A three-dimensional fluorescent water quality fingerprint distributed collection method based on flow table rules

Abstract

This invention discloses a distributed acquisition method for three-dimensional fluorescent water quality fingerprints based on flow table rules, relating to the field of intelligent dimming technology for LED lights. First, the target optical state is determined according to the dimming target, and channel control quantities are generated. When the target is not fully realized due to the minimum stable pulse width constraint, a target residual vector is generated. After a new dimming target arrives, a channel contribution basis is constructed based on the new target optical state, and the target residual vector is migrated and allocated to redetermine the channel control quantities for subsequent PWM cycles, thereby improving the brightness stability and overall color continuity of multi-channel dimming.

Description

Technical Field

[0001] This application relates to the field of water quality monitoring technology, and in particular to a three-dimensional fluorescent water quality fingerprint distributed acquisition method based on flow table rules. Background Technology

[0002] In existing water environment monitoring scenarios, multiple water quality sampling nodes can typically be deployed at various monitoring locations, including drainage pipe networks, river tributaries, outfall sections, and pollution propagation paths, to achieve continuous monitoring of water samples from different locations. Some monitoring schemes employ three-dimensional fluorescence technology to collect the full spectrum of water samples and utilize target peak regions, peak intensity relationships, or similarity features in the three-dimensional fluorescence water quality fingerprint to identify and track pollution events.

[0003] In practical applications, some pollution events are characterized by short duration, rapid propagation, and rapid decay. This is especially true in transient pollution scenarios such as intermittent emissions, short-term flushing emissions, and misdirected backflow, where pollution plumes often spread along pipelines or branches within a short period. If the approach of having a central platform perform the analysis first and then uniformly issue high-level sampling commands to multiple adjacent nodes is still adopted, the pollution plume may have already passed through neighboring nodes by the time the command takes effect, making it difficult for subsequent nodes to obtain continuous evidence. On the other hand, maintaining a long-term full-spectrum high-frequency sampling state at multiple nodes to avoid missed detections would significantly increase the node acquisition, transmission, and analysis loads.

[0004] Furthermore, most existing extended sampling methods directly expand the sampling range upon triggering. For example, after the first node detects an anomaly, multiple adjacent nodes are immediately required to enter a high-level sampling mode. While this approach can improve the capture probability, it is prone to problems such as excessive invalid expansion, high node load, and low convergence efficiency in propagation directions in scenarios with multiple adjacent branches, multiple scalable directions, or strong background disturbances.

[0005] Therefore, how to quickly wake up neighboring nodes with low node burden after the first node detects an anomaly, and how to expand sampling step by step when it is confirmed that the contamination continues to spread, and how to back down in time when the confirmation fails, has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0006] This application provides a three-dimensional fluorescence water quality fingerprinting distributed acquisition method based on flow table rules, which can balance the timeliness of pollution event capture and the operating load of distributed acquisition nodes in transient pollution propagation scenarios, and reduce the problems of excessive acquisition load, many invalid extensions and difficulty in forming continuous evidence caused by directly triggering multi-level node full-spectrum high-frequency sampling.

[0007] Firstly, this application provides a distributed acquisition method for three-dimensional fluorescence water quality fingerprints based on flow table rules. This method can be executed by a management device. In this method, the management device obtains event summary information sent by the first node, which is generated by the first node when its acquired three-dimensional fluorescence water quality fingerprint matches a target event rule. The management device determines at least one single-hop node directly adjacent to the first node from multiple acquisition nodes based on pre-stored node topology information. The management device sends a pre-wake-up flow table rule corresponding to the event summary information to each of the at least one single-hop node, so that each single-hop node performs a local peak area pre-scan of the target peak area according to the pre-wake-up flow table rule within the expected propagation window. After the local peak area pre-scan result of the target single-hop node meets the preset confirmation conditions, the management device sends an upgrade acquisition instruction to the target single-hop node and sends the updated pre-wake-up flow table rule corresponding to the event summary information to at least one double-hop node directly adjacent to the target single-hop node.

[0008] In the above method, after the first node hits the target event, the management device does not directly trigger the multi-level nodes to enter the full-spectrum high-frequency sampling state. Instead, it first pre-wakes up the single-hop node based on the event summary information, and the single-hop node performs a local peak pre-scan within the expected propagation window. Only when the local peak pre-scan result meets the preset confirmation conditions is the sampling mode of the target single-hop node upgraded and extended to the double-hop node. This scheme can improve the probability of neighboring nodes capturing the target event when the transient contamination propagation time is short, while reducing the node load and transmission load caused by invalid full-spectrum sampling.

[0009] In one possible design, obtaining the event summary information sent by the first node includes: obtaining event summary information extracted by the first node based on the full-spectrum acquisition results corresponding to the target event; wherein, the event summary information includes at least one of the following: event chain identifier, target peak region identifier, peak intensity ratio feature, anomaly level, first node identifier, and expected propagation time window. Through this design, the management device can utilize the key fingerprint features of the first node to construct the pre-wake-up basis for subsequent nodes, reducing the amount of transmitted data while retaining the core discriminative features of the target event.

[0010] In one possible design, when the at least one single-hop node includes multiple single-hop nodes, sending the pre-wake-up flow table rule corresponding to the event digest information to each of the at least one single-hop node includes: sending the pre-wake-up flow table rule corresponding to the event digest information to each of the multiple single-hop nodes in descending order of priority. Through this design, the management device can prioritize processing nodes more likely to capture the target event when multiple candidate single-hop nodes exist, thereby optimizing the usage order of distributed acquisition resources.

[0011] In one possible design, the priority of the plurality of single-hop nodes is related to the expected propagation time corresponding to each single-hop node; wherein, the shorter the expected propagation time, the higher the priority of the single-hop node corresponding to the expected propagation time. Through this design, the management device can prioritize waking up single-hop nodes that are expected to receive the contamination plume faster, thereby improving response time in transient contamination propagation scenarios.

[0012] In one possible design, the priority of the plurality of single-hop nodes is related to the branch risk information corresponding to the plurality of single-hop nodes; wherein, the higher the risk level indicated by the branch risk information, the higher the priority of the single-hop node corresponding to the branch risk information. Through this design, the management device can prioritize issuing pre-wake-up flow table rules to single-hop nodes corresponding to high-risk branches, thereby improving the collection coverage in high-risk propagation directions.

[0013] In one possible design, the pre-wake-up flow table rule includes a matching field and an action field; wherein, the matching field includes a target peak region identifier, a matching interval corresponding to the peak intensity ratio feature, and an expected propagation window; the action field includes entering the pre-wake-up state and performing a local peak region pre-scan. With this design, the management device can perform lightweight pre-wake-up control on nodes through flow table rules, allowing nodes to perform only a local pre-scan around the target peak region without directly switching to a high-load sampling mode.

[0014] In one possible design, the preset confirmation conditions include at least one of the following: the target peak region position offset in the local peak region pre-scan result is within a preset tolerance range; the peak intensity ratio in the local peak region pre-scan result is within a preset ratio range; and the target single-hop node obtains local peak region pre-scan results that meet the matching conditions at least twice consecutively within the expected propagation window. This design improves the stability of single-hop nodes confirming target events and reduces the probability of false triggering caused by occasional noise or background fluctuations.

[0015] In one possible design, sending the upgrade acquisition instruction information to the target single-hop node includes: instructing the target single-hop node to switch from a pre-wake-up state to a confirmation state; instructing the target single-hop node to switch from a local peak pre-scan mode to a full-spectrum acquisition mode or a high-frequency sampling mode; and updating the event summary information based on the confirmation result of the target single-hop node. With this design, the management device can upgrade the sampling level of the target single-hop node after a second confirmation is established, and generate the summary information required for subsequent expansion based on the new confirmation result.

[0016] In one possible design, when the at least one two-hop node comprises multiple two-hop nodes, the step of sending the pre-wake-up flow table rule corresponding to the updated event digest information to the at least one two-hop node directly adjacent to the target single-hop node includes: sending the pre-wake-up flow table rule corresponding to the updated event digest information to each of the multiple two-hop nodes in descending order of priority. Through this design, the management device can maintain orderly expansion as it expands to outermost nodes, thereby reducing resource waste caused by simultaneous expansion of multiple branches.

[0017] In one possible design, the method further includes: when a node in a pre-wake-up state at any level fails to meet the preset confirmation condition within the corresponding expected propagation window, the activation state of the pre-wake-up flow table rule corresponding to that node is revoked, restoring the node to a normal acquisition state and terminating subsequent expansion in the propagation direction of that node; wherein the pre-wake-up flow table rule is generated by activating a pre-loaded dormant rule template, the dormant rule template including a matching field, an action field, and a timeout field. Through this design, the management device can automatically end the expansion in the corresponding direction when the target event is not confirmed to continue propagating, avoiding nodes maintaining an invalid pre-wake-up state for extended periods.

[0018] Based on the above design, the method provided in this application embodiment can, after an anomaly is detected at the first node, use event summary information as the trigger to perform local peak area pre-scanning pre-wake-up on single-hop nodes, and upgrade sampling and extend to double-hop nodes after confirmation, and automatically back off when confirmation fails, thereby forming a hierarchical collection mechanism for transient pollution propagation scenarios.

[0019] The technical effects and advantages of the three-dimensional fluorescence water quality fingerprint distributed acquisition method based on flow table rules of this invention are as follows: Compared with existing schemes that directly trigger multiple nodes to enter high-level sampling after an anomaly is detected at the first node, this application uses a hierarchical acquisition mechanism to enable neighboring nodes to respond quickly to target events within the expected propagation window without having to maintain a full-spectrum high-frequency sampling state for a long time. The sampling range is expanded step by step when it is confirmed that the pollution continues to propagate, and the expansion in the corresponding direction is terminated in time when it is not confirmed. This can improve the event capture probability in transient pollution propagation scenarios, reduce the acquisition load, transmission load and analysis load caused by invalid expansion, and improve the convergence efficiency and continuous evidence formation capability of pollution propagation path identification. Attached Figure Description

[0020] Figure 1 A schematic diagram of the architecture of a three-dimensional fluorescence water quality fingerprint distributed acquisition system based on flow table rules provided in this application embodiment; Figure 2A schematic diagram of an event digest-driven hierarchical expansion provided in an embodiment of this application; Figure 3 A flowchart illustrating a distributed acquisition method for three-dimensional fluorescent water quality fingerprints based on event digest hierarchical wake-up provided in this application embodiment. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the embodiments of this application will be further described in detail below with reference to the accompanying drawings. In the description of the embodiments of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first" and "second" may explicitly or implicitly include one or more of that feature.

[0022] It should be understood that in the embodiments of this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects have an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items.

[0023] Figure 1 This is a schematic diagram of the architecture of a distributed three-dimensional fluorescence water quality fingerprint acquisition system based on flow table rules, provided as an embodiment of this application. (Reference) Figure 1 The system may include a management device, a head node, multiple single-hop nodes, and multiple double-hop nodes. The management device is used to maintain node topology information, rule template information, and event chain information, and to issue corresponding flow table rules or collection instruction information to each collection node. The head node, single-hop nodes, and double-hop nodes can all be water quality collection nodes deployed in drainage pipe networks, river branches, outlets, or monitoring sections.

[0024] In this embodiment, the management device can be an edge device, a server, or a cloud service implemented by a server; when the management device is a server or a cloud service, it can also be called a management platform. The management device can be used to manage information such as the device status, rule status, collection mode, node topology relationship, and event chain records of each collection node in the system.

[0025] The first node can be the first acquisition node to detect the target event. The first node can perform three-dimensional fluorescence full-spectrum acquisition on the current water sample and determine whether the current water sample has detected the target event based on preset anomaly rules. When the target event is detected, the first node can extract the corresponding event summary information and send it to the management device. The management device can determine the single-hop nodes directly adjacent to the first node, and the double-hop nodes directly adjacent to the single-hop nodes, based on pre-stored node topology information.

[0026] Figure 2 This diagram illustrates a hierarchical expansion driven by event summary, as provided in an embodiment of this application. In this diagram, after detecting a target event, the first node does not directly trigger multi-level nodes to perform full-spectrum high-frequency sampling. Instead, it first reports event summary information to the management device. The management device prioritizes issuing pre-wake-up flow table rules to single-hop nodes. Single-hop nodes perform local peak pre-scanning within the expected propagation window. When a single-hop node confirms that the target event continues to propagate, the management device instructs the single-hop node to upgrade its sampling mode and continue expanding to double-hop nodes.

[0027] In existing transient pollution monitoring schemes, if the analysis is directly performed by a central platform before high-level sampling commands are issued, the target pollution plume may have already passed through neighboring nodes by the time the command takes effect. Furthermore, maintaining multiple levels of nodes in a full-spectrum, high-frequency sampling state for an extended period significantly increases the operational load on the distributed acquisition nodes. Therefore, this application provides a distributed acquisition method for three-dimensional fluorescence water quality fingerprints based on event summary-based hierarchical wake-up.

[0028] Figure 3 A flowchart illustrating a distributed acquisition method for three-dimensional fluorescence water quality fingerprints based on event digest hierarchical wake-up provided in this application is shown below. Figure 3 The method may include the following steps: S101: The management device obtains the event summary information sent by the first node.

[0029] The event summary information is generated by the first node when its collected 3D fluorescence water quality fingerprint matches the target event rule. The first node can extract the event summary information based on the full-spectrum acquisition results corresponding to the target event and send the event summary information to the management device.

[0030] To enable those skilled in the art to clearly understand the target event determination process in the embodiments of this application, the implementation method of the target event rules is further described below. In the embodiments of this application, after the first node performs three-dimensional fluorescence full-spectrum acquisition on the current water sample, it can obtain the excitation-emission matrix data corresponding to the current water sample. The management device or the first node can determine the excitation-emission matrix data based on the preset target event rules. The target event rules can be pre-constructed based on historical abnormal water sample data, standard pollution source sample data, and normal background water sample data in the field. For example, the target event rules may include at least one of the following: a set of target peak regions, the peak region location range corresponding to each target peak region, a peak region response threshold, a peak intensity ratio range, an anomaly level threshold, and an event category identifier.

[0031] In one optional implementation, the management device can pre-store multiple target event templates, with different target event templates corresponding to different pollution types or different abnormal scenarios. Each target event template can include at least two target peak regions. For example, target event template E1 can include a first target peak region P1 and a second target peak region P2, wherein P1 corresponds to an excitation wavelength range of 225 nm to 235 nm and an emission wavelength range of 335 nm to 355 nm; P2 corresponds to an excitation wavelength range of 270 nm to 285 nm and an emission wavelength range of 430 nm to 455 nm. For any target event template, if the peak intensity of the current water sample in the corresponding target peak region is higher than its respective response threshold, and the peak intensity ratio between each target peak region is within a preset ratio range, then the current water sample is determined to have matched the corresponding target event rule.

[0032] In one optional implementation, let the peak intensities of the current water sample in target peak regions P1 and P2 be I1 and I2, respectively. Then, the peak intensity ratio feature R can be calculated, R = I1 / I2. If I1 ≥ T1, I2 ≥ T2, and R is within the interval [Rmin, Rmax], then the corresponding target event rule is determined to have been matched. T1, T2, Rmin, and Rmax can be determined from historical abnormal sample statistics or configured by on-site maintenance personnel based on the application scenario. For example, for target event template E1, T1 = 160, T2 = 90, Rmin = 1.20, and Rmax = 2.40 can be set.

[0033] After the first node matches the target event rule, it can generate event summary information. This event summary information may include an event chain identifier, a target peak region identifier, the center location of the peak region, the peak region width, peak intensity ratio characteristics, anomaly level, the first node identifier, and the sampling timestamp corresponding to the first node. The event chain identifier can be a string identifier, a combination of a timestamp and a node identifier, a universally unique identifier, or other identifiers capable of distinguishing different pollution propagation events.

[0034] In one optional implementation, the event summary information may further include initial calculation parameters related to the expected propagation window. Upon receiving the event summary information, the management device can use it as the basis for subsequent nodes to perform pre-wake-up and secondary confirmation.

[0035] S102: The management device identifies at least one single-hop node and sends the pre-wake-up flow table rule corresponding to the event digest information to the at least one single-hop node.

[0036] In this embodiment, the management device can determine at least one single-hop node directly adjacent to the first node from multiple acquisition nodes based on pre-stored node topology information. The management device then sends a pre-wake-up flow table rule corresponding to the event digest information to each of the at least one single-hop node, so that each single-hop node enters the pre-wake-up state within the expected propagation window.

[0037] To enable those skilled in the art to clearly understand the process of determining the expected propagation time window, it is further explained below. For the first node i and its directly adjacent single-hop node j, the management device can obtain the pipe segment length Lij, average flow velocity vij, and residence correction time δij between node i and node j, and calculate the expected arrival time of the contaminant plume from the first node to the single-hop node according to the formula t_ij=Lij / vij+δij, where t_ij represents the expected propagation time. Further, the management device can generate an expected propagation time window [t_ij-Δ1,t_ij+Δ2] around the expected arrival time, where Δ1 and Δ2 represent the advance tolerance time and lag tolerance time, respectively.

[0038] In one optional implementation, the average flow velocity vij can be determined by online flow meters, level gauges, pump station operating status, or historical statistical data; the residence correction time δij can be determined based on the number of bends between nodes, local water accumulation areas, storage structures, pump station start-up and shutdown status, or historical propagation delay statistics of similar events. For example, when Lij = 120 meters, vij = 0.60 meters per second, δij = 60 seconds, Δ1 = 40 seconds, and Δ2 = 60 seconds, the expected propagation time t_ij of the pollution plume from the first node to the single-hop node is 260 seconds, and the corresponding expected propagation window is from the 220th to the 320th second after the first node is triggered.

[0039] In one optional implementation, when at least one single-hop node includes multiple single-hop nodes, the management device can determine the priority of the multiple single-hop nodes and send pre-wake-up flow table rules in descending order of priority. For example, the priority of the multiple single-hop nodes can be related to the expected propagation time corresponding to the multiple single-hop nodes; the shorter the expected propagation time, the higher the priority of the corresponding single-hop node. Alternatively, the priority of the multiple single-hop nodes can be related to the tributary risk information corresponding to the multiple single-hop nodes; the higher the risk level indicated by the tributary risk information, the higher the priority of the corresponding single-hop node. Further, the tributary risk information can be determined by at least one of the following: the frequency of historical abnormal events, the key outfall level, the functional zone level to which the tributary belongs, or a manually configured risk coefficient.

[0040] In this embodiment, pre-wake-up flow table rules can be generated by activating a pre-loaded dormant rule template. The dormant rule template can be pre-stored in a single-hop node or management device and includes a matching field, an action field, a priority field, and a timeout field. After receiving the event summary information from the first node, the management device can fill the dormant rule template with the target peak identifier, peak center location, peak intensity ratio characteristics, and expected propagation window from the event summary information, thereby generating a pre-wake-up flow table rule corresponding to the target event, and then sending or activating it to the corresponding single-hop node.

[0041] In one optional implementation, the matching field may include a target peak region identifier, an excitation emission window corresponding to each target peak region, a peak intensity ratio range, and an expected propagation time window; the action field may include entering a pre-wake-up state, initiating a local peak region pre-scan task, outputting pre-scan results at a preset period, and exiting the pre-wake-up state upon confirmation failure; the priority field may be used to ensure that the pre-wake-up flow table rule is executed before the regular rules within the expected propagation time window when regular sampling rules already exist on the node; the timeout field may be used to automatically cancel the activation state of the pre-wake-up flow table rule after the expected propagation time window has expired. A single-hop node may include a rule execution module, which reads the matching field and action field from the pre-wake-up flow table rule and controls the node acquisition module to enter the pre-wake-up state and execute a local peak region pre-scan.

[0042] S103: The single-hop node performs a local peak area pre-scan according to the pre-wake-up flow table rules, and the management device performs a secondary confirmation based on the pre-scan results.

[0043] In this embodiment, after receiving the pre-wake-up flow table rule, the single-hop node does not directly switch to full-spectrum high-frequency sampling, but instead performs a local peak region pre-scan within the expected propagation window. The local peak region pre-scan can be understood as performing rapid acquisition only on the excitation-emission window near the target peak region. For example, if the center of the target peak region P1 is located at an excitation wavelength of 230 nm and an emission wavelength of 345 nm, the single-hop node can perform a pre-scan within a local window of excitation wavelengths from 224 nm to 236 nm and emission wavelengths from 337 nm to 353 nm; if the center of the target peak region P2 is located at an excitation wavelength of 278 nm and an emission wavelength of 442 nm, the single-hop node can perform a pre-scan within a local window of excitation wavelengths from 272 nm to 284 nm and emission wavelengths from 434 nm to 450 nm.

[0044] In one optional implementation, the single-hop node can perform multiple local peak region pre-scans at preset time intervals within the expected propagation window. For example, the single-hop node can perform a local peak region pre-scan every 20 seconds within the expected propagation window, for a total of 3 to 5 times. Each local peak region pre-scan can employ a fast stepping strategy larger than the full-spectrum scan step size, or it can adopt a method of acquiring only a few discrete sampling points near the center of the target peak region, thereby reducing acquisition time and node load. If the single-hop node itself does not support a complete windowed local spectrum scan, a preset discrete excitation emission point acquisition method can also be used to quickly detect the target peak region.

[0045] After completing the local peak region pre-scan, the single-hop node can send the local peak region pre-scan result to the management device. The management device can perform secondary confirmation based on preset confirmation conditions. For example, the preset confirmation conditions may include at least one of the following: First, the target peak region center offset in the local peak region pre-scan result is less than or equal to a preset offset threshold; second, the peak intensity ratio feature in the local peak region pre-scan result falls within a preset ratio range; third, the single-hop node obtains local peak region pre-scan results that meet the matching conditions at least twice consecutively within the expected propagation time window.

[0046] In one optional implementation, if the center position of the target peak region P1 in the first node event summary information is at an excitation wavelength of 230 nm and an emission wavelength of 345 nm, the management device can set the offset threshold to be no greater than 4 nm in the excitation direction and no greater than 6 nm in the emission direction; if the first node value of the peak intensity ratio feature R is 1.70, the confirmation ratio range of the single-hop node can be set to [1.45, 1.95]. Further, if the single-hop node obtains local peak region pre-scan results that satisfy the above-mentioned peak position offset and peak intensity ratio conditions twice consecutively within the expected propagation window, the management device can identify the single-hop node as the target single-hop node.

[0047] S104: After the target single-hop node meets the preset confirmation conditions, the management device sends upgrade collection instruction information to the target single-hop node and sends the pre-wake-up flow table rules corresponding to the updated event summary information to at least one double-hop node.

[0048] In this embodiment, after the target single-hop node meets the preset confirmation conditions, the management device can send upgrade acquisition instruction information to the target single-hop node. This upgrade acquisition instruction information can be used to instruct the target single-hop node to switch from a pre-wake-up state to a confirmation state, and from a local peak pre-scan mode to a full-spectrum acquisition mode or a high-frequency sampling mode.

[0049] In one optional implementation, the management device can instruct the target single-hop node to perform three-dimensional fluorescence full-spectrum acquisition at a frequency of once every 30 seconds or once every 60 seconds within a subsequent preset duration, and continuously send the full-spectrum acquisition results to the management device.

[0050] In this embodiment, the management device can also update the event summary information based on the confirmation result of the target single-hop node. For example, the management device can calculate the peak center position in the updated event summary information by weighting the results of the first node and the target single-hop node. Let the peak center corresponding to the first node be C0, and the peak center corresponding to the target single-hop node be C1. Then, the updated peak center Cnew can be calculated using the formula Cnew = αC0 + (1 - α)C1, where α is a weighting coefficient between 0 and 1. The management device can also recalculate the expected propagation window between the target single-hop node and the double-hop node, and write the updated peak center position, peak intensity ratio characteristics, expected propagation window, and event chain identifier into the new event summary information. The updated event summary information includes at least one update result from the node identifier, expected propagation window, and peak characteristics.

[0051] Based on the updated event summary information, the management device can send updated pre-wake-up flow table rules to at least one two-hop node directly adjacent to the target single-hop node. This allows the two-hop nodes to perform local peak pre-scanning in the same or similar manner as the single-hop nodes. Only after meeting preset confirmation conditions are the two-hop nodes further upgraded to full-spectrum acquisition mode. Thus, contamination propagation events can extend tier by tier along the node chain without simultaneously activating high-load sampling modes on multiple levels of nodes as soon as the first node detects an anomaly.

[0052] In one optional implementation, when at least one two-hop node includes multiple two-hop nodes, the management device can send the updated event digest information corresponding to the pre-wake-up flow table rules to the multiple two-hop nodes in descending order of their priority. The priority of the multiple two-hop nodes can be determined based on the expected propagation duration, tributary risk information, or other parameters related to propagation spread.

[0053] S105: If a node in the pre-wake-up state fails to meet the preset confirmation conditions within the corresponding expected propagation window, the management device cancels the activation state of the corresponding pre-wake-up flow table rule and terminates the subsequent expansion in that propagation direction.

[0054] In this embodiment of the application, when any node in the pre-wake-up state fails to meet the preset confirmation conditions within the corresponding expected propagation time window, the management device can cancel the activation state of the pre-wake-up flow table rule corresponding to the node, restore the node to the normal collection state, and terminate the subsequent expansion in the propagation direction where the node is located.

[0055] In one alternative implementation, the management device can switch the status bit in the pre-wake-up flow table rule from an active state to an inactive state, or delete the rule from the node-side rule table, and terminate the subsequent expansion in the propagation direction of the node.

[0056] In this embodiment, the management device can also record the pre-wake-up, local peak pre-scan, secondary confirmation, upgraded sampling, summary forwarding, and rollback termination processes corresponding to each node, to generate a hierarchical acquisition event chain record associated with the event chain identifier. This event chain record can be used to subsequently reconstruct the contamination propagation process, analyze the confirmation status of each node, and evaluate the triggering effect of the pre-wake-up rules.

[0057] The following is a specific example illustrating the embodiments of this application. Assume that the first node A, single-hop node B, and double-hop node C are sequentially deployed along a drainage branch. The pipe section length between node A and node B is 120 meters, and the pipe section length between node B and node C is 150 meters. Node A collects a full-spectrum sample at 10:00:00. Based on the target event rule, the peak intensity of target peak region P1 is 185, the peak intensity of target peak region P2 is 108, and the peak intensity ratio R is 1.71, thus hitting the target event template E1. Node A generates event summary information and reports it to the management device. Based on the current flow velocity of 0.60 m / s and a retention correction time of 60 seconds, the management device calculates the estimated propagation time of node B as 260 seconds and generates an estimated propagation window for node B from 10:03:40 to 10:05:20. The management device activates the pre-wake-up flow table rule for node B. Node B performs a local peak pre-scan every 20 seconds within the expected propagation window. It obtains pre-scan results that meet the confirmation criteria at 10:04:10 and 10:04:30. Based on this, the management device determines Node B as the target single-hop node and instructs Node B to perform full-spectrum acquisition every 30 seconds for the next 5 minutes. Simultaneously, the management device updates the event summary information based on Node B's confirmation results and issues the corresponding pre-wake-up flow table rule to Node C. If Node C does not obtain a pre-scan result that meets the confirmation criteria within its expected propagation window, the management device cancels Node C's pre-wake-up flow table rule and terminates the directional propagation.

[0058] Through the above implementation methods, those skilled in the art will understand that the embodiments of this application do not require all neighboring nodes to immediately enter the full-spectrum high-frequency sampling state after the first node is triggered. Instead, the neighboring nodes are first pre-scanned with low load around the target peak area through event summaries and pre-wake-up flow table rules. Only when it is confirmed that the pollution continues to spread will the sampling be upgraded level by level and extended to the next level node, thereby reducing the collection and transmission burden of the distributed collection nodes while ensuring the transient pollution capture capability.

[0059] It should be noted that the parameters and their value relationships regarding the target event rules, expected propagation window, pre-wake-up flow table rules, local peak area pre-scan, secondary confirmation, and event summary update are only examples and not limitations. In implementation, adjustments can be made according to different monitoring scenarios, node deployment methods, pipeline structures, and pollution types. This application embodiment does not limit these adjustments.

[0060] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the scope of protection of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A distributed acquisition method for three-dimensional fluorescence water quality fingerprints based on flow table rules, characterized in that, The method, applied to management equipment, includes: acquiring event summary information sent by a first node, wherein the event summary information is generated by the first node when its collected three-dimensional fluorescence water quality fingerprint hits a target event rule; determining at least one single-hop node directly adjacent to the first node from multiple acquisition nodes based on pre-stored node topology information; sending a pre-wake-up flow table rule corresponding to the event summary information to each of the at least one single-hop node, so that each single-hop node performs a local peak region pre-scan of the target peak region according to the pre-wake-up flow table rule within the expected propagation window; after the local peak region pre-scan result of the target single-hop node meets a preset confirmation condition, sending an upgrade acquisition instruction to the target single-hop node, and sending the updated pre-wake-up flow table rule corresponding to the event summary information to at least one double-hop node directly adjacent to the target single-hop node.

2. The three-dimensional fluorescence water quality fingerprint distributed acquisition method based on flow table rules according to claim 1, characterized in that, The step of obtaining the event summary information sent by the first node includes: obtaining the event summary information extracted by the first node based on the full spectrum acquisition results corresponding to the target event; wherein, the event summary information includes at least one of the following: event chain identifier, target peak region identifier, peak intensity ratio feature, anomaly level, first node identifier, and expected propagation window.

3. The distributed acquisition method for three-dimensional fluorescence water quality fingerprints based on flow table rules according to claim 1, characterized in that, When the at least one single-hop node includes multiple single-hop nodes, sending the pre-wake-up flow table rule corresponding to the event digest information to each of the at least one single-hop nodes includes: sending the pre-wake-up flow table rule corresponding to the event digest information to each of the multiple single-hop nodes in descending order of priority.

4. The three-dimensional fluorescence water quality fingerprint distributed acquisition method based on flow table rules according to claim 3, characterized in that, The priority of the plurality of single-hop nodes is related to the expected propagation time corresponding to the plurality of single-hop nodes; wherein, the shorter the expected propagation time, the higher the priority of the single-hop node corresponding to the expected propagation time.

5. The distributed acquisition method for three-dimensional fluorescence water quality fingerprints based on flow table rules according to claim 3 or 4, characterized in that, The priority of the plurality of single-hop nodes is related to the branch risk information corresponding to the plurality of single-hop nodes; wherein, the higher the risk level indicated by the branch risk information, the higher the priority of the single-hop node corresponding to the branch risk information.

6. The distributed acquisition method for three-dimensional fluorescence water quality fingerprints based on flow table rules according to claim 1, characterized in that, The pre-wake-up flow table rules include a matching field and an action field; wherein, the matching field includes the target peak region identifier, the matching interval corresponding to the peak intensity ratio feature, and the expected propagation time window; the action field includes entering the pre-wake-up state and performing local peak region pre-scan.

7. The distributed acquisition method for three-dimensional fluorescence water quality fingerprints based on flow table rules according to claim 1, characterized in that, The preset confirmation conditions include at least one of the following: the target peak region position offset in the local peak region pre-scan result is within a preset tolerance range; the peak intensity ratio in the local peak region pre-scan result is within a preset ratio range; and the target single-hop node obtains local peak region pre-scan results that meet the matching conditions at least twice consecutively within the expected propagation window.

8. The three-dimensional fluorescence water quality fingerprint distributed acquisition method based on flow table rules according to claim 7, characterized in that, Sending upgrade acquisition instruction information to the target single-hop node includes: instructing the target single-hop node to switch from a pre-wake-up state to a confirmation state; instructing the target single-hop node to switch from a local peak pre-scan mode to a full-spectrum acquisition mode or a high-frequency sampling mode; and updating the event summary information based on the confirmation result of the target single-hop node.

9. The three-dimensional fluorescence water quality fingerprint distributed acquisition method based on flow table rules according to claim 8, characterized in that, When the at least one double-hop node includes multiple double-hop nodes, the step of sending the pre-wake-up flow table rule corresponding to the updated event digest information to the at least one double-hop node directly adjacent to the target single-hop node includes: sending the pre-wake-up flow table rule corresponding to the updated event digest information to each of the multiple double-hop nodes in descending order of priority.

10. The distributed acquisition method for three-dimensional fluorescence water quality fingerprints based on flow table rules according to claim 8, characterized in that, The method further includes: when a node in the pre-wake-up state at any level fails to meet the preset confirmation condition within the corresponding expected propagation time window, the activation state of the pre-wake-up flow table rule corresponding to that node is revoked, the node is restored to the normal collection state, and the subsequent expansion in the propagation direction of that node is terminated; wherein, the pre-wake-up flow table rule is generated by activating a pre-loaded dormant rule template, and the dormant rule template includes a matching field, an action field, and a timeout field.

Images