A multi-intelligent sensor cooperative working gas monitoring and analyzing system

Abstract

The application provides a gas monitoring and analyzing system with multiple intelligent sensors working in cooperation, and belongs to the technical field of sensor monitoring, and comprises the following: a region splitting module, which is used for griding the atmospheric environment of the split regions of a factory according to the current region facilities; an initial monitoring module, which is used for controlling fixed sensors and mobile sensors to cooperatively monitor the current gas information of each grid unit to obtain an initial environment map; a frequency setting module, which is used for updating the collection frequency of each fixed sensor and mobile sensor under each divided region in dependence on the initial environment map and predicted meteorological data; and a gas quality early warning module, which is used for receiving new gas information collected by the fixed sensors and mobile sensors, constructing a gas quality early warning map of the target factory, and outputting early warning. The monitoring network formed by the application is of "static coverage + dynamic tracking", and the sampling frequency is adaptively adjusted to avoid control response errors as much as possible and improve the monitoring efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sensor monitoring, in particular to a gas monitoring and analyzing system with multiple intelligent sensors working in cooperation. BACKGROUND

[0002] With the acceleration of industrialization, the impact of various gas pollutants (such as volatile organic compounds, nitrogen oxides, sulfides, etc.) discharged by factories on the atmospheric environment is increasingly significant. The traditional fixed-frequency data collection of factories cannot adapt to the real-time changes of meteorological conditions (such as wind speed, wind direction, temperature), resulting in delayed pollution warning, and the atmospheric environment monitoring has the following pain points: single or sparse sensor arrangement cannot cover the complex factory pollution distribution, making the capture ability limited, and fixed sensors and mobile sensors (such as devices carried by unmanned aerial vehicles and inspection robots) often work independently, and due to the possible control response error of the sensors, the collected data is not accurate, which further leads to low efficiency of gas monitoring.

[0003] Therefore, the present application provides a gas monitoring and analyzing system with multiple intelligent sensors working in cooperation. SUMMARY

[0004] The present application provides a gas monitoring and analyzing system with multiple intelligent sensors working in cooperation to solve the above technical problems.

[0005] The present application provides a gas monitoring and analyzing system with multiple intelligent sensors working in cooperation, comprising:

[0006] The area splitting module is used to refine the functional attributes of the target factory to achieve area splitting, and to grid the atmospheric environment according to the current regional facilities of the split areas, wherein the atmospheric environment grid result is a first refinement unit of the independent deviation closed contour in the split area and a second refinement unit of the remaining area except the independent deviation closed contour;

[0007] The initial monitoring module is used to set fixed sensors and mobile sensors according to the current regional facilities of each divided area and the grid unit, and to control the fixed sensors and mobile sensors to cooperatively monitor the current gas information of each grid unit to obtain an initial environment map, wherein the fixed sensors and mobile sensors constitute multiple intelligent sensors;

[0008] The frequency setting module is used to update the collection frequency of each fixed sensor and mobile sensor in each divided area according to the initial environment map and the predicted meteorological data;

[0009] The gas quality early warning module is configured to receive new gas information collected by the fixed sensors and the mobile sensors based on the edge nodes in communication connection with the fixed sensors and the mobile sensors, construct a gas quality early warning map of the target factory, and output early warning.

[0010] Preferably, the area splitting module comprises:

[0011] The function splitting unit is configured to perform unmanned aerial vehicle monitoring on the target factory to obtain a factory deployment structure, split the factory deployment structure based on a construction facility map of the target factory to obtain a plurality of split areas, wherein the construction facility map comprises at least one function attribute, and the split areas are actual split areas;

[0012] The set determination unit is configured to compare and analyze the split areas and standard areas to construct a deviation construction-position set, and determine a plurality of independent deviation position contours;

[0013] The contour supplement unit is configured to input the position set of each independent deviation position contour into an adaptive contour filling model to perform adaptive contour supplementing to obtain an independent deviation closed contour, and rely on the actual ratio of the contour length of the independent deviation closed contour to the area contour of the corresponding split area and the actual change coefficient of the facility type in the contour to the facility type in the corresponding standard area to match a first refinement unit of the independent deviation position contour from a ratio-coefficient-setting comparison table;

[0014] The first meshing unit is configured to perform first meshing on the corresponding independent deviation position contour in the split area according to the first refinement unit;

[0015] The second meshing unit is configured to perform second meshing on the remaining areas in the split area except the independent deviation position contour according to a second refinement unit of the remaining areas from a function-setting comparison table according to the second refinement unit.

[0016] Preferably, the first meshing unit comprises:

[0017] The complete judgment subunit is configured to perform first meshing on the corresponding independent deviation closed contour according to the first refinement unit, and judge whether there is an incomplete first refinement unit in the edge meshing unit of the independent deviation closed contour;

[0018] If not, the first meshing unit of the corresponding independent deviation closed contour is kept unchanged;

[0019] If so, the peripheral facility attribute of the incomplete first refinement unit is determined;

[0020] The function attribute judgment subunit is configured to retain the incomplete first refined unit if the peripheral facility attribute is irrelevant to the function attribute of the corresponding split region.

[0021] The expansion subunit is configured to replace the incomplete first refined unit with the complete first refined unit if the connection line segment is a straight line segment.

[0022] The expansion subunit is configured to replace the incomplete first refined unit with the complete first refined unit if the connection line segment is a straight line segment.

[0023] The expansion subunit is configured to replace the incomplete first refined unit with the complete first refined unit if the connection line segment is a straight line segment.

[0024] The expansion subunit is configured to replace the incomplete first refined unit with the complete first refined unit if the connection line segment is a straight line segment.

[0025] The expansion subunit is configured to replace the incomplete first refined unit with the complete first refined unit if the connection line segment is a straight line segment.

[0026] Preferably, the method further comprises:

[0027] The vector construction module is configured to construct an analysis vector according to the first position of each grid unit in the division region, the correlation between the function attribute of the division region and the unit facility of the corresponding grid unit extracted from the current region facility.

[0028] The pollutant determination module is configured to input the analysis vector into a gas analysis model to obtain a gas pollution set of the corresponding grid unit, wherein the gas pollution set comprises gas pollutants and diffusion factors of each gas pollutant.

[0029] The sensor setting module is configured to find a first pollutant with a diffusion factor less than a preset factor, set a fixed sensor in the corresponding grid unit, and set a mobile sensor according to the diffusion path of each remaining pollutant of the unit facility in each grid unit.

[0030] Preferably, the initial monitoring module comprises:

[0031] The response set construction unit is configured to issue N times of synchronous working instructions to all sensors in each grid unit based on the controller, capture a first time point of each issued instruction and a second time point at which each sensor receives the corresponding issued instruction and starts working, and construct a response set S of each sensor.

[0032] A matrix construction unit is configured to regard the time length difference between each response time length in each response set S of sensors of the same model and a standard time length as a first set, and sort the time length differences in each first set to construct a difference matrix;

[0033] A function determination unit is configured to divide the difference matrix dynamically to obtain a response function, and predict a feedback compensation time of a corresponding sensor;

[0034] An environment map construction unit is configured to distribute a (N+1)th instruction to the sensors based on the controller-dependent feedback compensation time, and output current gas information of each grid unit to form an initial environment map.

[0035] Preferably, the function determination unit comprises:

[0036] An initial division sub-unit is configured to lock mutation elements in each row vector in the difference matrix according to a preset time scale window, and perform initial division on the difference matrix to obtain a first matrix and a second matrix, wherein the column number of the first matrix is the maximum column number in the left row vector in the initial division result, the column number of the second matrix is the maximum column number in the right row vector in the initial division result, and when there is a column number of 0 in the left row vector or the right row vector, the same value is filled according to a new scale, when there is a column number that is not the maximum column number in the left row vector, the same value is backward filled according to the last element in the corresponding side row vector until the maximum column number is filled, and when there is a column number that is not the maximum column number in the right row vector, the same value is forward filled according to the first element in the corresponding side row vector until the maximum column number is filled.

[0037] A secondary division sub-unit is configured to determine a new scale window by adding a preset time scale window to the total number of historical uses and the decay time length after N synchronous working instructions based on the controller, and perform secondary division on the difference matrix to obtain a third matrix and a fourth matrix.

[0038] A compensation determination sub-unit is configured to solve the characteristic functions of the first matrix, the second matrix, the third matrix and the fourth matrix respectively, and calculate the feedback compensation time of the corresponding sensor according to the corresponding characteristic coefficients of each sensor in all characteristic functions.

[0039] Preferably, the frequency setting module comprises:

[0040] A model analysis unit is configured to input the predicted meteorological data into a gas parameter influence analysis model, and input the influence coefficient of each gas parameter on the predicted meteorological data;

[0041] The coefficient judging unit is configured to keep the original collection frequency of the corresponding sensor unchanged if the influence coefficient is less than the preset coefficient, and the original collection frequency is the first collection frequency at this time.

[0042] The frequency updating unit is configured to update the original collection frequency to obtain the first collection frequency if the influence coefficient is not less than the preset coefficient, according to the influence level of the influence coefficient and the preset coefficient, and in combination with the gas collection weight of the corresponding fixed sensor.

[0043] Meanwhile, the original collection frequency is updated to obtain the second collection frequency according to the influence level of the influence coefficient and the preset coefficient, and in combination with the gas collection weight of the corresponding mobile sensor in the moving path based on the predicted meteorological data corresponding to the predicted time period.

[0044] Preferably, the gas quality early warning module comprises:

[0045] The significant bar construction unit is configured to determine the quality situation of each gas parameter in each grid unit according to the new gas information, and draw a quality significant bar based on the corresponding grid unit, wherein the quality significant bar comprises a quality situation color block of each gas parameter and a comprehensive early warning block.

[0046] The early warning map construction unit is configured to construct a gas quality early warning map based on the quality significant bars of all grid units and output the map.

[0047] Compared with the prior art, the application has the following beneficial effects:

[0048] Based on the functional attributes and the pollution source characteristics, the region is refined, the "independent deviation closed contour" (such as a closed equipment area) and the "remaining region" are distinguished, the differentiated design of the monitoring unit is realized, the fixed sensor provides basic data, the mobile sensor dynamically fills in the blind, the monitoring network of "static coverage + dynamic tracking" is formed, the sampling frequency is adaptively adjusted in combination with the initial environment map and the meteorological prediction, the control response error is as far as possible avoided, the monitoring efficiency is improved, the real-time early warning map is constructed, and the pollution response time is effectively shortened.

[0049] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be achieved and obtained by the structure particularly pointed out in the written description and the accompanying drawings.

[0050] The technical solutions of the present application will be further described in detail below with the help of the accompanying drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0051] The accompanying drawings are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and are used to explain the application without restricting it. In the drawings:

[0052] Figure 1 A structural diagram of a gas monitoring and analyzing system with multiple intelligent sensors working in cooperation in an embodiment of the application;

[0053] Figure 2 A deployment diagram of multiple intelligent sensors in an embodiment of the application;

[0054] Figure 3 A structural diagram of a quality significant strip in an embodiment of the application;

[0055] Figure 4 A structural diagram of an initial rectangular frame in an embodiment of the application;

[0056] Figure 5 Unit expansion based on a straight line segment in an embodiment of the application;

[0057] Figure 6 Unit expansion based on a curved line segment in an embodiment of the application. DETAILED DESCRIPTION

[0058] The preferred embodiments of the application are described below in conjunction with the accompanying drawings, which should be understood as merely illustrative and explanatory, and are not intended to limit the application.

[0059] A gas monitoring and analyzing system with multiple intelligent sensors working in cooperation in an embodiment of the application, as shown in Figure 1 includes:

[0060] A region splitting module, configured to implement region splitting by refining the functional attributes of a target factory, and to grid the atmospheric environment of the split regions according to the current regional facilities, wherein the grid result of the atmospheric environment is a first refined unit of an independent deviation closed contour in the split region and a second refined unit of the remaining region except the independent deviation closed contour;

[0061] An initial monitoring module, configured to set fixed sensors and mobile sensors according to the current regional facilities and the grid units of each divided region, and to control the fixed sensors and the mobile sensors to cooperatively monitor the current gas information of each grid unit to obtain an initial environment map, wherein the fixed sensors and the mobile sensors constitute multiple intelligent sensors;

[0062] A frequency setting module, configured to update the collection frequency of each fixed sensor and mobile sensor under each divided region according to the initial environment map and predicted meteorological data;

[0063] The gas quality early warning module is configured to receive new gas information collected by the fixed sensors and the mobile sensors based on the edge nodes in communication connection with the fixed sensors and the mobile sensors, construct a gas quality early warning map of the target factory, and output early warning.

[0064] In this embodiment, the target factory refers to a petroleum refinery, and the functional attributes include high-temperature combustion attributes, chemical reaction attributes, material storage attributes, and wastewater treatment attributes. The high-temperature combustion attributes correspond to the heating furnace and boiler area of the petroleum refinery, the chemical reaction attributes correspond to the reaction tower and reactor area of the petroleum refinery, the material storage attributes correspond to the crude oil tank and product oil tank area of the petroleum refinery, and the wastewater treatment attributes correspond to the aeration tank and sludge dewatering room area of the petroleum refinery, thereby realizing area splitting.

[0065] In this embodiment, the lower area facility refers to the facility conditions corresponding to the split area, including facility layout and facility products. For example, the facilities existing in the crude oil tank and product oil tank area include fixed roof tanks, floating roof tanks, loading and unloading trestles, and oil gas recovery devices.

[0066] In this embodiment, the independent deviation closed contour is determined based on the difference between the actual factory facility and the pre-planned factory facility, and the first refinement unit and the second refinement unit are both divided in a grid form.

[0067] In this embodiment, the fixed sensors include, for example, electrochemical sensors (monitoring H2S, SO2), PID photoionization detectors (monitoring VOC S ), and laser scattering instruments (monitoring PM 2.5 , PM 1.0 ).

[0068] The mobile sensors include, for example, multi-gas detectors carried by inspection robots, monitoring oxygen, carbon monoxide, etc. The training path of each mobile sensor is pre-planned.

[0069] In this embodiment, the collaborative monitoring refers to trying to ensure that the gas data collected by the sensors are collected at the same time, facilitating global analysis of the overall gas condition of the factory at a certain time.

[0070] In this embodiment, the initial environment map contains the gas data collected by each sensor in each grid unit and is displayed on the geographical location map corresponding to the factory.

[0071] In this embodiment, the predicted meteorological data is obtained based on a weather station.

[0072] In this embodiment, the frequency update is based on the data synchronization collection of the same type of sensor, and can ensure the comprehensiveness of gas monitoring according to the actual situation of the factory, so as to avoid monitoring omission caused by abnormality.

[0073] In this embodiment, the edge node adopts an industrial-grade edge computing gateway.

[0074] In this embodiment, the early warning map is composed of quality significant strips of all grid units.

[0075] The beneficial effects of the above technical scheme are: based on the functional attribute and the pollution source characteristic, the region is refined, the 'independent deviation closed contour' (such as a closed equipment area) and the'remaining area' are distinguished, the monitoring unit is designed differently, the fixed sensor provides basic data, the mobile sensor dynamically fills in the blind area, a monitoring network of'static coverage + dynamic tracking' is formed, the initial environment map and the weather forecast are combined, the sampling frequency is adaptively adjusted, the control response error is avoided as much as possible, the monitoring efficiency is improved, a real-time early warning map is constructed, and the pollution response time is effectively shortened.

[0076] The application discloses a gas monitoring and analyzing system based on multi-intelligent sensor cooperation.

[0077] The function splitting unit is used for obtaining a factory deployment structure through unmanned aerial vehicle monitoring of a target factory, splitting a function structure of the factory deployment structure based on a construction facility map of the target factory to obtain a plurality of split regions, wherein the construction facility map contains at least one function attribute, and the split regions are actual split regions;

[0078] The set determination unit is used for comparing and analyzing the split regions and standard regions to construct a deviation construction-position set, and determining a plurality of independent deviation position contours;

[0079] The contour supplement unit is used for inputting a position set of each independent deviation position contour into an adaptive contour filling model to perform adaptive contour supplementing to obtain an independent deviation closed contour, and matching a first refinement unit of the independent deviation position contour from a ratio-coefficient-setting comparison table according to an actual ratio of a contour length of the independent deviation closed contour to a region contour of a corresponding split region and an actual change coefficient of a facility type in the contour to a facility type in a corresponding standard region.

[0080] The first grid unit is used for performing first gridding on a corresponding independent deviation position contour in the split region according to the first refinement unit.

[0081] The second meshing unit is configured to mesh the remaining area in the split area except the independent deviation position contour according to the second refinement unit matched with the remaining area from the function-setting correspondence table according to the area function, and to perform second meshing on the remaining area according to the second refinement unit.

[0082] In this embodiment, the plant deployment structure is the physical layout structure of the refinery obtained by aerial photography of a drone and three-dimensional modeling technology, the construction facility map is a CAD drawing in the design stage of the refinery, and the actual split area is realized based on the plant deployment structure.

[0083] In this embodiment, the standard area is an area corresponding to a function determined based on a CAD drawing.

[0084] In this embodiment, the deviation construction-position set is a set of areas and position coordinates that are different from the actual construction and the standard design, for example, new temporary loading and unloading ports, expansion of the tank area, and other operations, which result in that the actual plant deployment structure is different from the structure planned on the CAD drawing, so it is necessary to determine the existing deviation positions to provide a basis for subsequent meshing.

[0085] In this embodiment, the independent deviation position contour is a geometric contour of a deviation area with a clear boundary obtained by sequentially connecting the deviation construction-position set, and the contour can be an unclosed line segment, so meshing is required, and it is more reasonable to realize the independent deviation closed contour based on the closed area. The independent deviation closed contour is constructed based on the independent deviation position contour.

[0086] In this embodiment, the adaptive contour filling model is based on a machine learning algorithm, and is a model for predicting a complete closed contour according to known contour points, and is implemented using a U-Net network.

[0087] In this embodiment, the first refinement unit is a fine monitoring unit designed for the independent deviation area, and the second refinement unit is a regular monitoring unit set for the remaining area (standard area).

[0088] In this embodiment, the ratio-coefficient-setting correspondence table is as follows:

[0089]

[0090]

[0091] For example, the contour length ratio is 0.9, and the facility coefficient changes to 0.8, at this time, the size of the first refinement unit obtained is 8m x 8m. After the meshing unit is determined, a sensor can be directly installed according to the requirements of the plant gas monitoring. Generally, one set of solid sensors (solid sensors for monitoring different gas parameters) and one mobile sensor are installed in each meshing unit.

[0092] In this embodiment, the function-setting reference table is a function attribute of a region function and a region division unit size (a second refinement unit) matched with the function attribute, which is directly obtained, for example, by using 10m*10m as the second refinement unit to grid the corresponding remaining region.

[0093] In this embodiment, the independent deviation region is focused on grid division to ensure the comprehensiveness of monitoring.

[0094] In this embodiment, the facility change coefficient is realized based on the facility layout complexity of the factory.

[0095] In this embodiment, the product of (the actual type quantity of the facility type in the contour / the type quantity of the facility type in the standard region) and (the total quantity of the facility involved in the actual type quantity / the total quantity of the facility of the type quantity of the facility type in the standard region) is the facility change coefficient.

[0096] The beneficial effects of the above technical solution are: through region refinement splitting and differentiated sensor configuration, the reliability and comprehensiveness of monitoring of the gas information of the target factory are ensured.

[0097] The application discloses a multi-intelligent sensor cooperative gas monitoring and analyzing system.

[0098] The complete judgment subunit is used for performing first grid division on the corresponding independent deviation closed contour according to the first refinement unit, and judging whether there is an incomplete first refinement unit in the edge grid unit of the independent deviation closed contour.

[0099] If not, the first grid unit of the corresponding independent deviation closed contour is kept unchanged.

[0100] If yes, the peripheral facility attribute of the incomplete first refinement unit is determined.

[0101] The function attribute judgment subunit is used for judging whether the peripheral facility attribute is irrelevant to the function attribute of the corresponding split region.

[0102] If the peripheral facility attribute is relevant to the function attribute of the corresponding split region, and the corresponding peripheral facility belongs to a non-deviation region, the incomplete first refinement unit and the non-deviation region of the peripheral facility are connected by a line segment.

[0103] The expansion subunit is used for judging whether the connecting line segment is a straight line segment.

[0104] If the connecting line segment is a curve segment, then the outermost lateral boundary point and the outermost longitudinal boundary point of the curve segment are determined, and an initial rectangular frame is drawn;

[0105] The initial rectangular frame is used as an expansion area to expand the corresponding incomplete first refinement unit;

[0106] If the peripheral facility attribute is related to the functional attribute of the corresponding split area, and the corresponding peripheral facility belongs to a deviation area, then the incomplete first refinement unit is retained.

[0107] In this embodiment, the independent deviation closed contour is meshed according to the requirements of the first refinement unit, so as to more accurately monitor the gas information in the area. For example, the small tank area is divided into square meshes with a side length of 5 meters, and each mesh is a first meshed unit.

[0108] In this embodiment, the incomplete first refinement unit is a meshed unit at the edge of the contour, which does not fully meet the initial design standards of the first refinement unit in terms of shape, size, or contained facilities. For example, at the edge of the tank area, a meshed unit has an actual area of only half of the standard mesh due to the blocking of the surrounding wall, and it is impossible to deploy the planned number of complete sensors. This is an incomplete first refinement unit.

[0109] In this embodiment, the peripheral facility attribute is the function, type, risk level, and other characteristics of the facilities around the incomplete first refinement unit. There is an oil pipeline (function: transporting crude oil) around the incomplete unit, and there is also a small oil pump house (type: auxiliary oil transportation facility, with a certain risk of oil and gas leakage). These are the peripheral facility attributes.

[0110] In this embodiment, the non-deviation area refers to an area that meets the factory design standards and functional planning and has no construction deviation.

[0111] In this embodiment, the connecting line segment is a line segment formed at the connection between the incomplete first refinement unit and the peripheral non-deviation area, which can be a straight line or a curve. When the incomplete unit is adjacent to a regular non-deviation tank area, the boundary line between them is the connecting line segment. If the boundary line is a straight wall, it is a straight line segment; if the boundary line is a curved pipeline, it is a curve segment.

[0112] In this embodiment, for the curve segment connecting line segment, the most edge points in the horizontal direction and the vertical direction are the outermost lateral boundary point and the outermost longitudinal boundary point, as shown in FIG. 1C. Figure 4 The curve segment is A1, and the obtained outermost lateral boundary points are r1 and r2, and the outermost longitudinal boundary points are r3 and r4. After connecting the lateral and longitudinal directions, an initial rectangular frame J is obtained.

[0113] As shown in the embodiment, Figure 5 B1 is an incomplete first refinement unit, B2 is partially replaced by a complete first refinement unit to the incomplete first refinement unit, that is, to realize the replacement.

[0114] As shown in the embodiment, Figure 6 C1 is an incomplete first refinement unit, and J is an initial rectangular frame, at this time, the position and size of J are the expansion of the incomplete first refinement unit.

[0115] The beneficial effects of the above technical solutions are: for the incomplete unit irrelevant to the functional attribute of the split region, the present situation is retained, and unnecessary area is avoided. When the incomplete unit is connected with the related non-biased region and the connection line segment is a straight line segment or can be expanded by drawing an initial rectangular frame, the sensor can be reasonably deployed to the key area, the incomplete unit is reasonably expanded to meet the first refinement unit standard, the consistency and standardization of sensor deployment, sampling and other operations in each monitoring unit are ensured, the data deviation caused by incomplete unit is reduced, the initial rectangular frame is expanded for the curved segment connection, which can better adapt to the complex facility layout in the refinery, and according to the actual situation of the incomplete unit and the peripheral facility attribute, it is determined whether to expand, so that the monitoring system can better adapt to various construction deviations in the refinery.

[0116] The gas monitoring and analyzing system of the multi-intelligent sensor cooperative work further comprises:

[0117] The vector construction module is configured to construct an analysis vector according to a first position of each grid unit in the division region, a correlation between the functional attribute of the division region, and a unit facility of the corresponding grid unit extracted from the current region facility.

[0118] The pollutant determination module is configured to input the analysis vector into a gas analysis model to obtain a gas pollution set of the corresponding grid unit, wherein the gas pollution set comprises gas pollutants and diffusion factors of each gas pollutant.

[0119] The sensor setting module is configured to find a first pollutant with a diffusion factor less than a preset factor, and set a fixed sensor to the corresponding grid unit, and set a mobile sensor according to a diffusion path of each remaining pollutant of the unit facility in each grid unit.

[0120] In the embodiment, the first position is a specific spatial position coordinate of the grid unit in the division region to which the grid unit belongs, for example, the coordinate of the grid unit is (X=100 meters, Y=200 meters), at this time, it is the first position.

[0121] In this embodiment, when the sub-area facility is the various production equipment, auxiliary facilities and the like currently possessed in the divided area, the unit facility is the specific facility contained in each grid unit, which can include a section of catalyst conveying pipeline, a small valve and the like, and these are the unit facilities of the unit,

[0122] In this embodiment, the analysis vector is a multi-dimensional vector composed of the first position of the grid unit, the correlation with the functional attribute of the divided area and the unit facility and the like, which is used for inputting the model for analysis. The first position coordinates are (X=150 meters, Y=180 meters), the correlation with the functional attribute of the area (catalytic cracking reaction, generation of sulfur-containing waste gas and the like) is 0.6, and the unit facility is a section of reaction pipeline and a control valve, so the analysis vector can be represented as [150, 180, correlation coefficient, reaction pipeline, control valve].

[0123] In this embodiment, the correlation coefficient is directly obtained based on the facility function matching degree, and specifically: matching facility quantity / total quantity of facilities corresponding to the functional area.

[0124] In this embodiment, the gas analysis model uses a neural network model constructed by a deep learning algorithm, and the vector obtained by combining the historical gas monitoring data, facility layout, production process and the like of the refinery in the required dimension is used as the input sample of the model, and the pollution condition corresponding to the corresponding vector is used as the output sample, and the neural network model is trained to obtain.

[0125] In this embodiment, the gas pollution set contains the types of gas pollutants in the grid unit and the set of diffusion factors of each pollutant. In a certain grid unit, the gas pollution set is {[sulfur dioxide, 0.3], [catalyst dust, 0.5]}, indicating that there are two pollutants, sulfur dioxide and catalyst dust, in the unit, and their diffusion factors are 0.3 and 0.5 respectively. The diffusion factor reflects the diffusion difficulty or diffusion speed of the pollutant in the area.

[0126] The diffusion factor is a parameter for measuring the diffusion characteristics of the gas pollutant in the grid unit. The smaller the value, the more difficult or slower the diffusion. For example, the diffusion factor of sulfur dioxide in the gas pollution set is 0.3, which means that it is relatively difficult to diffuse in the grid unit. The diffusion factor of catalyst dust is 0.5, which is relatively easier to diffuse.

[0127] In this embodiment, the preset factor is a critical value artificially set to judge the diffusion characteristics of the pollutant. According to the actual monitoring experience and needs of the refinery, the preset factor is set to 0.4. If the diffusion factor of a certain pollutant is less than 0.4, it is considered to be relatively difficult to diffuse. The diffusion factor of sulfur dioxide is 0.3, which is less than the preset factor 0.4, so sulfur dioxide is the first pollutant in the grid unit.

[0128] In this embodiment, the diffusion path is the route followed by the diffusion movement of pollutants in the environment, and catalyst dust may diffuse from the reaction tower to the surrounding area along the direction of the airflow in the workshop, and the specific movement route thereof in the corresponding grid unit is the diffusion path, which can be determined by airflow simulation, actual monitoring tracking, etc.

[0129] As shown in Figure 2 , it is a structure diagram of a multi-intelligent sensor, which includes, for example, a grid unit 1, a grid unit 2, and a grid unit 3, and each unit includes fixed sensors and mobile sensors.

[0130] The beneficial effects of the above technical solution are: by determining the first pollutant, fixed sensors are arranged in the area where the substance is relatively difficult to diffuse but may cause serious pollution, and mobile sensors are arranged according to the diffusion path of the remaining pollutants, so that the mobile sensors can dynamically monitor along the actual diffusion route of the pollutants, and the specific pollution condition in each grid unit can be understood according to the gas pollution set output by the gas analysis model, the monitoring pertinence is improved, and blind monitoring is avoided.

[0131] The application discloses a gas monitoring and analyzing system with multiple intelligent sensors working cooperatively, and the initial monitoring module comprises:

[0132] A response set construction unit is configured to issue N times of synchronous working instructions to all sensors under each grid unit based on the controller, capture a first time point of each issued instruction and a second time point at which each sensor receives the corresponding issued instruction and starts working, and construct a response set S of each sensor.

[0133] A matrix construction unit is configured to regard a time length difference between each response time length and a standard time length of sensors of the same type in each response set S as a first set, sort the time length differences in each first set, and construct a difference matrix.

[0134] A function determination unit is configured to dynamically divide the difference matrix to obtain a response function and predict a feedback compensation time of the corresponding sensor.

[0135] An environment map construction unit is configured to issue an N+1th instruction to the sensor based on the controller depending on the feedback compensation time, output current gas information of each grid unit, and construct an initial environment map.

[0136] In this embodiment, due to the delay of the sensor in receiving the instruction issued by the controller, but the required measurement of gas information needs to be executed synchronously (the gas measurement results of the target factory at the same time point need to be acquired synchronously), so it is necessary to adjust the time of the instruction received by the sensor to ensure that the sensors of the same type can work at the same time point as much as possible, and ensure the accuracy of the measurement of the same gas parameter of the target factory.

[0137] In this embodiment, the controller is used to manage and control the equipment or system of the sensor, and can send instructions, collect data and process, in the oil refinery, the industrial computer system arranged in the central control room is used as the controller, which is connected with the sensors of each grid unit through wired or wireless network, and uniformly coordinates the operation of the sensors.

[0138] In this embodiment, the first time point is the specific time when the controller issues the synchronous working instruction, and the second time point is the time when each sensor receives the synchronous working instruction and starts to work.

[0139] In this embodiment, the set S is composed of the first time point and the second time point when each sensor receives the instruction each time, and is used to record the response time of the sensor.

[0140] The sulfur dioxide sensor is tested for three times of synchronous working instructions, and the corresponding first time points are 8:00:00, 12:00:00 and 16:00:00, and the second time points are 8:00:03, 12:00:02 and 16:00:04, so the response set S of the sensor is {(8:00:00, 8:00:03), (12:00:00, 12:00:02), (16:00:00, 16:00:04)}.

[0141] In this embodiment, the response time is the time interval from the first time point (instruction issuing) to the second time point (starting to work).

[0142] In this embodiment, the standard time length is the response time of the sensor under the ideal state, for example, according to the performance parameters of the sensor and the requirements of the oil refinery, the standard time length of the sensor of this type is set to 2 seconds.

[0143] In this embodiment, in the response set of the same type sensor, the set composed of the time length difference between each response time and the standard time length is the first set, the difference matrix is composed of the time length difference in the first set after being sorted from small to large, the first set contains the time length difference {3, 1, 0, 2}, and the difference matrix is [0, 1, 2, 3] after sorting.

[0144] In this embodiment, the feedback compensation time is predicted according to the response function, so that the sensor can work more accurately and synchronously, and the time needs to be advanced or delayed when the next instruction is issued, the response time difference of a certain sensor is in the interval (1-2], and according to the response function, the feedback compensation time of the sensor is 1 second. This means that the controller needs to send an instruction to the sensor 1 second in advance to compensate for its response delay when the next instruction is issued.

[0145] In this embodiment, based on the controller issuing the N+1th instruction according to the feedback compensation time, a graph is drawn by collecting the current gas information output by the sensor in each grid unit, which is used to intuitively show the gas environment in the oil refinery.

[0146] The beneficial effects of the above technical solutions are: by constructing a response set, analyzing the time difference and performing feedback compensation, the same type of sensors can start working more closely after receiving the instruction, the improvement of synchronization enables multiple sensors to work more effectively in cooperation, and the instruction issuing time is adjusted according to the feedback compensation time, avoiding monitoring confusion caused by sensor response delay or advance.

[0147] The present application is a kind of multi-intelligent sensor cooperative gas monitoring and analyzing system, the function determination unit comprises:

[0148] The initial division sub-unit is used for locking the mutation elements in each row vector in the difference matrix according to the preset time scale window, and the initial division of the difference matrix obtains a first matrix and a second matrix, wherein the column number of the first matrix is the maximum column number in the left row vector in the initial division result, the column number of the second matrix is the maximum column number in the right row vector in the initial division result, and when there is a column number of 0 in the left row vector or the right row vector, the same value is filled according to the new scale, when there is a column number that is not the maximum column number in the left row vector, the same value is backward filled according to the last element in the corresponding side row vector until the maximum column number is filled, and when there is a column number that is not the maximum column number in the right row vector, the same value is forward filled according to the first element in the corresponding side row vector until the maximum column number is filled.

[0149] The secondary division sub-unit is used for determining the new scale window by adding the historical total number of uses and the decay time after N synchronous working instructions based on the controller and the preset time scale window, and performing secondary division on the difference matrix to obtain a third matrix and a fourth matrix.

[0150] The compensation determination sub-unit is used for solving the characteristic function of the first matrix, the second matrix, the third matrix and the fourth matrix respectively, and calculating the corresponding characteristic coefficients in all characteristic functions to obtain the feedback compensation time of the corresponding sensor.

[0151] In this embodiment, the preset time scale window is 1.5 seconds.

[0152] In this embodiment, N is set to 5 here for simple calculation, assuming that there are 3 sensors of the same model, and it should be noted that the actual test number N is greater than 10, and each row corresponds to the test result of a sensor.

[0153] Since the preset time scale window is 1.5 seconds, after the difference matrix is split, the following is obtained At this time, the maximum number of columns for the first matrix is 5, and the maximum number of columns for the second matrix is 5.

[0154] In this embodiment, the decay duration is obtained from the total number of N times of synchronization instructions from the total number of historical use times of the controller since the start of use, and is matched from the number-duration table, at this time, the obtained decay coefficient is 0.5, and then the new scale window is 2 seconds, and it should be noted that the number-duration table contains different use times and the decay duration of the corresponding instruction issuing time, which is set before the factory and can be directly used, at this time, the obtained It should be noted that the acquisition of the third matrix and the fourth matrix is similar to the acquisition principle of the first matrix and the second matrix, which will not be repeated here.

[0155] In this embodiment, the characteristic functions of the four matrices obtained by solving the characteristic function of each matrix are respectively:

[0156] T1=a1x1+a2x2+...+anxn

[0157] T2=b1x1+b2x2+...+bnxn

[0158] T3=c1x1+c2x2+...+cnxn

[0159] T4=d1x1+d2x2+...+dnxn

[0160] Wherein, T1, T2, T3, T4 are the characteristic functions of the first matrix, the second matrix, the third matrix and the fourth matrix respectively, that is, the characteristic functions obtained by solving the matrix are the public knowledge in mathematics, and x1, x2...xn are each sensor under the same model; a1, a2...an are the characteristic coefficients corresponding to each sensor based on the characteristic function of the first matrix; b1, b2...bn are the characteristic coefficients corresponding to each sensor based on the characteristic function of the second matrix; c1, c2...cn are the characteristic coefficients corresponding to each sensor based on the characteristic function of the third matrix; d1, d2...dn are the characteristic coefficients corresponding to each sensor based on the characteristic function of the fourth matrix;

[0161] The feedback compensation time Bi of the i-th sensor under the same model is calculated according to the following formula:

[0162]

[0163] Wherein, m0 represents the column number of the first matrix; m2 represents the column number of the third matrix; M1 represents the total number of sensors under the same model; ai, ci, bi, di respectively represent the characteristic coefficients of the i-th sensor under the same model based on the first matrix, the third matrix, the second matrix and the fourth matrix.

[0164] The characteristic coefficients are calculated in time units, so the basic unit of the matrix is a time unit.

[0165] In this embodiment, the characteristic coefficients of the four matrices are combined in a certain proportion, so that the influence of each factor on the feedback compensation time is balanced, The weights of the characteristic coefficients of the first matrix and the third matrix are determined respectively, The weights of the characteristic coefficients of the second matrix and the fourth matrix are determined, avoiding the excessive amplification or reduction of the influence of a certain factor, and ensuring that the calculation result can comprehensively and balancedly reflect the combined action of multiple factors.

[0166] The beneficial effects of the above technical solution are: based on the multiple information reflected by multiple matrices, the feedback compensation time of each sensor is accurately calculated. In this way, the sensors can work more accurately when receiving instructions, reducing the data deviation and monitoring error caused by inconsistent response time, and improving the monitoring accuracy and efficiency of the entire sensor network.

[0167] The application discloses a gas monitoring and analyzing system for collaborative work of multiple intelligent sensors, and the frequency setting module comprises:

[0168] The model analysis unit is configured to input the predicted meteorological data into a gas parameter influence analysis model, and input an influence coefficient of each gas parameter on the predicted meteorological data.

[0169] The coefficient judgment unit is configured to keep the original collection frequency of the corresponding sensor unchanged if the influence coefficient is less than a preset coefficient, and in this case, the original collection frequency is the first collection frequency.

[0170] The frequency updating unit is configured to update the original collection frequency to obtain the first collection frequency according to the influence level of the influence coefficient and the preset coefficient and in combination with the gas collection weight of the corresponding fixed sensor if the influence coefficient is not less than the preset coefficient.

[0171] Meanwhile, according to the influence level of the influence coefficient and the preset coefficient, and in combination with the gas collection weight of the corresponding mobile sensor based on the moving path of the mobile sensor in the predicted time period corresponding to the predicted meteorological data, the original collection frequency is updated to obtain a second collection frequency.

[0172] In this embodiment, the gas parameter influence analysis model is a mathematical model established based on the relationship between meteorological data and gas parameters, which is used to analyze the influence degree of the predicted meteorological data on various gas parameters (such as gas concentration, diffusion speed, etc.). After the predicted meteorological data is determined, the diffusion of the corresponding gas parameter can be directly obtained, and then the influence coefficient is obtained. For example, the model can analyze the influence of these meteorological conditions on the concentration and diffusion of sulfur dioxide, volatile organic compounds (VOCs) and other gases discharged by the oil refinery. Specifically, it is found through the calculation of the gas parameter influence analysis model that under the above predicted meteorological conditions, the influence coefficient of wind speed on the diffusion speed of VOCs discharged by the oil refinery is 0.8.

[0173] In this embodiment, the gas parameter is a description of the type of gas that the sensor type needs to measure.

[0174] In this embodiment, the preset coefficient is a threshold value preset according to experience or actual demand. According to the historical monitoring data and actual operation experience of the oil refinery, the preset coefficient of the influence of wind speed on gas diffusion is set to 0.6. When the calculated influence coefficient is compared with the preset coefficient, it can be determined whether to adjust the sensor collection frequency.

[0175] In this embodiment, the original collection frequency is the frequency of collecting data preset by the sensor without considering the influence of meteorological data. A certain fixed sensor is used to monitor the sulfur dioxide concentration in the catalytic cracking unit area of the oil refinery, and the original collection frequency is set to collect data every 10 minutes. The original collection frequency of a certain mobile sensor (such as a sensor carried by a drone) is to collect data every 30 minutes for designated area inspection.

[0176] In this embodiment, the first collection frequency is obtained by updating the original collection frequency of the fixed sensor according to the comparison result of the influence coefficient and the preset coefficient and related conditions. For example, according to the influence level and the gas collection weight of the sensor, the original collection frequency is updated from every 10 minutes to every 5 minutes. This every 5 minutes is the first collection frequency. According to the different levels divided according to the difference between the influence coefficient and the preset coefficient, the adjustment range of the collection frequency is determined more finely. The difference between the influence coefficient and the preset coefficient is divided into three levels, the difference is between 0-0.2 for low influence level, the difference is between 0.2-0.5 for medium influence level, and the difference is greater than 0.5 for high influence level. Different influence levels correspond to different collection frequency adjustment strategies, such as a larger adjustment range of the collection frequency under high influence level.

[0177] In this embodiment, the gas collection weight reflects the importance or contribution of the sensor in collecting certain gas data, and is pre-set and directly used.

[0178] In this embodiment, the second collection frequency is for a mobile sensor, and the original collection frequency of the mobile sensor carried by a certain unmanned aerial vehicle is once every 30 minutes. According to the predicted meteorological data, it is analyzed that the influence coefficient of the monitored gas parameter is greater than the preset coefficient, and the gas collection weight and the influence level of the mobile sensor in the corresponding mobile path are combined to update the collection frequency to once every 15 minutes, which is the second collection frequency.

[0179] The beneficial effects of the above technical scheme are: when the meteorological data changes to cause the influence coefficient to be greater than the preset coefficient, the sensor collection frequency is timely increased, and when the influence coefficient is less than the preset coefficient, the original collection frequency of the sensor is maintained, unnecessary high-frequency collection is avoided when the meteorological condition has little influence on the gas parameter, and the data transmission amount and processing pressure of the sensor are reduced. The collection frequency is adjusted in combination with the gas collection weight, so that the sensor at the key position can collect data at a higher frequency when the meteorological condition changes, and the sensor at the relatively secondary position is adjusted according to the actual situation. In this way, the monitoring resources can be concentrated in the more important area, the monitoring accuracy and efficiency of the key pollution source are improved, and the collection frequency is adjusted according to the gas collection weight of the mobile sensor in the mobile path in different prediction periods, so that the mobile sensor can more reasonably plan the monitoring path and collection time.

[0180] The gas monitoring and analyzing system with multiple intelligent sensors working in cooperation comprises a gas quality early warning module, a gas quality prediction module, a gas quality monitoring module, a gas quality analysis module, and a gas quality early warning module.

[0181] A significant bar construction unit is configured to determine the quality of each gas parameter in each grid unit based on the new gas information, and draw a quality significant bar based on the corresponding grid unit, wherein the quality significant bar comprises a quality color block of each gas parameter and a comprehensive early warning block.

[0182] An early warning map construction unit is configured to construct a gas quality early warning map based on the quality significant bars of all grid units and output the map.

[0183] In this embodiment, it is assumed that the sulfur dioxide concentration threshold of a petroleum refinery is 4 ppm, and when the sulfur dioxide concentration in a certain grid unit is 5 ppm, the quality is considered to be slightly over standard. If the nitrogen oxide concentration threshold is 2 ppm, and the nitrogen oxide concentration in a certain unit is 3 ppm, the quality is also slightly over standard. It is assumed that if it exceeds 2 times the corresponding threshold, it is considered to be moderately over standard, and if it exceeds 3 times the threshold, it is considered to be severely over standard.

[0184] In this embodiment, the comprehensive early warning block gives an overall early warning mark according to the quality of all gas parameters in a grid unit, which is used to quickly reflect the overall state of the gas quality of the unit. The display result of the comprehensive early warning block is obtained by matching the combination of whether the gas parameters meet the standard or not with a combination-comprehensive comparison table. The table contains the combination of different gas parameters and the comprehensive result for the combination, which are all pre-stored and can be directly used. It is assumed that green represents that the gas parameter meets the standard, yellow represents that the gas parameter slightly exceeds the standard, orange represents that the gas parameter moderately exceeds the standard, and red represents that the gas parameter severely exceeds the standard.

[0185] For example, the quality significant bar as shown in FIG. 6 shows that if the sulfur dioxide slightly exceeds the standard (yellow), the nitrogen oxides severely exceed the standard (red), and the VOCs meet the standard (green) in a grid unit, a black exclamation mark is displayed in the comprehensive early warning block after comprehensive judgment, indicating that the unit is in a mild early warning state. Figure 3

[0186] If the sulfur dioxide severely exceeds the standard (red), the nitrogen oxides moderately exceed the standard (orange), and the VOCs meet the standard (green), a black cross is displayed in the comprehensive early warning block, indicating that the unit is in a serious early warning state.

[0187] The beneficial effects of the above technical solutions are that the gas quality early warning chart is used to intuitively display the gas quality of each grid unit, and the comprehensive early warning block in the quality significant bar provides the overall early warning information of each unit, which helps the decision maker to quickly judge the severity of pollution and to comprehensively understand the state of various pollutants in a region, so as to more accurately evaluate the overall gas environment quality of the oil refinery and provide strong data support for formulating more scientific and reasonable environmental protection measures and production scheduling strategies.

[0188] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.​

Claims

1. A multi-intelligent sensor cooperative gas monitoring and analyzing system, characterized in that, The method comprises the following steps: A region splitting module is used to refine the functional attributes of the target factory to achieve region splitting, and to grid the atmospheric environment of the split region according to the current regional facilities, wherein the atmospheric environment grid result is a first refinement unit of the independent deviation closed contour in the split region and a second refinement unit of the remaining region except the independent deviation closed contour; An initial monitoring module is used to set fixed sensors and mobile sensors according to the current regional facilities of each division region and the grid unit, and to control the fixed sensors and mobile sensors to cooperatively monitor the current gas information of each grid unit to obtain an initial environment map, wherein the fixed sensors and mobile sensors constitute a multi-intelligent sensor; A frequency setting module is used to update the collection frequency of each fixed sensor and mobile sensor in each division region according to the initial environment map and the predicted meteorological data; A gas quality early warning module is used to receive new gas information collected by the fixed sensors and mobile sensors based on the edge node in communication connection with the fixed sensors and mobile sensors, to construct a gas quality early warning map of the target factory for outputting early warning; The region splitting module comprises: A function splitting unit is used to obtain a factory deployment structure by unmanned aerial vehicle monitoring of the target factory, to split the factory deployment structure based on a construction facility map of the target factory to obtain a plurality of split regions, wherein the construction facility map contains at least one functional attribute, and the split region is an actual split region; A set determination unit is used to compare and analyze the split region with a standard region to construct a deviation construction-position set, and to determine a plurality of independent deviation position contours; A contour supplement unit is used to input the position set of each independent deviation position contour into an adaptive contour filling model respectively to obtain an independent deviation closed contour by adaptive contour supplement, and to match the first refinement unit of the independent deviation position contour from a ratio-coefficient-setting comparison table according to the actual ratio of the contour length of the independent deviation closed contour to the region contour of the corresponding split region and the actual change coefficient of the facility type in the contour to the facility type in the corresponding standard region; A first grid unit is used to perform first grid on the corresponding independent deviation position contour in the split region according to the first refinement unit; A second grid unit is used to perform second grid on the remaining region in the split region except the independent deviation position contour according to the second refinement unit of the remaining region from a function-setting comparison table according to the region function.

2. The multi-intelligent sensor synergic gas monitoring and analyzing system according to claim 1, wherein, The first grid unit comprises: A complete judgment subunit is used to perform first grid on the corresponding independent deviation closed contour according to the first refinement unit, to judge whether there is an incomplete first refinement unit in the edge grid unit of the independent deviation closed contour; If not, the first grid unit of the corresponding independent deviation closed contour is kept unchanged; If yes, the peripheral facility attribute of the incomplete first refinement unit is determined. The function attribute judgment subunit is configured to retain the incomplete first refined unit if the peripheral facility attribute is irrelevant to the function attribute of the corresponding split region; The expansion subunit is configured to replace the incomplete first refined unit with the complete first refined unit if the connection line segment is a straight line segment; The expansion subunit is configured to replace the incomplete first refined unit with the complete first refined unit if the connection line segment is a straight line segment; The expansion subunit is configured to replace the incomplete first refined unit with the complete first refined unit if the connection line segment is a straight line segment; The expansion subunit is configured to replace the incomplete first refined unit with the complete first refined unit if the connection line segment is a straight line segment; The vector construction module is configured to construct an analysis vector according to the first position of each grid unit in the division region, the correlation between the function attribute of the division region and the unit facility of the corresponding grid unit extracted from the current region facility.

3. The multi-intelligent sensor synergic gas monitoring and analyzing system according to claim 1, wherein, The pollutant determination module is configured to input the analysis vector into a gas analysis model to obtain a gas pollution set of the corresponding grid unit, wherein the gas pollution set includes gas pollutants and diffusion factors of each gas pollutant. The sensor setting module is configured to find a first pollutant with a diffusion factor less than a preset factor, set a fixed sensor in the corresponding grid unit, and set a mobile sensor according to the diffusion path of each remaining pollutant in the unit facility of each grid unit. The initial monitoring module includes: The response set construction unit is configured to issue N times of synchronous working instructions to all sensors in each grid unit based on the controller, capture a first time point of each issued instruction and a second time point at which each sensor receives the corresponding issued instruction and starts working, and construct a response set S of each sensor.

4. The multi-intelligent sensor synergic gas monitoring and analyzing system according to claim 3, wherein, The matrix construction unit is configured to regard the time length difference between each response time length and a standard time length of the sensors of the same model in each response set S as a first set, sort the time length differences in each first set, and construct a difference matrix. The function determination unit is configured to dynamically divide the difference matrix to obtain a response function, and predict the feedback compensation time of the corresponding sensor. The environment map construction unit is configured to issue an (N+1)th instruction to the sensor based on the controller depending on the feedback compensation time, and output the current gas information of each grid unit to construct an initial environment map. The function determination unit includes: ​ 5. The multi-intelligent sensor synergic gas monitoring and analyzing system according to claim 4, wherein, ​ An initial partitioning subunit is configured to perform initial partitioning on the difference matrix to obtain a first matrix and a second matrix according to a preset time scale window, lock of mutation elements of each row vector in the difference matrix, wherein the number of columns of the first matrix is the maximum number of columns in the left row vector in the initial partitioning result, the number of columns of the second matrix is the maximum number of columns in the right row vector in the initial partitioning result, and when there is a column number of 0 in the left row vector or the right row vector, the same value is filled according to a new scale, when there is a column number that is not the maximum column number in the left row vector, the same value is backward filled according to the last element in the corresponding side row vector until the maximum column number is filled, and when there is a column number that is not the maximum column number in the right row vector, the same value is forward filled according to the first element in the corresponding side row vector until the maximum column number is filled; A secondary partitioning subunit is configured to determine a new scale window by adding the preset time scale window to the total number of historical uses and the decay duration after N synchronization operation instructions based on the controller, and perform secondary partitioning on the difference matrix to obtain a third matrix and a fourth matrix; A compensation determination subunit is configured to solve characteristic functions of the first matrix, the second matrix, the third matrix and the fourth matrix respectively, and calculate corresponding characteristic coefficients of each sensor in all characteristic functions to obtain feedback compensation time of the corresponding sensor.

6. The multi-intelligent sensor synergic gas monitoring and analyzing system according to claim 1, wherein, The frequency setting module comprises: A model analysis unit is configured to input the predicted meteorological data into a gas parameter influence analysis model, and input an influence coefficient of each gas parameter on the predicted meteorological data; A coefficient judgment unit is configured to keep the original acquisition frequency of the corresponding sensor unchanged if the influence coefficient is less than a preset coefficient, wherein the original acquisition frequency is the first acquisition frequency at this time. A frequency updating unit is configured to update the original acquisition frequency to obtain the first acquisition frequency according to the influence level of the influence coefficient and the preset coefficient, and in combination with the gas acquisition weight of the corresponding fixed sensor at this time. Meanwhile, the original acquisition frequency is updated to obtain the second acquisition frequency according to the influence level of the influence coefficient and the preset coefficient, and in combination with the gas acquisition weight of the corresponding mobile sensor based on the moving path of the corresponding predicted time period of the predicted meteorological data.

7. The multi-intelligent sensor synergic gas monitoring and analyzing system according to claim 1, wherein, The gas quality early warning module comprises: A significant bar construction unit is configured to determine the quality of each gas parameter in each grid unit according to the new gas information, and draw a quality significant bar based on the corresponding grid unit, wherein the quality significant bar comprises a quality color block of each gas parameter and a comprehensive early warning block. An early warning map construction unit is configured to construct a gas quality early warning map based on the quality significant bars of all grid units and output the map.

Images