示范性流程图。
如图1所示,该方法100包括:
步骤101:基于传感器与潜在污染源的位置关系,确定目标风向下的、传感器与潜在污染源之间的因果关系。
在这里,传感器的数目可以为一或多个,潜在污染源的数目也可以为一或多个。传感器适配于对潜在污染源排放出的污染物的物理量(比如,浓度)进行监测。而且,传感器的部署位置固定,潜在污染源的部署位置也固定。当潜在污染源排放的污染物的种类为多种时,传感器为适配于检测各自污染物类型的多种传感器。
申请人发现:当风向与传感器的检测方向一致时,传感器较容易监测到污染物,当风向与传感器的检测方向相反时,传感器较难监测到污染物。因此,可以首先确定若干个预定的目标风向,再基于传感器与潜在污染源的位置关系,确定出各自目标风向下的、传感器与潜在污染源之间的因果关系。
以工业园区为例进行说明。图2为本发明实施方式的工业园区拓扑图的示范性示意图。
在图2中,工业园区10内存在有建筑物11、建筑物12和建筑物13。其中,在建筑物11内具有潜在污染源B,建筑物12内具有潜在污染源D,建筑物13内具有潜在污染源A和潜在污染源C。潜在污染源A、潜在污染源B、潜在污染源C和潜在污染源D适配于排放同种类型的污染物,比如二氧化硫气体。箭头N所示方向为朝北方向。
另外,在工业园区10的西边边缘处布置有传感器X,在工业园区10的南边边缘处布置有传感器Y,在工业园区10的东边边缘处布置有传感器Z。传感器X、传感器Y和传感器Z分别适配于监测潜在污染源A、潜在污染源B、潜在污染源C和潜在污染源D所排放的污染物。优选地,传感器X、传感器Y和传感器Z分别设置有浓度门限值,当监测到的污染物超过该浓度门限值时,认定监测到污染物,并记录浓度门限值。
基于图2所示的工业园,可以确定出目标风向下传感器与潜在污染源之间的因果关系。其中,基于风向角度的不同划分情形,目标风向的数目可以为多个。比如,当以90度为风向角度进行划分时,可以划分为四个目标风向:西北风、东北风、西南风和东南风。风向角度越小,划分出的目标风向的数目越多。
下面以四个目标风向:西北风、东北风、西南风和东南风为例进行说明。此时,传感器X、传感器Y和传感器Z与潜在污染源A、潜在污染源B、潜在污染源C和潜在污染源D之间的因果关系包括:
(1)、当目标风向为西北风(即风从西北方向吹来)时,潜在污染源A、潜在污染源B和潜在污染源C排放的污染物,将被传感器Y监测到。
(2)、当目标风向为东北风(即风从东北方向吹来)时,潜在污染源A排放的污染物, 将被传感器X监测到;潜在污染源B排放的污染物,将被传感器X和传感器Y监测到;潜在污染源D排放的污染物,将被传感器Y监测到。
(3)、当目标风向为西南风(即风从西南方向吹来)时,潜在污染源A排放的污染物,将被传感器Z监测到;潜在污染源B排放的污染物,将被传感器Z监测到;潜在污染源C排放的污染物,将被传感器Z监测到;潜在污染源D排放的污染物,将被传感器Z监测到。
(4)、当目标风向为东南风(即风从东南方向吹来)时,潜在污染源A排放的污染物,将被传感器X监测到;潜在污染源C排放的污染物,将被传感器X监测到;潜在污染源D排放的污染物,将被传感器X和传感器Z监测到。
以上以四个目标风向为例,对传感器与潜在污染源之间的因果关系进行示范性描述,本领域技术人员可以意识到,这种描述仅是示范性的,并不用于限定本发明实施方式的保护范围。实际上,基于风向的不同划分以及传感器与潜在污染源的位置关系,传感器与潜在污染源之间的因果关系还可以具有多种实施方式。
步骤102:建立对应于目标风向的、包含因果关系的贝叶斯网络。
在这里,当目标风向的数目为多个时,对应于目标风向的贝叶斯网络也为多个。
具体地,对应于各自的目标风向的每个贝叶斯网络分别是一个有向无环图(Directed Acyclic Graph,DAG),由节点及连接这些节点的有向边构成。节点代表随机变量,结点间的有向边代表结点间的互相关系(由父结点指向其子结点),其中用条件概率表达关系强度,没有父结点的用先验概率进行表达。每个目标风向的贝叶斯网络的随机变量包括传感器和潜在污染源。其中每个箭头分别表示该潜在污染源排放的污染物可以被该传感器在该目标风向发生时监测到,是一种因果关系建模。
针对图2所示的工业园区,图3为本发明实施方式对应于目标风向的贝叶斯网络的示范性示意图。在图3中,N箭头指向北边方向;E箭头指向东边方向。
在图3中:
对应于东北风的贝叶斯网络31包括节点A、节点B、节点D、节点X和节点Y。其中:节点A表示潜在污染源A、节点B表示潜在污染源B、节点C表示潜在污染源C,节点D表示潜在污染源D,节点X表示传感器X,节点Y表示传感器Y。节点A和节点B分别指向传感器X,节点B和节点D分别指向传感器Y。
对应于西北风的贝叶斯网络32包括节点A、节点B、节点C和节点Y。其中:节点A表示潜在污染源A、节点B表示潜在污染源B、节点C表示潜在污染源C,节点Y表示传感器Y。节点A、节点B和节点C分别指向传感器Y。
对应于西南风的贝叶斯网络33包括节点A、节点B、节点C、节点D和节点Z。其中:节点A表示潜在污染源A、节点B表示潜在污染源B、节点C表示潜在污染源C,节点D表示潜在污染源D,节点Z表示传感器Z。节点A、节点B、节点C和节点D分别指向传感器Z。
对应于东南风的贝叶斯网络34包括节点A、节点C、节点D、节点X和节点Z。其中:节点A表示潜在污染源A、节点C表示潜在污染源C,节点D表示潜在污染源D,节点X表示传感器X,节点Z表示传感器Z。节点A、节点C和节点D分别指向传感器X,节点D还指向传感器Z。
以上以四个目标风向为例,对各个目标风向的贝叶斯网络进行示范性描述,本领域技术人员可以意识到,这种描述仅是示范性的,并不用于限定本发明实施方式的保护范围。
步骤103:基于风向为所述目标风向时的传感器历史数据训练所述贝叶斯网络。
在这里,可以利用风向为目标风向时的传感器历史数据分别训练对应于该目标风向的贝叶斯网络。其中:其历史数据用于训练对应于目标风向的贝叶斯网络的传感器,属于对应于该目标风向的贝叶斯网络。
比如,利用东北风时的传感器历史数据训练对应于东北风的贝叶斯网络31。对应于东北风的贝叶斯网络31包含传感器X和传感器Y。因此,利用风向为东北风时的、传感器X的历史数据和风向为东北风时的、传感器Y的历史数据训练对应于东北风的贝叶斯网络31。
利用西北风时的传感器历史数据训练对应于西北风的贝叶斯网络32。对应于西北风的贝叶斯网络32包括传感器Y。因此,利用风向为西北风时的、传感器Y的历史数据训练对应于西北风的贝叶斯网络32。
利用西南风时的传感器历史数据训练对应于西南风的贝叶斯网络33。对应于西南风的贝叶斯网络33包括传感器Z。因此,利用风向为西南风时的、传感器Z的历史数据训练对应于西南风的贝叶斯网络33。
利用东南风时的传感器历史数据训练对应于东南风的贝叶斯网络34。对应于东南风的贝叶斯网络34包括传感器X和传感器Z。因此,利用风向为东南风时的、传感器X的历史数据和风向为东南风时的、传感器Z历史数据训练对应于东南风的贝叶斯网络34。
在一个实施方式中,可以基于风向为目标风向时的传感器实际历史数据训练对应于该目标风向的贝叶斯网络。其中,可以通过传感器的历史数据库获取历史上的、风向为该目标风向时传感器实际历史数据(监测值),并利用该历史上的、风向为该目标风向时的传感器实际历史数据训练对应于该目标风向的贝叶斯网络。比如,利用对应于东北风的贝叶斯网络31中 所包含的传感器X和传感器Y的实际历史数据训练对应于东北风的贝叶斯网络31。
在一个实施方式中,可以基于目标风向时的传感器仿真历史数据训练对应于该目标风向的贝叶斯网络。在这里,预先建立传感器仿真模型,并利用传感器仿真模型获取该目标风向时的传感器仿真历史数据。然后,利用该目标风向的传感器仿真历史数据训练对应于该目标风向的贝叶斯网络。比如,利用对应于东北风的贝叶斯网络31中所包含的传感器X和传感器Y的仿真历史数据训练对应于东北风的贝叶斯网络31。
在一个实施方式中,基于风向为目标风向时的传感器历史数据训练贝叶斯网络包括:分别确定每个对应于目标风向的贝叶斯网络的先验概率分布;基于对应于目标风向的贝叶斯网络的先验概率分布和风向为目标风向时的传感器历史数据,以置信传播算法计算对应于目标风向的贝叶斯网络的后验概率分布。
置信传播算法,又名和-积信息传递算法,是一种在图模型上进行推断的消息传递算法,可用在贝叶斯网络和马尔科夫随机域中。
下面以图2所示拓扑结构为例,描述计算贝叶斯网络的先验概率分布的过程。
(1)、假设潜在污染源A、潜在污染源B、潜在污染源C和潜在污染源D分别服从均匀分布U(0,m),其中m可设定为传感器X、传感器Y和传感器Z在历史上该风向所检测到的最大值。
以潜在污染源A为例,在东南风时:
m=max{y
1,y
2,y
3,...y
T};其中y
1y
2,y
3,...y
T为传感器Y在各个时间点的检测值;
潜在污染源A服从均匀分布,则潜在污染源A的概率密度函数f(a)为:
f(a)=0,a≥m或者a≤0;
(2)、对于传感器X、传感器Y和传感器Z,假设其分别服从离散正态分布。以传感器X为例,满足X~N(μ
x,σ
x),μ
x为其历史上该风向的均值;σ
x为其历史上在该风向监测到的方差:
其中f(x)为传感器X的概率密度函数。
在基于上述描述确定贝叶斯网络的先验概率分布后,可以利用对应于风向的传感器历史数据以置信传播算法计算贝叶斯网络的后验概率分布。在一个实施方式中,以置信传播算法计算所述贝叶斯网络的后验概率分布包括:在置信传播算法的每步迭代计算时,将风向为目标风向时的风速作为影响对应于该目标风向的贝叶斯网络中的传感器的监测值的参数。其中: 风速越大,传感器监测值越小。相应地,风速越小,传感器监测值越大。
示范性地,在东南风时,置信传播算法如下:
输入:随机变量A、B、C、Y以及各自的先验概率f(a)、f(b)、f(c)和f(y),收敛阈值ε,风速s;
定义:节点N∈{A,B,C,Y};
初始化:
(1)、t=0时,初始信息等于先验概率:m(n)=f(n);
(2)、定义初始信息队列M={m(a),m(b),m(c),m(y)};
(3)、从N中选择一个初始节点N1开始传递信息。
开始循环迭代过程(其中,当t={2,3,4,…}时),迭代过程包括:(1)、节点N1传递消息到节点N2(N1与N2之间需要有箭头直接链接,比如A->Y,Y->B,Y->C);(2)、计算N1,N2的联合概率分布p(n1,n2),其中p(n1,n2)=m(n1)*m(n2);(3)、更新消息
如果收敛:Belief
(t)(n2)–Belief
(t-1)(n2)<=ε,则将节点N2移除出信息队列M,当信息队列变为空集
时,返回后验概率分布Pr(N),其概率密度函数为p(n)=Belief
(t)(n)。
以上以东南风为例,描述了置信传播算法的实施实例,本领域技术人员可以意识到,这种描述仅是示范性的,并不用于限定本发明实施方式的保护范围。
步骤104:基于风向为目标风向时的传感器当前数据和已训练的贝叶斯网络,从潜在污染源中确定出污染源。
在这里,将风向为目标风向时的传感器当前数据输入到对应于该目标风向的、已训练的贝叶斯网络,从而由该贝叶斯网络从潜在污染源中确定出污染源。
在一个实施方式中,步骤104的基于风向为目标风向时的传感器当前数据和已训练的贝叶斯网络,从潜在污染源中确定出污染源,具体包括:将风向为目标风向时的传感器当前数据输入已训练的贝叶斯网络;基于已训练的贝叶斯网络的后验概率和风向为目标风向时的传感器当前数据,计算每个潜在污染源的排放概率分布;对每个潜在污染源的排放概率分布的概率最高点所对应的排放值进行排序;将最大排放值所对应的潜在污染源确定为所述污染源。
比如,当确定当前风向为西北风时,将西北风时的传感器的当前数据(比如,西北风时的、图3中的传感器Y的当前数据)输入到对应于西北风的、已训练的贝叶斯网络;基于对应于西北风的、已训练的贝叶斯网络的后验概率和西北风时的传感器当前数据,计算每个潜 在污染源的排放概率分布;对每个排放概率分布的概率最高点所对应的排放值进行排序;将最大排放值所对应的潜在污染源确定为污染源。
再比如,当确定当前风向为东北风时,将东北风时的传感器的当前数据(比如,东北风时的、图3中的传感器X的当前数据和传感器Y的当前数据)输入到对应于东北风的、已训练的贝叶斯网络;基于对应于东北风的、已训练的贝叶斯网络的后验概率和东北风时的传感器当前数据,计算每个潜在污染源的排放概率分布;对每个排放概率分布的概率最高点所对应的排放值进行排序;将最大排放值所对应的潜在污染源确定为污染源。
图4为本发明实施方式确定污染源的示范性示意图。
在图4中,横坐标为排放值,纵坐标为概率,当前风向为西北风。曲线41为潜在污染源A的排放概率分布曲线;曲线42为潜在污染源B的排放概率分布曲线;曲线43为潜在污染源C的排放概率分布曲线。
曲线41的概率最高点所对应的排放值(即潜在污染源A的排放值)为MA,曲线42的概率最高点所对应的排放值(即潜在污染源B的排放值)为MB,曲线43的概率最高点所对应的排放值(即潜在污染源C的排放值)为MC。可见,MC大于MB,MB大于MA,因此确定出污染源为潜在污染源C。
在图4中,以当前风向为西北风为例描述了确定污染源的具体实例。本领域技术人员可以意识到,对于其他风向,也可以采用类似方式确定出污染源,本发明实施方式对此不再赘述。
针对图2所示的工业园拓扑结构,图5为本发明实施方式在工业园中确定污染源的方法的示范性流程图。
如图5所示,该方法包括:
步骤501:基于工业园区中传感器与潜在污染源的位置关系,确定预定的目标风向下的、传感器与潜在污染源之间的因果关系。
假定预定的目标风向为东北风、西北风、东南风和西南风,则确定的因果关系包括:(1)、当风向为西北风时,潜在污染源A、潜在污染源B和潜在污染源C排放的污染物,将被传感器Y监测到。(2)、当风向为东北风时,潜在污染源A排放的污染物,将被传感器X监测到;潜在污染源B排放的污染物,将被传感器X和传感器Y监测到;潜在污染源D排放的污染物,将被传感器Y监测到。(3)、当风向为西南风时,潜在污染源A排放的污染物,将被传感器Z监测到;潜在污染源B排放的污染物,将被传感器Z监测到;潜在污染源C排放的污染物,将被传感器Z监测到;潜在污染源D排放的污染物,将被传感器Z监测到。(4)、当 风向为东南风时,潜在污染源A排放的污染物,将被传感器X监测到;潜在污染源C排放的污染物,将被传感器X监测到;潜在污染源D排放的污染物,将被传感器X和传感器Z监测到。
步骤502:建立对应于目标风向的、包含所述因果关系的贝叶斯网络。
具体地,建立对应于东北风的贝叶斯网络31、对应于西北风的贝叶斯网络32、对应于西南风的贝叶斯网络33和对应于东南风的贝叶斯网络34。其中:
对应于西北风的贝叶斯网络32包括节点A、节点B、节点C和节点Y。节点A表示潜在污染源A、节点B表示潜在污染源B、节点C表示潜在污染源C,节点Y表示传感器Y。节点A、节点B和节点C分别指向传感器Y。
对应于东北风的贝叶斯网络31包括节点A、节点B、节点D、节点X和节点Y。节点A表示潜在污染源A、节点B表示潜在污染源B、节点C表示潜在污染源C,节点D表示潜在污染源D,节点X表示传感器X,节点Y表示传感器Y。节点A和节点B分别指向传感器X,节点B和节点D分别指向传感器Y。
对应于西南风的贝叶斯网络33包括节点A、节点B、节点C、节点D和节点Z。节点A表示潜在污染源A、节点B表示潜在污染源B、节点C表示潜在污染源C,节点D表示潜在污染源D,节点Z表示传感器Z。节点A、节点B、节点C和节点D分别指向传感器Z。
对应于东南风的贝叶斯网络34包括节点A、节点C、节点D、节点X和节点Z。节点A表示潜在污染源A、节点C表示潜在污染源C,节点D表示潜在污染源D,节点X表示传感器X,节点Z表示传感器Z。节点A、节点C和节点D分别指向传感器X,节点D还指向传感器Z。
步骤503:获取各自目标风向时的传感器实际历史数据。
在这里,分别获取东北风时的传感器实际历史数据、西北风时的传感器实际历史数据、东南风时的传感器实际历史数据和西南风时的传感器实际历史数据。
步骤504:利用各自风向时的传感器实际历史数据,训练对应于各自风向的贝叶斯网络。
在这里,利用东北风时的传感器实际历史数据训练对应于东北风的贝叶斯网络31。优选地,首先确定对应于东北风的贝叶斯网络31的先验概率分布;再基于对应于东北风的贝叶斯网络31的先验概率分布和东北风时的传感器实际历史数据,以置信传播算法计算对应于东北风的贝叶斯网络31的后验概率分布。更优选地,在置信传播算法的每步迭代计算时,将该东北风的风速作为影响贝叶斯网络31中的传感器X与传感器Y的监测值的参数。其中:风速越大,传感器X的监测值与传感器Y的监测值越小;风速越小,传感器X的监测值与传感 器Y的监测值越大。
利用西北风时的传感器实际历史数据训练对应于西北风的贝叶斯网络32。优选地,首先确定对应于西北风的贝叶斯网络32的先验概率分布;再基于对应于西北风的贝叶斯网络32的先验概率分布和西北风时的传感器实际历史数据,以置信传播算法计算对应于西北风的贝叶斯网络32的后验概率分布。更优选地,在置信传播算法的每步迭代计算时,将该西北风的风速作为影响贝叶斯网络32中的传感器Y的监测值的参数。其中:风速越大,传感器Y的监测值越小;风速越小,传感器Y的监测值越大。
利用西南风时的传感器实际历史数据训练对应于西南风的贝叶斯网络33。优选地,首先确定对应于西南风的贝叶斯网络33的先验概率分布;再基于对应于西南风的贝叶斯网络33的先验概率分布和西南风时的传感器实际历史数据,以置信传播算法计算对应于西南风的贝叶斯网络33的后验概率分布。更优选地,在置信传播算法的每步迭代计算时,将该西南风的风速作为影响贝叶斯网络33中的传感器Z的监测值的参数。其中:风速越大,传感器Z的监测值越小;风速越小,传感器Z的监测值越大。
利用东南风时的传感器实际历史数据训练对应于东南风的贝叶斯网络34。优选地,首先确定对应于东南风的贝叶斯网络34的先验概率分布;再基于对应于东南风的贝叶斯网络34的先验概率分布和东南风时的传感器实际历史数据,以置信传播算法计算对应于东南风的贝叶斯网络34的后验概率分布。更优选地,在置信传播算法的每步迭代计算时,将该东南风的风速作为影响贝叶斯网络34中的传感器X和传感器Z的监测值的参数。其中:风速越大,传感器X和传感器Z的监测值越小;风速越小,传感器X和传感器Z的监测值越大。
步骤505:基于当前风向时的传感器当前数据和对应于当前风向的已训练的贝叶斯网络,从潜在污染源中确定出污染源。
比如,假定当前风向为西北风,选择对应于西北风的贝叶斯网络32,然后将当前风向(即西北风)时的传感器当前数据输入对应于西北风的贝叶斯网络32(优选的,还将当前西北风的风速输入到对应于西北风的贝叶斯网络32)。对应于西北风的贝叶斯网络32输出潜在污染源A、潜在污染源B和潜在污染源C的排放概率分布。接着,对潜在污染源A、潜在污染源B和潜在污染源C的各自概率最高点所对应的排放值进行排序;将最大排放值所对应的潜在污染源确定为污染源。
再比如,假定当前风向为东北风。则选择对应于东北风的贝叶斯网络31,然后将当前风向(即东北风)时的传感器当前数据输入对应于东北风的贝叶斯网络31(优选的,还将当前东北风的风速输入到对应于东北风的贝叶斯网络31)。对应于东北风的贝叶斯网络31输出潜 在污染源A、潜在污染源B和潜在污染源D的排放概率分布。接着,对每个排放概率分布的概率最高点所对应的排放值进行排序;将最大排放值所对应的潜在污染源确定为污染源。
基于上述描述,本发明实施方式还提出了一种确定污染源的装置。
图6为本发明实施方式的确定污染源的装置的示范性结构图。
如图6所示,确定污染源的装置600包括:
关系确定模块601,用于基于传感器与潜在污染源的位置关系,确定目标风向下的、传感器与潜在污染源之间的因果关系;
网络建立模块602,用于建立对应于所述目标风向的、包含所述因果关系的贝叶斯网络;
训练模块603,用于基于风向为所述目标风向时的传感器历史数据训练所述贝叶斯网络;
确定模块604,用于基于风向为所述目标风向时的传感器当前数据和已训练的所述贝叶斯网络,从所述潜在污染源中确定出污染源。
在一个实施方式中,训练模块603,用于确定所述贝叶斯网络的先验概率分布;基于所述贝叶斯网络的先验概率分布和风向为所述目标风向时的传感器历史数据,以置信传播算法计算所述贝叶斯网络的后验概率分布。
在一个实施方式中,训练模块603,用于在所述置信传播算法的每步迭代计算时,将风向为目标风向时的风速作为影响传感器监测值的参数,其中风速越大,传感器监测值越小。
在一个实施方式中,确定模块604,用于将风向为所述目标风向时的传感器当前数据输入所述已训练的所述贝叶斯网络;基于所述贝叶斯网络的后验概率和风向为所述目标风向时的传感器当前数据,计算每个潜在污染源的排放概率分布;对每个潜在污染源的排放概率分布的概率最高点所对应的排放值进行排序;将最大排放值所对应的潜在污染源确定为所述污染源。
在一个实施方式中,训练模块603,基于风向为所述目标风向时的传感器实际历史数据训练所述贝叶斯网络;或基于风向为所述目标风向时的传感器仿真历史数据训练所述贝叶斯网络。
图7为本发明实施方式的确定污染源的装置的示范性结构图。
在图7中,确定污染源的装置700包括一个存储器702和一个处理器701;存储器702中存储有可被处理器701执行的应用程序,用于使得处理器701执行如上任一项所述的确定污染源的方法。
其中,存储器702具体可以实施为电可擦可编程只读存储器(EEPROM)、快闪存储器(Flash memory)、可编程程序只读存储器(PROM)等多种存储介质。处理器701可以实施 为包括一或多个中央处理器或一或多个现场可编程门阵列,其中现场可编程门阵列集成一或多个中央处理器核。具体地,中央处理器或中央处理器核可以实施为CPU或MCU或DSP等等。
需要说明的是,上述各流程和各结构图中不是所有的步骤和模块都是必须的,可以根据实际的需要忽略某些步骤或模块。各步骤的执行顺序不是固定的,可以根据需要进行调整。各模块的划分仅仅是为了便于描述采用的功能上的划分,实际实现时,一个模块可以分由多个模块实现,多个模块的功能也可以由同一个模块实现,这些模块可以位于同一个设备中,也可以位于不同的设备中。
各实施方式中的硬件模块可以以机械方式或电子方式实现。例如,一个硬件模块可以包括专门设计的永久性电路或逻辑器件(如专用处理器,如FPGA或ASIC)用于完成特定的操作。硬件模块也可以包括由软件临时配置的可编程逻辑器件或电路(如包括通用处理器或其它可编程处理器)用于执行特定操作。至于具体采用机械方式,或是采用专用的永久性电路,或是采用临时配置的电路(如由软件进行配置)来实现硬件模块,可以根据成本和时间上的考虑来决定。
本发明还提供了一种机器可读的存储介质,存储用于使一机器执行如本文所述方法的指令。具体地,可以提供配有存储介质的***或者装置,在该存储介质上存储着实现上述实施例中任一实施方式的功能的软件程序代码,且使该***或者装置的计算机(或CPU或MPU)读出并执行存储在存储介质中的程序代码。此外,还可以通过基于程序代码的指令使计算机上操作的操作***等来完成部分或者全部的实际操作。还可以将从存储介质读出的程序代码写到***计算机内的扩展板中所设置的存储器中或者写到与计算机相连接的扩展单元中设置的存储器中,随后基于程序代码的指令使安装在扩展板或者扩展单元上的CPU等来执行部分和全部实际操作,从而实现上述实施方式中任一实施方式的功能。
用于提供程序代码的存储介质实施方式包括软盘、硬盘、磁光盘、光盘(如CD-ROM、CD-R、CD-RW、DVD-ROM、DVD-RAM、DVD-RW、DVD+RW)、磁带、非易失性存储卡和ROM。可选择地,可以由通信网络从服务器计算机或云上下载程序代码。
以上所述,仅为本发明的较佳实施方式而已,并非用于限定本发明的保护范围。凡在本发明的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本发明的保护范围之内。
Claims (12)
- 确定污染源的方法(100),其特征在于,包括:基于传感器与潜在污染源的位置关系,确定(101)目标风向下的、传感器与潜在污染源之间的因果关系;建立(102)对应于所述目标风向的、包含所述因果关系的贝叶斯网络;基于风向为所述目标风向时的传感器历史数据训练(103)所述贝叶斯网络;基于风向为所述目标风向时的传感器当前数据和已训练的所述贝叶斯网络,从所述潜在污染源中确定(104)出污染源。
- 根据权利要求1所述的确定污染源的方法(100),其特征在于,所述基于风向为目标风向时的传感器历史数据训练(103)所述贝叶斯网络包括:确定所述贝叶斯网络的先验概率分布;基于所述贝叶斯网络的先验概率分布和风向为目标风向时的所述传感器历史数据,以置信传播算法计算所述贝叶斯网络的后验概率分布。
- 根据权利要求2所述的确定污染源的方法(100),其特征在于,所述以置信传播算法计算所述贝叶斯网络的后验概率分布包括:在所述置信传播算法的每步迭代计算时,将风向为所述目标风向时的风速作为影响传感器监测值的参数,其中所述风速越大,所述传感器监测值越小。
- 根据权利要求1所述的确定污染源的方法(100),其特征在于,所述基于风向为目标风向时的传感器当前数据和已训练的所述贝叶斯网络,从所述潜在污染源中确定(104)出污染源包括:将风向为目标风向时的传感器当前数据输入所述已训练的贝叶斯网络;基于所述已训练的贝叶斯网络的后验概率和所述风向为目标风向时的传感器当前数据,计算每个潜在污染源的排放概率分布;对每个潜在污染源的排放概率分布的概率最高点所对应的排放值进行排序;将最大排放值所对应的潜在污染源确定为所述污染源。
- 根据权利要求1-4中任一项所述的确定污染源的方法(100),其特征在于,所述基于风向为目标风向时的传感器历史数据训练(103)所述贝叶斯网络包括:基于风向为目标风向时的传感器实际历史数据训练所述贝叶斯网络;或基于风向为目标风向时的传感器仿真历史数据训练所述贝叶斯网络。
- 确定污染源的装置(600),其特征在于,包括:关系确定模块(601),用于基于传感器与潜在污染源的位置关系,确定目标风向下的、 传感器与潜在污染源之间的因果关系;网络建立模块(602),用于建立对应于所述目标风向的、包含所述因果关系的贝叶斯网络;训练模块(603),用于基于风向为所述目标风向时的传感器历史数据训练所述贝叶斯网络;确定模块(604),用于基于风向为所述目标风向时的传感器当前数据和已训练的所述贝叶斯网络,从所述潜在污染源中确定出污染源。
- 根据权利要求6所述的确定污染源的装置(600),其特征在于,所述训练模块(603),用于确定所述贝叶斯网络的先验概率分布;基于所述贝叶斯网络的先验概率分布和风向为目标风向时的所述传感器历史数据,以置信传播算法计算所述贝叶斯网络的后验概率分布。
- 根据权利要求7所述的确定污染源的装置(600),其特征在于,所述训练模块(603),用于在所述置信传播算法的每步迭代计算时,将风向为所述目标风向时的风速作为影响传感器监测值的参数,其中所述风速越大,所述传感器监测值越小。
- 根据权利要求6所述的确定污染源的装置(600),其特征在于,所述确定模块(604),用于将风向为目标风向时的传感器当前数据输入所述已训练的所述贝叶斯网络;基于所述贝叶斯网络的后验概率和所述风向为目标风向时的传感器当前数据,计算每个潜在污染源的排放概率分布;对每个潜在污染源的排放概率分布的概率最高点所对应的排放值进行排序;将最大排放值所对应的潜在污染源确定为所述污染源。
- 根据权利要求6-9中任一项所述的确定污染源的装置(600),其特征在于,所述训练模块(603),用于基于风向为目标风向时的传感器实际历史数据训练所述贝叶斯网络;或基于风向为目标风向时的传感器仿真历史数据训练所述贝叶斯网络。
- 确定污染源的装置(700),其特征在于,包括处理器(701)和存储器(702);所述存储器(702)中存储有可被所述处理器(701)执行的应用程序,用于使得所述处理器(701)执行如权利要求1至5中任一项所述的确定污染源的方法(100)。
- 计算机可读存储介质,其特征在于,其中存储有计算机可读指令,该计算机可读指令用于执行如权利要求1至5中任一项所述的确定污染源的方法(100)。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202080105955.7A CN116324833A (zh) | 2020-12-17 | 2020-12-17 | 确定污染源的方法、装置和计算机可读存储介质 |
PCT/CN2020/137343 WO2022126544A1 (zh) | 2020-12-17 | 2020-12-17 | 确定污染源的方法、装置和计算机可读存储介质 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2020/137343 WO2022126544A1 (zh) | 2020-12-17 | 2020-12-17 | 确定污染源的方法、装置和计算机可读存储介质 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022126544A1 true WO2022126544A1 (zh) | 2022-06-23 |
Family
ID=82058858
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2020/137343 WO2022126544A1 (zh) | 2020-12-17 | 2020-12-17 | 确定污染源的方法、装置和计算机可读存储介质 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN116324833A (zh) |
WO (1) | WO2022126544A1 (zh) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115902114A (zh) * | 2022-12-07 | 2023-04-04 | 北京工业大学 | 基于半定量法的小尺度大气污染溯源方法 |
CN115936543A (zh) * | 2023-03-15 | 2023-04-07 | 湖北君邦环境技术有限责任公司 | 一种突发水污染事故污染溯源方法、***、设备及介质 |
CN117269443A (zh) * | 2023-09-11 | 2023-12-22 | 杭州智驳科技有限公司 | 一种基于大数据的智慧数字乡村环境监测*** |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110313958A1 (en) * | 2009-02-16 | 2011-12-22 | Institutt For Energiteknikk | System and method for empirical ensemble-based virtual sensing of particulates |
CN103914622A (zh) * | 2014-04-04 | 2014-07-09 | 清华大学 | 一种化学品泄漏快速预测预警应急响应决策方法 |
CN106770930A (zh) * | 2016-11-24 | 2017-05-31 | 北京蓝色星语科技有限公司 | 一种使用无人机进行危险品检测的方法及*** |
CN107341576A (zh) * | 2017-07-14 | 2017-11-10 | 河北百斛环保科技有限公司 | 一种大数据可视化的空气污染溯源和走向预测方法 |
CN107561600A (zh) * | 2017-09-15 | 2018-01-09 | 深圳市卡普瑞环境科技有限公司 | 一种大气质量检测方法以及检测设备 |
CN110531030A (zh) * | 2019-08-16 | 2019-12-03 | 北京慧辰资道资讯股份有限公司 | 一种基于物联网大数据大气污染物分析的方法及装置 |
-
2020
- 2020-12-17 WO PCT/CN2020/137343 patent/WO2022126544A1/zh active Application Filing
- 2020-12-17 CN CN202080105955.7A patent/CN116324833A/zh active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110313958A1 (en) * | 2009-02-16 | 2011-12-22 | Institutt For Energiteknikk | System and method for empirical ensemble-based virtual sensing of particulates |
CN103914622A (zh) * | 2014-04-04 | 2014-07-09 | 清华大学 | 一种化学品泄漏快速预测预警应急响应决策方法 |
CN106770930A (zh) * | 2016-11-24 | 2017-05-31 | 北京蓝色星语科技有限公司 | 一种使用无人机进行危险品检测的方法及*** |
CN107341576A (zh) * | 2017-07-14 | 2017-11-10 | 河北百斛环保科技有限公司 | 一种大数据可视化的空气污染溯源和走向预测方法 |
CN107561600A (zh) * | 2017-09-15 | 2018-01-09 | 深圳市卡普瑞环境科技有限公司 | 一种大气质量检测方法以及检测设备 |
CN110531030A (zh) * | 2019-08-16 | 2019-12-03 | 北京慧辰资道资讯股份有限公司 | 一种基于物联网大数据大气污染物分析的方法及装置 |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115902114A (zh) * | 2022-12-07 | 2023-04-04 | 北京工业大学 | 基于半定量法的小尺度大气污染溯源方法 |
CN115936543A (zh) * | 2023-03-15 | 2023-04-07 | 湖北君邦环境技术有限责任公司 | 一种突发水污染事故污染溯源方法、***、设备及介质 |
CN115936543B (zh) * | 2023-03-15 | 2023-06-06 | 湖北君邦环境技术有限责任公司 | 一种突发水污染事故污染溯源方法、***、设备及介质 |
CN117269443A (zh) * | 2023-09-11 | 2023-12-22 | 杭州智驳科技有限公司 | 一种基于大数据的智慧数字乡村环境监测*** |
CN117269443B (zh) * | 2023-09-11 | 2024-05-03 | 杭州智驳科技有限公司 | 一种基于大数据的智慧数字乡村环境监测*** |
Also Published As
Publication number | Publication date |
---|---|
CN116324833A (zh) | 2023-06-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2022126544A1 (zh) | 确定污染源的方法、装置和计算机可读存储介质 | |
CN111814308B (zh) | 一种面向自动驾驶***的加速测试*** | |
CN111537023B (zh) | 一种工业园区大气污染物扩散模拟与溯源方法 | |
KR101507397B1 (ko) | 대기 모델링 서비스 제공 시스템 | |
CN113033029A (zh) | 自动驾驶仿真方法、装置、电子设备及存储介质 | |
US11466992B2 (en) | Method, apparatus, device and medium for detecting environmental change | |
Jacquemart et al. | Adaptive interacting particle system algorithm for aircraft conflict probability estimation | |
CN112131336B (zh) | 一种污染源定位方法、装置、计算机设备及存储介质 | |
CN114511077A (zh) | 使用基于伪元素的数据扩增来训练点云处理神经网络 | |
Paul et al. | Formal verification of safety-critical aerospace systems | |
CN114329996A (zh) | 基于vbem方法的污染物点源扩散模型构建方法 | |
CN117421433A (zh) | 一种图文智能舆情分析方法及*** | |
Zhang et al. | Accelerated testing for highly automated vehicles: A combined method based on importance sampling and normalizing flows | |
Leufen et al. | Calculating the turbulent fluxes in the atmospheric surface layer with neural networks | |
Shahidi et al. | Hidden markov model based graph matching for calibration of localization maps | |
CN115238245B (zh) | 污染物监测方法、装置、存储介质及电子设备 | |
Zeng et al. | Efficient path evaluation for AUVs using adaptive B-spline approximation | |
Chandio et al. | Towards adaptable and tunable cloud-based map-matching strategy for GPS trajectories | |
CN113465613B (zh) | 一种城市轨道交通中隧道网络定位的地图匹配优化方法 | |
KR102619131B1 (ko) | 위성데이터 결측값 추정방법 및 시스템 | |
Zhang et al. | Empirical approximation for the stochastic fundamental diagram of traffic flow on signalized intersection | |
Vertovec et al. | Verification of safety critical control policies using kernel methods | |
CN117407319B (zh) | 用于雷达***软件测试的目标生成方法、装置及电子设备 | |
Espinoza-Huerta et al. | A multivariable computational fluid dynamics validation method based in Bayesian networks applied in a greenhouse | |
CN115051870B (zh) | 一种基于因果发现检测未知网络攻击的方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20965560 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20965560 Country of ref document: EP Kind code of ref document: A1 |