CN-121980499-A - Multi-source olfactory data space-time consistency self-adaptive acquisition method and system

CN121980499ACN 121980499 ACN121980499 ACN 121980499ACN-121980499-A

Abstract

The invention provides a multisource olfactory data space-time consistency self-adaptive collection method and a system, which relate to the technical field of olfactory information acquisition and processing, and comprise a multisource olfactory data access module which is used for accessing olfactory data streams from olfactory collection nodes and acquiring space-time identification information corresponding to the olfactory data streams, the olfactory collection nodes comprise fixed collection nodes and mobile collection nodes, and the multisource olfactory data space-time consistency self-adaptive collection method and system solve the problem that the multisource olfactory data collection lacks comparability and reproducibility among different collection nodes by introducing space-time consistency evaluation indexes. By carrying out real-time monitoring and feedback on the time-space consistency, the method can automatically adjust acquisition parameters such as sampling frequency, reporting frequency and the like, ensure the data consistency among all nodes, avoid the influence of data inconsistency on the subsequent analysis and early warning effect, and improve the quality and reliability of the data.

Inventors

ZHU XIAODONG
ZHAO MING
ZHOU YI
CHEN SHUAIDONG
SUN MING
XIA YUHANG
ZHANG YIHANG
SUN KE
SHEN WENFENG

Assignees

中科微感(宁波)科技有限公司

Dates

Publication Date: 20260505
Application Date: 20260119

Claims (10)

1. A multi-source olfactory data spatiotemporal consistency adaptive collection system, comprising: The system comprises a multisource olfactory data access module, a sensor module and a sensor module, wherein the multisource olfactory data access module is used for accessing an olfactory data stream from an olfactory acquisition node and acquiring space-time identification information corresponding to the olfactory data stream, the olfactory acquisition node comprises a fixed acquisition node and a mobile acquisition node, and the space-time identification information comprises time stamp information and space position identification information; The space-time consistency index quantification module is used for calculating space-time consistency evaluation indexes for representing comparability and reproducibility of cross-node olfactory data based on the olfactory data stream and the space-time identification information, wherein the space-time consistency evaluation indexes comprise time consistency indexes and space consistency indexes; The consistency deviation judging module is used for comparing the space-time consistency evaluation index with a preset consistency criterion and outputting a consistency deviation result, wherein the consistency deviation result comprises a deviation type identifier and a deviation degree parameter, and the deviation type identifier comprises a time deviation type and a space deviation type; The self-adaptive acquisition regulation and control module is used for generating an acquisition regulation and control instruction based on the consistency deviation result, and transmitting the acquisition regulation and control instruction to a corresponding olfactory acquisition node to perform on-line adjustment on acquisition parameters of the olfactory acquisition node, wherein the acquisition parameters comprise sampling frequency, reporting frequency and triggering threshold, the on-line adjustment is performed so that the space-time consistency evaluation index is not lower than a preset index threshold and is a consistency target, and the adjustment quantity of the acquisition parameters is determined under the condition that a preset resource constraint condition is met; And the closed loop updating module is used for re-executing the calculation of the space-time consistency evaluation index and the consistency deviation judgment on the updated olfactory data stream after the acquisition regulation instruction is effective, and outputting the olfactory data stream meeting the preset consistency criterion as a consistency enhancement data set.
2. The system of claim 1, wherein the time consistency index is obtained by performing alignment calculation on a smell event feature sequence extracted from the smell data stream in the same sliding time window, the alignment calculation adopts dynamic time alignment, a time offset and an alignment similarity are output through the alignment calculation, and the alignment similarity and the time offset are normalized and combined to obtain the time consistency index.
3. The system of claim 1, wherein the spatial consistency index is calculated after a spatial adjacency relation is constructed, the spatial adjacency relation is determined by adjacency rules with a distance between nodes not larger than a preset distance threshold and neighbor rules based on a preset number of neighbors, the spatial consistency index comprises a spatial gradient consistency metric and a neighborhood residual consistency metric, the spatial gradient consistency metric is calculated based on the consistency of the change direction of the olfactory data of adjacent nodes, and the neighborhood residual consistency metric is calculated based on residual statistics between observed values of the adjacent nodes and predicted values of spatial interpolation.
4. The system for adaptively collecting the spatio-temporal consistency of the multisource olfactory data according to claim 1, wherein the spatio-temporal consistency evaluation index is a comprehensive evaluation index and is obtained by the time consistency index and the space consistency index according to a preset fusion rule, and the fusion rule comprises respectively carrying out normalization processing on the time consistency index and the space consistency index and carrying out weighted fusion according to preset weights.
5. The system of claim 4, wherein the determining of the preset weights includes searching a candidate weight set for a weight value that maximizes a ratio of the spatiotemporal consistency evaluation index to a preset index threshold or minimizes a statistical value of a consistency deviation degree parameter based on a historical running data set, and taking the obtained weight value as the preset weight.
6. The multi-source olfactory data space-time consistency self-adaptive acquisition system of claim 1, wherein the adjustment amount of the acquisition parameters is determined by adopting an incremental updating rule, the incremental updating rule meets the condition that a monotonic corresponding relation exists between the adjustment amount of the acquisition parameters and the deviation degree parameters, dead zone constraint, limiting constraint and speed limiting constraint are applied to the adjustment amount, the dead zone constraint is that parameter updating is not triggered when the deviation degree parameters are lower than a preset dead zone threshold, the limiting constraint is that the maximum adjustment amplitude of the adjustment amount is updated for a single time, and the speed limiting constraint is that a minimum time interval between two adjacent acquisition regulation and control instruction issuing is limited.
7. The system of claim 1, wherein the resource constraint conditions include a terminal power consumption constraint, a communication bandwidth constraint, a storage occupancy constraint, and a terminal computational load constraint, the resource constraint conditions are configured in a budget form and participate in the adjustment amount determination, and the budget includes at least a unit time energy consumption budget, a unit time communication data amount budget, a local cache occupancy budget, and a terminal computational load budget.
8. The system of claim 1, wherein the adaptive acquisition control module is configured to increase a sampling frequency or a reporting frequency and to lower a trigger threshold when the olfactory data stream satisfies an odor event trigger condition, and to decrease the sampling frequency or the reporting frequency and to raise the trigger threshold when the olfactory data stream does not satisfy the odor event trigger condition and the spatiotemporal consistency evaluation index is not lower than a preset index threshold, wherein the odor event trigger condition includes a concentration change rate exceeding a preset change rate threshold and an accumulated change amount within a preset time window exceeding a preset accumulated threshold.
9. The system of claim 1 wherein the closed loop update module terminates closed loop iterations based on a convergence criterion, the convergence criterion comprising that the spatio-temporal consistency evaluation index is not lower than a predetermined index threshold for N consecutive predetermined iterations, and the maximum fluctuation amplitude of the spatio-temporal consistency evaluation index in the N iterations does not exceed a predetermined fluctuation threshold, and the consistency enhancement dataset is output after the convergence criterion is satisfied.
10. A method for adaptively acquiring space-time consistency of multisource olfactory data, which is realized by a multisource olfactory data space-time consistency adaptive acquisition system according to any one of claims 1to 9, and is characterized by comprising the following steps: S1, accessing olfactory data streams from a plurality of olfactory acquisition nodes, and acquiring space-time identification information corresponding to the olfactory data streams, wherein the space-time identification information comprises time stamp information and space position identification information; S2, calculating space-time consistency evaluation indexes based on the olfactory data stream and the space-time identification information, wherein the space-time consistency evaluation indexes comprise time consistency indexes and space consistency indexes; s3, comparing the space-time consistency evaluation index with a preset consistency criterion, and outputting a consistency deviation result, wherein the consistency deviation result comprises a deviation type identifier and a deviation degree parameter, and the deviation type identifier comprises a time deviation type and a space deviation type; S4, generating an acquisition regulation and control instruction based on the consistency deviation result, and transmitting the acquisition regulation and control instruction to a corresponding olfactory acquisition node to perform online adjustment on acquisition parameters of the olfactory acquisition node, wherein the acquisition parameters comprise sampling frequency, reporting frequency and a trigger threshold, the online adjustment is performed so that the space-time consistency evaluation index is not lower than a preset index threshold to be a consistency target, and the adjustment quantity of the acquisition parameters is determined under the condition that a preset resource constraint condition is met; S5, after the acquisition regulation instruction takes effect, re-executing the step S2 and the step S3 on the updated olfactory data stream to form a closed-loop self-adaptive acquisition process, and outputting the olfactory data stream meeting the preset consistency criterion as a consistency enhancement data set.

Description

Multi-source olfactory data space-time consistency self-adaptive acquisition method and system Technical Field The invention relates to the technical field of olfactory information acquisition and processing, in particular to a multi-source olfactory data space-time consistency self-adaptive acquisition method and system. Background In the prior art, with the development of electronic noses, gas sensing arrays, volatile organic compounds monitoring and internet of things communication technologies, researchers and engineering application departments have started to arrange multipoint olfactory monitoring systems in scenes such as factory workshops, park perimeters, urban roads, garbage/sewage treatment plants and the like, collect environmental odor information through fixed monitoring terminals, mobile inspection terminals and gas sensors of various principles, and upload multisource olfactory data to an edge side or cloud platform through a wired/wireless network for application such as gas leakage monitoring, environmental odor supervision, industrial process gas quality control, public safety early warning and the like. And the partial system also combines the auxiliary environmental parameters such as temperature, humidity, wind speed, wind direction, pressure and the like to perform time sequence analysis and space distribution estimation on the collected smell data so as to support subsequent source item positioning, anomaly identification and risk assessment. However, the existing multisource olfactory data acquisition scheme generally regards each acquisition terminal as an independent sampling point, and lacks a systematic space-time consistency constraint and self-adaptive regulation mechanism, so that a series of common defects are exposed in multisource cooperative application, namely, a quantifiable space-time consistency evaluation index is not introduced into an acquisition side and a closed loop for evaluation and regulation is formed, so that the comparability and reproducibility of multisource olfactory data in different scenes and different time periods are insufficient, and the demands of fine tracing and stable early warning are difficult to meet. Disclosure of Invention Aiming at the defects of the prior art, the invention provides a multi-source olfactory data space-time consistency self-adaptive acquisition method and a system, and the technical problem to be solved by the invention is how to solve the problem that the multi-source olfactory data lacks comparability and reproducibility in different scenes and different time periods through self-adaptive acquisition regulation and control based on space-time consistency evaluation indexes. The multi-source olfactory data space-time consistency self-adaptive acquisition system comprises a multi-source olfactory data access module, a data acquisition module and a data acquisition module, wherein the multi-source olfactory data access module is used for accessing an olfactory data stream from an olfactory acquisition node and acquiring space-time identification information corresponding to the olfactory data stream, the olfactory acquisition node comprises a fixed acquisition node and a movable acquisition node, and the space-time identification information comprises time stamp information and space position identification information. The space-time consistency index quantification module is used for calculating space-time consistency evaluation indexes for representing comparability and reproducibility of the inter-node olfactory data based on the olfactory data stream and the space-time identification information, and the space-time consistency evaluation indexes comprise time consistency indexes and space consistency indexes. The consistency deviation judging module is used for comparing the space-time consistency evaluation index with a preset consistency criterion and outputting a consistency deviation result, wherein the consistency deviation result comprises a deviation type identifier and a deviation degree parameter, and the deviation type identifier comprises a time deviation type and a space deviation type. The self-adaptive acquisition regulation and control module is used for generating an acquisition regulation and control instruction based on the consistency deviation result and transmitting the acquisition regulation and control instruction to a corresponding olfactory acquisition node to perform on-line adjustment on acquisition parameters of the olfactory acquisition node, wherein the acquisition parameters comprise sampling frequency, reporting frequency and triggering threshold, the on-line adjustment is performed so that the space-time consistency evaluation index is not lower than a preset index threshold and is a consistency target, and the adjustment quantity of the acquisition parameters is determined under the condition that a preset resource constraint condition is met. And the closed loop updating module is