Multi-stage linkage industrial wastewater comprehensive toxicity real-time monitoring and emergency response method

Abstract

The present application relates to the technical field of wastewater toxicity monitoring, in particular to a multi-stage linkage industrial wastewater comprehensive toxicity real-time monitoring and emergency response method, comprising the following steps: obtaining multi-point spectral data and comparing main peak shift, establishing discharge timing relationship, checking node connection order to restore path structure, extracting continuous monitoring section calibration disturbance period, interfacing action identification to complete missing position, and constructing toxicity linkage response closed-loop state set. In the present application, through the cooperative comparison of main peak shift trend among multiple monitoring points, the time sequence correlation between pollution change process and spatial nodes is established, combined with the path connection order to promote node arrangement, the disturbance response time range is extracted from the continuous numerical change section, the sequence corresponding relationship is formed by the action number interfacing the path node, combined with the state feedback result to construct the closed-loop process of instruction issuance and response recovery, and the through connection between pollution process identification, response action scheduling and state return result is completed.

Description

Technical Field

[0001] This invention relates to the field of wastewater toxicity monitoring technology, and in particular to a multi-level linkage method for real-time monitoring and emergency response of the comprehensive toxicity of industrial wastewater. Background Technology

[0002] The field of wastewater toxicity monitoring technology involves the identification, assessment, and dynamic monitoring of toxic and hazardous substances in industrial wastewater. Core aspects include the extraction of spectral and mass spectrometric features of typical pollutants, identification of toxicity response patterns, methods for identifying anomalies in water quality indicators, pollution source location and tracing technologies, and multi-source fusion and comprehensive processing of monitoring data. This field widely utilizes fluorescence spectroscopy, mass spectrometry fingerprinting, principal component analysis, and artificial intelligence modeling to achieve real-time monitoring and risk warning of the entire wastewater discharge process in key areas such as industrial parks and wastewater treatment plants. Traditional multi-level linkage methods for real-time monitoring and emergency response of comprehensive toxicity in industrial wastewater involve setting up sampling points at key nodes such as major polluting enterprises, sewage trunk lines, and treatment facilities to collect wastewater samples and extract their fluorescence and mass spectrometry fingerprint features to construct a pollution source fingerprint database. Spectral similarity comparison and principal component analysis are then used to identify and classify pollution sources. Multi-level early warning thresholds are set based on indicators such as ammonia nitrogen, COD, and TOC to establish linkage early warning rules based on water quality fluctuations. Furthermore, neural network algorithms are used to construct a regression model between spectral and toxicity responses, driving the intelligent management system to complete the entire process of pollution identification, risk classification, and response scheduling. However, this approach relies heavily on manual comparison, has long monitoring cycles, and lacks system integration and a real-time closed-loop response mechanism.

[0003] In the process of pollution monitoring, feature extraction and path identification are decoupled, resulting in a lack of structural connection between response identification and pollution diffusion evolution. When faced with simultaneous fluctuations in indicators at multiple locations or when pollution chains have hidden distribution characteristics, the failure of temporal and spatial mapping can easily lead to omissions in the response process. The method of triggering early warning by relying on preset thresholds cannot effectively deal with nonlinear fluctuations in indicators caused by continuous disturbances or sudden diffusion. The manual participation in the comparison process causes data feedback delays and uneven instruction transmission, making it difficult to cover all node response scenarios in the chain path. For example, when the toxic response deviates in multiple segments during the transmission process along the pipeline network, it is difficult to achieve dynamic linkage between early warning and disposal actions. Summary of the Invention

[0004] To address the technical problems existing in the prior art, embodiments of the present invention provide a multi-level linkage method for real-time monitoring and emergency response of the comprehensive toxicity of industrial wastewater; To achieve the above objectives, the present invention adopts the following technical solution: a multi-level linkage method for real-time monitoring and emergency response of comprehensive toxicity of industrial wastewater, comprising the following steps: S1: Acquire spectral data from industrial wastewater inlet monitoring wells, terminal discharge outlets, and online sensing monitoring chambers; analyze the characteristics of the starting position and direction of movement of the main peak; and analyze the temporal relationship of the characteristics in the emission sequence based on the main peak shift time point to obtain a set of linkage response feature segments. S2: Based on the time period number in the linkage response feature map segment set, retrieve the corresponding discharge node in the drainage path map, advance the node downstream along the path, check the time sequence by comparing the upstream and downstream connection relationship, and obtain the pollution chain segment structure path group. S3: Based on the pollution chain segment structure path group, extract the continuous monitoring value sequence and toxicity response parameters within the corresponding time period, analyze the changing trend of node monitoring values ​​and toxicity characterization data, identify continuous segments with consistent changing trends based on the direction of numerical changes between continuous time points, and define the start and end time range of toxicity disturbances in combination with the magnitude of changes in toxicity parameters to obtain a list of toxicity disturbance response time periods. S4: Based on the list of toxic disturbance response time periods, extract the corresponding response action identifiers, match the pollution chain segment path node numbers with the action identifiers, and fill in the unmatched node action numbers to obtain the linkage response structure action sequence. S5: Based on the action sequence of the linkage response structure, the action commands are sequentially issued to the emission control link, and the corresponding response status information is extracted to obtain the closed-loop state set of toxic linkage response.

[0005] As a further embodiment of the present invention, the linkage response feature segment set includes the starting position of the main peak, the direction of movement of the main peak, the continuous offset number of the main peak band, and the time axis position mark; the pollution chain segment structure path group includes the time period number, the emission node number, the drainage path connection sequence, and the node advancement sequence; the toxic disturbance response time period list includes the continuous monitoring value sequence, the direction of numerical change, the start and end time of the change segment, and the direction consistency indicator; the linkage response structure action sequence includes the action identifier, the node number correspondence, the action gap completion mark, and the action sequence; and the toxic linkage response closed-loop state set includes the action execution instruction, the response point return status, the non-response identifier item, and the status and action comparison sequence.

[0006] As a further aspect of the present invention, the starting position of the main peak refers to the initial wavelength value of the spectral characteristics of the pollutant; The continuous monitoring value sequence refers to the sequence of monitoring data values ​​of emission nodes arranged in chronological order within a specified time period.

[0007] As a further aspect of the present invention, the unmatched node refers to a node that has not yet been assigned an action number during the process of connecting the path node with the response action identifier, and needs to be completed by interpolation of neighboring nodes and replacement paragraphs; The emission control component refers to the actual operational part that executes the linkage response action command.

[0008] As a further aspect of the present invention, the specific steps of S1 are as follows: S101: Acquire raw spectral data from water intake monitoring wells, terminal outlets, and online sensing monitoring chambers; extract the initial wavelength value of the main peak at each monitoring point; track the wavelength change trend; identify the spectral region where the main peak continuously shifts within a continuous time axis; and obtain a set of continuously shifted main peak bands. S102: Based on the main peak continuous offset band set, extract the acquisition time corresponding to the spectral region at the point, then filter the spectral data segments that show the offset trajectory according to the time, and aggregate the parts with overlapping bands and time progression relationship to obtain the main peak advancement band sequence. S103: Based on the main peak advancing band sequence, compare the band direction of the spectral data segments in the differentiated monitoring points according to the time advancement relationship, extract the continuous data segment group with consistent offset characteristics, and string them together according to the position of the band corresponding to the time to obtain the linkage response feature map segment set.

[0009] As a further aspect of the present invention, the specific steps of S2 are as follows: S201: Based on the time period number and monitoring point number corresponding to the linkage response feature map segment set, retrieve the node distribution information in the drainage path map, and according to the correspondence between the monitoring point number and the discharge node number, retrieve the coordinate position of the discharge node corresponding to the time period number in the path map, compare the node coordinates according to the order of the time period numbers, and obtain the path node positioning dataset. S202: Based on the path node positioning dataset, call the connection information in the path graph, extract the connection direction of the node pairs with continuous direction from the graph coordinates, analyze the line segment direction of the node pairs in the path graph, and obtain the node flow direction construction relationship set. S203: Based on the node flow direction, construct a relationship set, starting with the node located at the beginning of the time number in the graph, and advance the nodes in sequence according to the connection direction. Connect and extend the nodes according to the path direction to obtain the pollution chain segment structure path group.

[0010] As a further aspect of the present invention, the specific steps of S3 are as follows: S301: Based on the time period corresponding to the emission node in the pollution chain structure path group, extract the monitoring data and toxicity response parameters of the node in the corresponding time period, arrange the monitoring values ​​and toxicity characterization data corresponding to the node in chronological order, and obtain a continuous monitoring value sequence set. S302: Based on the continuously monitored numerical sequence set, compare the direction of change of the monitored values ​​and the trend of change of the toxicity parameters between consecutive time points in the data sequence, identify continuous segments with the same direction of change, and locate the start and end positions of the continuous segments on the time axis to obtain the time range set of trend change segments. S303: Based on the set of time ranges of the trend change intervals, screen the monitoring data segments with overlapping time ranges from the emission nodes, analyze whether the direction of numerical change of the differentiated nodes in the corresponding time period is consistent with the trend of toxicity parameter change, and define the time range of toxicity disturbance according to the magnitude of toxicity parameter change to obtain a list of toxicity disturbance response periods.

[0011] As a further aspect of the present invention, in the process of comparing the direction of numerical differences at consecutive time points within the data sequence: identifying data pairs between adjacent time points that exhibit consistent changes in both magnitude and direction, and connecting them sequentially into a continuous sequence according to the chronological order; In the process of identifying continuous segments with the same direction of numerical change: analyze the distribution of data pairs with the same direction on the time axis, filter out mutually continuous and continuous sequence segments, and divide the time limits at the beginning and end of the sequence segments; In the process of filtering data segments with overlapping time ranges from nodes: by comparing the time range of sequence segments in emission nodes, data segments with overlapping time ranges are extracted, and the direction of change of emission nodes in overlapping time ranges is analyzed to see if they are consistent.

[0012] As a further aspect of the present invention, the specific steps of S4 are as follows: S401: Based on the time range in the toxic disturbance response time period list, extract the action identifiers associated with the time period from the response action information set, and compare them according to the chronological order to obtain the action sequence matching set; S402: Based on the node number sequence in the pollution chain path, match each group of action identifiers in the action sequence matching set with the node number, locate the action content to which the node belongs, and obtain the action node docking result; S403: Based on the nodes with missing positions in the action node docking results, extract the corresponding action identifiers of adjacent nodes and fill in the missing points to obtain the linkage response structure action sequence.

[0013] As a further aspect of the present invention, the specific steps of S5 are as follows: S501: Based on the action identifier in the action sequence of the linkage response structure, the action instructions are sequentially assigned to the control execution nodes corresponding to the emission control link, and the action execution process of the control execution nodes is monitored, the corresponding action status data is collected, and the action response status set is obtained after the control execution nodes return the corresponding status information. S502: Based on the action state content in the action response state set, compare the action identifier text with the state field by referring to the action identifier content in the action sequence of the linkage response structure, analyze whether there are any incomplete action items in the response process, and obtain the missing action response information. S503: Based on the missing action response information, extract the corresponding action identifier and state feedback content, trace each state according to the action sequence, and obtain the closed-loop state set of toxic linkage response.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by coordinating the main peak shift trend among multiple monitoring points, a time series correlation between the pollution change process and spatial nodes is established. The node arrangement is advanced by combining the path connection sequence, and the range of disturbance response time is defined by extracting the continuously changing numerical segments. The action number is used to connect the path nodes to form a sequence correspondence. The closed-loop process of instruction issuance and response recovery is constructed by combining the status feedback results, thus completing the connection between pollution process identification, response action scheduling and status feedback results. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the steps of the present invention; Figure 2 This is a detailed schematic diagram of S1 of the present invention; Figure 3 This is a detailed schematic diagram of S2 of the present invention; Figure 4 This is a detailed schematic diagram of S3 of the present invention; Figure 5 This is a detailed schematic diagram of S4 of the present invention; Figure 6 This is a detailed schematic diagram of S5 of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0018] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0019] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0020] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0021] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0022] Please see Figure 1 This invention provides a multi-level linkage method for real-time monitoring and emergency response of the comprehensive toxicity of industrial wastewater, including the following steps: S1: Obtain the raw spectral data of the water inlet detection well, the terminal outlet, and the online sensing detection chamber on the same time axis, compare the starting position of the main peak and the direction of the main peak movement at each monitoring point, identify the continuous shift of the main peak band at the point, assign corresponding numbers to spectral segments with the same shift trend, and then trace the corresponding positions of the segments according to the time sequence to obtain the linkage response feature map segment set. S2: Based on the time period number of the linkage response feature map segment set, refer to the drainage path map to find the corresponding node, move the time number downstream node by node according to the drainage direction, check the connection order of each node in the drainage path, advance the node along the path one by one, restore the flow relationship in the discharge channel according to the order of the time number, and obtain the pollution chain segment structure path group. S3: Based on the pollution chain segment structure path group, extract the continuous monitoring value sequence and toxicity response parameters within the corresponding time period, analyze the changing trend of node monitoring values ​​and toxicity characterization data, identify continuous segments with consistent changing trends based on the direction of numerical changes between continuous time points, and define the start and end time range of toxicity disturbances in combination with the magnitude of changes in toxicity parameters to obtain a list of toxicity disturbance response time periods. S4: Based on the time range in the list of toxic disturbance response periods, extract the action identifiers corresponding to the time periods from the existing response action content, connect the node numbers in the pollution chain path with the action identifiers in sequence, and directly fill in the adjacent action identifiers at the positions where action gaps occur in the path to obtain the linkage response structure action sequence. S5: Based on the action identifiers in the action sequence of the linkage response structure, the action instructions are assigned to the corresponding execution parts of the emission control link in sequence. After the status information is returned at the response point, the action identifiers are compared to analyze whether there are any unresponded identifiers. The status information is then exported after being matched with the action sequence to obtain the closed-loop status set of toxicity linkage response.

[0023] The linkage response feature segment set includes the starting position of the main peak, the direction of movement of the main peak, the continuous offset number of the main peak band, and the time axis position mark. The pollution chain segment structure path group includes the time period number, the emission node number, the drainage path connection sequence, and the node advancement sequence. The toxic disturbance response time period list includes the continuous monitoring value sequence, the direction of numerical change, the start and end time of the change segment, and the direction consistency indicator. The linkage response structure action sequence includes the action identifier, the node number correspondence, the action gap completion mark, and the action sequence. The toxic linkage response closed-loop state set includes the action execution command, the response point return status, the non-response identifier item, and the status and action comparison sequence.

[0024] Please see Figure 2 The specific steps of S1 are as follows: S101: Acquire raw spectral data from water intake monitoring wells, terminal outlets, and online sensing monitoring chambers; extract the initial wavelength value of the main peak at each monitoring point; track the wavelength change trend; identify the spectral region where the main peak continuously shifts within a continuous time axis; and obtain a set of continuously shifted main peak bands. First, the spectral data frames of each monitoring point need to be traversed frame by frame to extract the distribution curve of its reflectance intensity value across the entire wavelength range. The location where the reflectance value reaches its maximum value is determined as the main peak wavelength value. By calculating the wavelength index of the main peak point on the spectral curve of each frame of data, a wavelength sequence vector is constructed to represent the main peak position at each time point. Then, based on this wavelength sequence vector, continuous time frames are read to retrieve the direction of change of wavelength values. If the wavelength value consistently shows an increasing trend over multiple consecutive time periods, it is determined that the shift direction is extending towards the long wavelength band; otherwise, it is considered to be moving towards the short wavelength band. In the judgment process, the difference between two adjacent wavelength values ​​is compared, and a non-zero difference between the two values ​​is set as the basis for the shift. Then, by sequentially moving the comparison window to traverse all time frame pairs in the wavelength sequence vector, the shift trend in the entire spectral data stream can be tracked and identified. For example, if the corresponding main peak wavelength values ​​in five consecutive time slices are 43... If the sequence is 6, 437, 439, 440, 441, then it meets the criteria for continuous shift judgment, confirming the existence of wavelength shift within this time interval. The judgment results are then mapped back to the original spectral data frames, and the spectral range corresponding to the frames is extracted. For example, the wavelength range is positioned between 430 and 450. Further, a lateral consistency search is performed on the main peak intensity of each time frame in this band, and the intensity peak change pattern is extended. Band frame index positions that do not have a continuous trend are eliminated, and all spectral frame numbers within the continuous shift area are compressed and stored to obtain the wavelength coverage range of the continuous shift segment. For example, if the main peak is continuously identified to shift from 425 to 444 in data frames numbered 21 to 32, and the band intensity change direction is consistent, then this segment is classified as a continuous shift trajectory of the main peak. Multiple such shift trajectories are then distinguished according to their respective monitoring points to achieve response mapping between spectral monitoring points, and finally, a set of continuous shift bands of the main peak is obtained.

[0025] S102: Based on the continuous shift band set of the main peak, extract the acquisition time corresponding to the spectral region at the point, then filter the spectral data segments that show the shift trajectory according to the time, and aggregate the parts with overlapping bands and time progression relationship to obtain the main peak advancement band sequence. First, each band segment is located according to the monitoring point number, and the corresponding acquisition timestamp is retrieved from the original dataset. By traversing the data index positions in all offset band segments, the time points bound to the start and end frames of each segment are extracted and assigned as the start and end times of the segment, respectively. After the time range of all segments is extracted, a spectral data segment sequence with the acquisition time as a reference is constructed. The time sequence between segments in this sequence is searched. If the end time of the previous segment is less than the start time of the next segment, the two segments are determined to be in a continuous progression relationship. For example, if the end time of segment A is 21 and the start time of segment B is 22, then the two segments are in a continuous progression relationship. Then, the ranges of these segments with a continuous progression relationship are compared by searching whether there is an overlap in wavelength values ​​in their corresponding band ranges. If the wavelength range of segment A is 435 to 447 and the wavelength range of segment B is 442 to 452, then the two segments are in a continuous progression relationship because their band ranges are different. There is overlap between 442 and 447, which satisfies the band crossover relationship. Such segments that have both time progression and band overlap are considered as progressive segment pairs. The chain connection is further searched among multiple progressive segment pairs. If segment B and segment C also constitute a progressive segment pair, then segments A, B, and C are regarded as the same progressive chain. In the process of constructing the progressive chain, each newly added segment needs to have its start time compared with the end time of the previous segment. At the same time, the wavelength range is judged for interval crossover to ensure that the progressive chain has continuity in time axis and band interval. For example, if the time of three segments is 10 to 15, 16 to 20, and 21 to 25 respectively, and the bands are 430 to 440, 438 to 448, and 446 to 456 respectively, the three data segments are both continuous in time and overlapped in band, which can be used to construct a complete progressive chain. All segments of this type of progressive chain are aggregated into a group of progressive data, and then the segments are connected in sequence according to time to obtain the main peak progressive band sequence.

[0026] S103: Based on the main peak advancing band sequence, the band direction of the spectral data segment in the differentiated monitoring points is compared according to the time advancement relationship. A continuous data segment group with consistent offset characteristics is extracted and connected in series according to the position of the band corresponding to the time to obtain the linkage response feature map segment set. First, the acquisition time and monitoring point number are extracted from each spectral segment. Then, spectral segments with the same or similar time points are selected among multiple monitoring points as comparison benchmarks. The band range and time value in each group of spectral segments are doubly indexed, and the band change direction at different monitoring points is compared sequentially according to the time progression. By comparing the movement trend of the band center position on the time axis, the consistency of its offset in multiple monitoring points is identified. If the band center moves from 432 to 440 at point A and from 434 to 442 at point B, it can be determined that they have the same offset direction. Based on this, a judgment relationship of consistent offset trend is established. Segments that meet the same offset trend are grouped into the same data segment group, and then the data segment group is searched. Find the acquisition time points corresponding to the start and end segments, and arrange the corresponding positions of each segment on the time axis in chronological order. For example, if the start time of segment M is 12, the start time of segment N is 14, and the start time of segment P is 16, then the three form a segment concatenation relationship in sequence. Then, number and map the data segment group after the above time sequence concatenation, and extend its continuity on the time axis according to the time progression path of each group of segments. In this way, sequence data with time progression attributes are obtained. Then, combined with the offset consistency judgment result, group data with the same center offset direction of all bands are extracted from all data segment groups, and these group segments are concatenated into a time progression trajectory in time sorting order, and finally the linkage response feature map segment set is obtained.

[0027] Please see Figure 3 The specific steps of S2 are as follows: S201: Based on the time period number and monitoring point number corresponding to the linkage response feature map segment set, retrieve the node distribution information in the drainage path map, and according to the correspondence between the monitoring point number and the discharge node number, retrieve the coordinate position of the discharge node corresponding to the time period number in the path map, compare the node coordinates according to the order of the time period number, and obtain the path node positioning dataset. First, the complete node distribution pattern is read from the drainage path map. The time segments indicated by each node number are retrieved one by one and mapped to the corresponding coordinates on the path map. The coordinates of each node are sequentially extracted according to their node number order. For example, if a segment is numbered A1 and the corresponding node number is N5, the coordinates of node N5 in the path map are extracted as [285, 710]. Based on these coordinates, their order within the time segment numbers is compared, matching the time sequence with the coordinate sequence. This establishes a connection between time index and spatial location across multiple nodes. Next, all extracted coordinate points are sequence-normalized. During this normalization process, each coordinate pair is reconstructed into a linear data frame according to its number order. For example, [A1-N5-285]... The coordinates of points A, B, C, D, E, F ...

[0028] S202: Based on the path node localization dataset, call the connection information in the path graph, extract the connection direction of the directionally continuous node pairs from the graph coordinates, analyze the line segment direction of the node pairs in the path graph, and obtain the node flow direction construction relationship set. First, the adjacent pairing relationships between node numbers are extracted item by item. Then, the spatial connection paths are queried using the connection information marked in the path diagram. When calling the connection parameters of each node in the diagram, the order of the node numbers is used to set the preceding and following relationships. For example, nodes N5 and N6 form a pairing combination, and their spectral coordinates are obtained as [285, 710] and [289, 712]. After extracting the start and end coordinates of the corresponding connection lines, the direction vector is derived based on the difference between the horizontal and vertical directions. A positive horizontal change value indicates a connection direction from left to right, and a positive vertical change value indicates a connection direction from top to bottom. The direction of coordinate change is used as a judgment criterion, and the horizontal direction change flag D1 and the vertical direction change flag are set accordingly. The flag D2 defines the direction as a reverse change when either D1 or D2 is negative. If both D1 and D2 are zero, it indicates that the nodes are in a static connected state and do not participate in the direction determination. Based on this, the range of connection slope changes between each pair of nodes is calculated. For example, if D1 is 4 and D2 is 2, the slope ratio of the node pair is 2. A direction classification matrix is ​​constructed according to the slope results, dividing the node pairs with positive slope ratios into forward channel segments and the rest into reverse channel segments. Then, cross-location is performed using this matrix and the node connection order. The node index set dominated by forward connection is extracted in the same channel segment. Finally, the direction feature attributes of each pair of nodes are sorted in numerical order to obtain the node flow direction construction relationship set.

[0029] S203: Construct a relation set based on node flow direction. Starting with the node located at the beginning of the time number in the graph, advance the nodes in sequence according to the connection direction, and connect and extend the nodes according to the path direction to obtain the pollution chain segment structure path group. First, retrieve the starting node index corresponding to the starting position of the time number from the graph. Then, determine the first node of the channel it belongs to in the graph by the connection direction of the preceding node. For example, set node N11, which is the starting point, in the flow relation set, call the direction field and advance one node unit according to the forward relation set in the direction field. Extract the next node number that is directly connected to the starting point and has the same direction, such as N13. Then, find the backward connection target of N13 in the flow relation set, which is N15. Continue to extend the connection chain in this way. For each pair of connections, check the direction consistency to determine whether there is a direction in the chain. In the event of offset or breakage, if a node marked as a broken chain or intersection is encountered during advancement, advancement is stopped, and all node numbers in the current chain are temporarily stored in the chain data temporary storage set. The above operation is repeated with the next starting node N21 as the starting point of the new chain. For each advancement chain segment, the coordinate information and connection index value in the path graph are called for alignment comparison. All nodes in the continuous chain are linked according to the direction advancement order to form a continuous path sequence. The path groups are classified according to the visible trajectory partitions connected in the graph during the node advancement process. Finally, all advancement path segments are summarized to obtain the contaminated chain segment structure path group.

[0030] Please see Figure 4 The specific steps of S3 are as follows: S301: Based on the time period corresponding to the emission node in the pollution chain structure path group, extract the monitoring data and toxicity response parameters of the node in the corresponding time period, arrange the monitoring values ​​and toxicity characterization data corresponding to the node in chronological order, and obtain a continuous monitoring value sequence set. First, for each node sequence within the path group, the original monitoring records corresponding to the monitoring stage are retrieved. The node number and time period index are then matched one by one to obtain all monitoring data and toxicity response parameters for the node within the corresponding time period. The monitoring record for each node is further divided into multiple fields: original reading, toxicity characterization value, sampling trigger sequence, and sampling point number, which serve as the data frame for subsequent calculations. Next, the node sequence information within the path group is retrieved, and the start times of the corresponding time periods for nodes within the same path are searched side-by-side. The field content of the monitoring readings and toxicity characterization values ​​for each node is evaluated. The reading field is compared with the monitoring range field within the time period, and the toxicity characterization value is matched with the preset toxicity parameter range to determine if the corresponding field value falls within the specified range. If the field value exceeds the corresponding range, the abnormal field value is replaced with the boundary value of the corresponding range to ensure consistency between the node monitoring sequence and the toxicity parameter sequence under a unified dimension. Finally, the monitoring reading field of each node is... The toxicity characterization fields are sorted from smallest to largest according to the time period index. The data fields are rearranged by comparing the positions of the sorted data. During the rearrangement process, the absolute difference between the monitoring readings and the changes in toxicity parameters is used as the criterion. If the difference between adjacent points is within the allowable deviation range, the sequence is kept continuous. If the difference exceeds the limit, the corresponding point is marked as an abnormal jump point and replaced with a jump flag field in the sequence. Then, the sorted sequences of all nodes are aggregated into a continuous node sequence according to the path order. The number of jump flag fields in the sequence is counted and used as the criterion for judging the smoothness of the sequence and the stability of toxicity changes. If the number of jump flags is within a reasonable range, the sequence is retained as a path-stable monitoring sequence. Finally, all retained sequences are combined and aligned by path to ensure that the starting points of adjacent path sequences are consistent on the time axis. The monitoring sequences and toxicity parameter sequences of each node are connected under a unified time reference to obtain a continuous monitoring value sequence set.

[0031] S302: Based on the continuous monitoring numerical sequence set, compare the direction of change of monitoring values ​​and the trend of change of toxic parameters between consecutive time points in the data sequence, identify continuous segments with the same direction of change, and locate the start and end positions of the continuous segments on the time axis to obtain the time range set of trend change segments. First, for each continuous monitoring sequence, the monitoring value field and toxicity parameter field corresponding to two adjacent time points are extracted according to time sequence. The direction of numerical difference and toxicity change trend between each pair of adjacent data are calculated. An upward trend judgment indicator is obtained by judging whether the previous monitoring value is less than the next monitoring value, or a downward trend judgment indicator is obtained by judging whether the previous monitoring value is greater than the next monitoring value. At the same time, the corresponding toxicity trend indicator is obtained according to the increase or decrease of the toxicity parameter. Then, consecutive identical marker segments are extracted from the trend judgment indicator and toxicity trend indicator in the sequence, and a trend switching marker is set at the position where the trend indicator changes, dividing the entire sequence into several trend segments with the same direction of change. Subsequently, the start and end time point indices are extracted for each trend segment, and the corresponding start and end times are extracted from the continuous monitoring value sequence. The start and end times are combined to form the distribution range of the trend segment on the time axis. Based on this, it is further determined whether the monitored values ​​and toxicity parameters within the segment continuously show a unidirectional increase or a unidirectional decrease. If there is a drop in values, local fluctuations, or a reversal of the toxicity trend, the corresponding segment is removed, and only the trend segments with continuous and stable changes are retained as valid segments. Then, the time ranges of all valid segments are cross-searched to exclude segments with insufficient overlap in time spans between different sequences. Furthermore, the trend change segments in each path are extracted sequentially according to the emission path direction, so that the trend change segments in multiple path directions form a continuous connection relationship on the time axis. Finally, the boundaries of all time segments are extracted from the coverage of the continuous trend segments, and the corresponding toxicity change intervals are determined by combining the change amplitude of toxicity parameters to obtain the set of time ranges of trend change segments.

[0032] S303: Based on the time range set of trend change segments, screen the monitoring data segments with overlapping time ranges from the emission nodes, analyze whether the direction of numerical change of differentiated nodes in the corresponding time period is consistent with the trend of toxicity parameter change, and define the time range of toxicity disturbance according to the magnitude of toxicity parameter change to obtain a list of toxicity disturbance response periods. First, monitoring data segments and toxicity parameter data segments for each emission node within the corresponding time period are extracted from the continuous monitoring numerical sequence set. Cross-referencing is then performed on the data segments between multiple nodes, using time points as references. By comparing the direction of change in monitoring values ​​and the trend of change in toxicity parameters at the same time point for each node, consistency analysis is conducted on the change trends of different nodes within each time period in the continuous change segment. Specifically, the difference in monitoring values ​​and toxicity parameters between two adjacent time points within each node is calculated, and the direction of change is vectorized into positive / negative relationship identifier sequences and toxicity trend identifier sequences. Through parallel comparison of the identifier sequences within corresponding time periods between nodes, it is identified that all nodes exhibit the same direction of change and consistent toxicity trends within the same time period. The process involves first identifying a set of time periods with consistent trends; then further removing directional reversals caused by sensing errors, data jitter, or local interference from this set, and filtering out invalid segments with insufficient duration based on the minimum duration of continuous segments; then extracting time segments that simultaneously exhibit unidirectional changes and synchronous changes in toxicity parameters at multiple nodes along the emission path, and splicing them together according to their relative order along the path flow to identify continuous toxicity response segments with temporal consistency during the emission path propagation; finally, selecting a set of segments covering multiple nodes and exhibiting both trend consistency and toxicity change consistency from all nodes, and defining the toxicity disturbance time range based on the magnitude of toxicity parameter changes to obtain a list of toxicity disturbance response time periods.

[0033] Please see Figure 5 The specific steps of S4 are as follows: S401: Based on the time range in the list of toxic disturbance response periods, extract the action identifiers associated with the time periods from the response action information set, and compare them according to the chronological order to obtain the action sequence matching set; First, each response period is broken down into a sequence of start and end time points. Then, the time point data recorded in the action information set is scanned sequentially. For any action information item, if its time point falls within a certain response period, the action identifier of that action information item is extracted. The extracted action identifiers are aggregated into a group by time period. Then, the time point is used as a sorting reference to sort the action identifiers in each group in chronological order. The position of the action identifiers in the group is adjusted by comparing the time points numerically. Further, time interval analysis is used to identify the response sequence correlation between adjacent action identifiers. When the interval between any pair of action identifiers is too short or equal, they are identified as parallel or simultaneous response actions. Parallel marking information is added for subsequent calls. All sorted and analyzed action identifier sets are converted into a set of serialized action trajectories corresponding to the response period. Then, the action identifier sequence sets corresponding to all response periods are connected in chronological order to extract the mapping relationship between time points and action identifiers. They are then concatenated in chronological order to obtain the action sequence matching set.

[0034] S402: Based on the node number sequence in the contamination chain path, match each group of action identifiers in the action sequence matching set with the node number, locate the action content to which the node belongs, and obtain the action node docking result; First, the path nodes are sequentially numbered according to their progression order in the channel, forming a linearly advancing node sequence. Each node number corresponds to a specific physical location. Then, the action identifier sequence from the action sequence matching set is called, and each group of action identifiers is compared one-to-one with the path node sequence in the order of appearance. During the comparison, the action identifiers are mapped to node numbers sequentially according to time. When the number of nodes exceeds the number of action identifiers, the consecutive corresponding positions of the action identifiers in the preceding part of the node sequence are retained. For the remaining unmatched nodes, the action identifiers of the nodes before and after them that have completed mapping are searched. Among these two action identifiers, the one with the closest time point is selected as the action pointer for that node, and the action identifier information is added back to that node. When the number of action identifiers exceeds the number of node numbers, the excess actions are reserved in a spare queue for subsequent node additions. The matching process is used in the supplementation stage. Subsequently, the correspondence between the mapped node numbers and action identifiers is organized to construct a three-field combination containing number, action identifier, and time sequence. When generating the node action correspondence, it is necessary to check the uniformity of the distribution of action identifiers in the path nodes. Possible missing pairing areas are identified by calculating the repetition rate of action identifiers or the magnitude of changes in position spacing. For example, when three consecutive nodes have not been assigned action identifiers, but their adjacent nodes have been paired, it is necessary to infer the time sequence of the three middle nodes based on the time sequence of the action identifiers at the two adjacent ends. The action identifier with earlier time is assigned to the earlier node, and the action identifier with later time is assigned to the later node. Finally, each number in the node sequence is bound to the corresponding action identifier, forming a matching structure with consistency in node order and action order, and the action node docking result is obtained.

[0035] S403: Based on nodes with missing positions in the action node docking results, extract the corresponding action identifiers of adjacent nodes and fill in the missing points to obtain the linkage response structure action sequence. First, identify data points in the node number sequence that do not have action labels. Then, search forward for each data point's immediately preceding node number to confirm its action label content. Next, search forward for its immediately following node number and determine if it has an action label. If both adjacent nodes have action labels, record their respective positions in the overall path sequence. Calculate the position of the missing node in the sequence between these two nodes and determine the proportional relationship between its position and the distance between the preceding and following action labels. Interpolate the action labels between the two action contents according to this proportion. That is, if the proportion of the number between the preceding and following action labels is close to that of the preceding node, assign the preceding action label; if it is close to that of the following node, assign the following action label; if they are relatively balanced, use the order of appearance of the two action labels as the criterion, assigning the action label that appears first. If there are multiple consecutive missing nodes in the number sequence, the nodes at both ends that have already been assigned actions need to be numbered. The numbers are extracted separately to form closed segment intervals. Then, the action identifiers are evenly distributed according to the segment length. By determining the position of the action identifiers at the upper and lower endpoints in the original sequence, the required action content for all nodes in the missing segment is determined. If there is a critical node with a large difference in action identifiers, such as the difference in action types between the upper and lower endpoints exceeding three categories, the endpoint filling method is not directly used. Instead, other numbered segments with similar adjacency relationships in the path are introduced for matching. Path segments with the same action combination order in other segments are searched, and the corresponding action content at the missing position is extracted and mapped to fill the current missing position. If no alternative segment is found in the entire path, the node identifier is left empty, and a separate list of data to be filled is generated. This list is then filled in with newly added data. Finally, all node numbers processed by the completion operation are filled with the corresponding action identifiers, forming a complete sequence matching chain between numbers and actions, resulting in a linkage response structure action sequence.

[0036] Please see Figure 6 The specific steps of S5 are as follows: S501: Based on the action identifier in the action sequence of the linkage response structure, the action instructions are sequentially assigned to the corresponding control execution nodes of the emission control link, and the action execution process of the control execution nodes is monitored, the corresponding action status data is collected, and the action response status set is obtained after the control execution nodes return the corresponding status information. First, in the emission control process, action commands are sequentially transmitted to the corresponding response points according to the order indicated by the action identifiers. The execution process requires first deconstructing the correspondence between the action identifier and the controlled object. For example, if the action identifier points to a specific emission valve, the valve number and opening / closing direction must be clearly stated in the command content. Then, the same deconstruction process is performed on subsequent points according to the node sequence. Next, the performance of each execution point after receiving the command is monitored at the emission control site. Quantifiable action status data needs to be collected for these performance metrics, such as the valve displacement angle after the command is triggered, the flow rate change value generated by the flow meter, and the sampling signal change at online sampling points due to command execution. The data collection process must be completed sequentially according to the action sequence, without skipping any points. When collecting action status data, the collection nodes need to be numbered to avoid confusion between different action positions. Then, the collected action status data is compared item by item, based on a direct comparison between two parameters. For example, comparing the expected position value of the command with the actual position value... The difference between actual action position values ​​is obtained by subtracting the two values. The difference is then used as the criterion for judgment if the difference exceeds a preset difference limit. This difference limit needs to be defined according to the characteristics of the emission control link. For example, the allowable deviation limit for emission valve action can be set to within five units, and the allowable deviation limit for flow meter action can be set to within three units. If the difference falls within the allowable range, the action of that node is recorded as completed. If the difference does not fall within the allowable range, it is recorded as the action not meeting the command requirements. The action results of each node are represented by independent data units. Then, the corresponding action identifier is matched with the collected data units one by one. In the matching process, it must be completed in the order of the action sequence so that the data units corresponding to the action identifier are arranged in the same order as the original action sequence. After all data is matched, an action response state set is formed. For example, if there are actions A, B, and C in the action sequence, the corresponding data units are a, b, and c, respectively. Finally, the data is output in the format Aa, Bb, Cc to obtain the action response state set.

[0037] S502: Based on the action state content in the action response state set, compare the action identifier text with the status field by referring to the action identifier content in the action sequence of the linkage response structure, analyze whether there are any incomplete action items in the response process, and obtain the missing action response information. First, the action identifiers are matched one-to-one with the text in the status field and extracted into two lists. List 1 contains sequentially arranged action identifier text items, and List 2 contains the corresponding action status text value items. Each pair of texts is matched by its sequence number. Then, the execution feedback information of each action is read from the status text items. For example, the status item of action A is "complete" or "not completed". "Completed" is defined as the text field completely matching the preset status string. For example, text with the same status words as "complete", "executed", "in place" is considered as completed. Text fields that do not match are defined as not completed. Next, the action identifiers in the action identifier sequence are compared with the status feedback values ​​to see if they are in a completed state. If the text feedback is not in the defined completed state list, it is recorded that the action has a missing response. During the comparison process, the action identifier index number at the corresponding position is recorded, thereby obtaining the action number corresponding to all incomplete actions in the entire action sequence. Suppose there is an action identifier sequence of A, B, C, D, and its corresponding status feedback is "complete", "no response", "executed", "not completed ... The process involves selecting "Failure" and "Completion" as the missing action locations, with the second and third items extracted as B and C respectively. The output list of missing actions is [B, C]. In the feedback field, the specific text content of each incomplete item is further extracted. For example, "No response" can be used as a classification label for the missing type, and "Execution failure" can be used to extract failure type keywords for further classification. In the above process, the keywords used in each action status field need to be extracted from a predefined list. They cannot rely on reasoning and must be explicitly matched with the literal content of the characters. The presence of keyword phrases is used as the judgment criterion. For example, "No response", "Stuck", and "No feedback" are all classified as missing. Note that case sensitivity must be consistent during text comparison. All text can be converted to lowercase before comparison. At the same time, spaces and special characters in the text are removed to ensure consistent comparison results. The position sequence and text results are then combined to form the corresponding missing action description table. Finally, the missing action and its status text content corresponding to each item are output to obtain the action response missing information.

[0038] S503: Based on missing action response information, extract the corresponding action identifier and state feedback content, trace each state according to the action sequence, and obtain the closed-loop state set of toxicity linkage response. First, extract the action identifier and its corresponding status feedback text from each missing data item. Rearrange the extracted results according to the action sequence. Each action identifier and its corresponding status field are combined into a traceability item. Within this traceability item, compare the status fields before and after its sequence position to analyze whether there are execution interruptions, repetitions, jumps, or abnormal states. Execution interruptions can be determined by identifying missing items between consecutive status fields. For example, if the sequence of action identifiers A, B, C, D, E has status fields of "Completed," "No Response," "Completed," "Completed," and "Completed," then action B is an interrupted item. Subsequent actions completing consecutively indicate that the interruption did not cause a link blockage. This type of status combination is categorized as a partially closed item. Then, output this type of combination sequentially and mark the interruption point and its preceding and following nodes on the corresponding item. State repetition can be determined by identifying two consecutive action identifiers whose corresponding status feedback fields are completely identical and whose values ​​point to a state word with no execution content, such as "Not Started" or "Initialized." If action C and... If all actions D are in this state, then this group is marked as a repeated but not triggered item in the output. If there are skipped items in the action identifiers in the sequence, it is confirmed by the break in the continuity of the numbers. For example, if the action numbers are A, B, D, and E, and the missing number C is presumed to be missing, then the status feedback fields between adjacent actions are compared to confirm whether it is a skipped trigger. If action B is "completed" and action D is "completed", but there is no action feedback item in between, it is marked as a skipped action. If the status field contains unexpected content, such as an empty string, an abnormal string, or an unrecognizable phrase, it is recorded as a status anomaly item and included in the subsequent marking results. After all the above status tracing steps are completed, all the interrupted, repeated, skipped, and abnormal status action items that have been traced and confirmed are integrated, and each item is appended with its index position in the action identifier sequence and a status value text description. At the same time, the content of its adjacent action items is retained for subsequent comparison. Finally, a set of tracing results containing action identifiers, status feedback, position indexes, adjacent action identifiers, and status descriptions is output, resulting in a closed-loop status set of toxic linkage response.

[0039] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A multi-level linkage method for real-time monitoring and emergency response of comprehensive toxicity of industrial wastewater, characterized in that, Includes the following steps: S1: Acquire spectral data from industrial wastewater inlet monitoring wells, terminal discharge outlets, and online sensing monitoring chambers; analyze the characteristics of the starting position and direction of movement of the main peak; and analyze the temporal relationship of the characteristics in the emission sequence based on the main peak shift time point to obtain a set of linkage response feature segments. S2: Based on the time period number in the linkage response feature map segment set, retrieve the corresponding discharge node in the drainage path map, advance the node downstream along the path, check the time sequence by comparing the upstream and downstream connection relationship, and obtain the pollution chain segment structure path group. S3: Based on the pollution chain segment structure path group, extract the continuous monitoring value sequence and toxicity response parameters, analyze the changing trend of node monitoring values ​​and toxicity characterization data, identify continuous segments with consistent changing trends, and define the start and end time range of toxicity disturbances in combination with the change amplitude of toxicity parameters to obtain a list of toxicity disturbance response time periods. S4: Based on the list of toxic disturbance response time periods, extract the corresponding response action identifiers, match the pollution chain segment path node numbers with the action identifiers, and fill in the unmatched node action numbers to obtain the linkage response structure action sequence. S5: Based on the action sequence of the linkage response structure, the action commands are sequentially issued to the emission control link, and the corresponding response status information is extracted to obtain the closed-loop state set of toxic linkage response.

2. The multi-level linkage method for real-time monitoring and emergency response of comprehensive toxicity of industrial wastewater according to claim 1, characterized in that, The linkage response feature segment set includes the starting position of the main peak, the direction of movement of the main peak, the continuous offset number of the main peak band, and the time axis position mark. The pollution chain segment structure path group includes the time period number, the emission node number, the drainage path connection sequence, and the node advancement sequence. The toxic disturbance response time period list includes the continuous monitoring value sequence, the direction of numerical change, the start and end time of the change segment, and the direction consistency indicator. The linkage response structure action sequence includes the action identifier, the node number correspondence, the action gap completion mark, and the action sequence. The toxic linkage response closed-loop state set includes the action execution command, the response point return status, the non-response identifier, and the status and action comparison sequence.

3. The multi-level linkage method for real-time monitoring and emergency response of comprehensive toxicity of industrial wastewater according to claim 1, characterized in that, The starting position of the main peak refers to the initial wavelength value of the spectral characteristics of the pollutant; The continuous monitoring value sequence refers to the sequence of monitoring data values ​​of emission nodes arranged in chronological order within a specified time period.

4. The multi-level linkage method for real-time monitoring and emergency response of comprehensive toxicity of industrial wastewater according to claim 1, characterized in that, The unmatched node refers to a node that has not yet been assigned an action number during the process of connecting the path node with the response action identifier. It needs to be completed by interpolating with neighboring nodes and replacing paragraphs. The emission control component refers to the actual operational part that executes the linkage response action command.

5. The multi-level linkage method for real-time monitoring and emergency response of comprehensive toxicity of industrial wastewater according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: Acquire raw spectral data from water intake monitoring wells, terminal outlets, and online sensing monitoring chambers; extract the initial wavelength value of the main peak at each monitoring point; track the wavelength change trend; identify the spectral region where the main peak continuously shifts within a continuous time axis; and obtain a set of continuously shifted main peak bands. S102: Based on the main peak continuous offset band set, extract the acquisition time corresponding to the spectral region at the point, then filter the spectral data segments that show the offset trajectory according to the time, and aggregate the parts with overlapping bands and time progression relationship to obtain the main peak advancement band sequence. S103: Based on the main peak advancing band sequence, compare the band direction of the spectral data segments in the differentiated monitoring points according to the time advancement relationship, extract the continuous data segment group with consistent offset characteristics, and string them together according to the position of the band corresponding to the time to obtain the linkage response feature map segment set.

6. The multi-level linkage method for real-time monitoring and emergency response of comprehensive toxicity of industrial wastewater according to claim 1, characterized in that, The specific steps of S2 are as follows: S201: Based on the time period number and monitoring point number corresponding to the linkage response feature map segment set, retrieve the node distribution information in the drainage path map, and according to the correspondence between the monitoring point number and the discharge node number, retrieve the coordinate position of the discharge node corresponding to the time period number in the path map, compare the node coordinates according to the order of the time period numbers, and obtain the path node positioning dataset. S202: Based on the path node positioning dataset, call the connection information in the path graph, extract the connection direction of the node pairs with continuous direction from the graph coordinates, analyze the line segment direction of the node pairs in the path graph, and obtain the node flow direction construction relationship set. S203: Based on the node flow direction, construct a relationship set, starting with the node located at the beginning of the time number in the graph, and advance the nodes in sequence according to the connection direction. Connect and extend the nodes according to the path direction to obtain the pollution chain segment structure path group.

7. The multi-level linkage method for real-time monitoring and emergency response of comprehensive toxicity of industrial wastewater according to claim 1, characterized in that, The specific steps for S3 are as follows: S301: Based on the time period corresponding to the emission node in the pollution chain structure path group, extract the monitoring data and toxicity response parameters of the node in the corresponding time period, arrange the monitoring values ​​and toxicity characterization data corresponding to the node in chronological order, and obtain a continuous monitoring value sequence set. S302: Based on the continuously monitored numerical sequence set, compare the direction of change of the monitored values ​​and the trend of change of the toxicity parameters between consecutive time points in the data sequence, identify continuous segments with the same direction of change, and locate the start and end positions of the continuous segments on the time axis to obtain the time range set of trend change segments. S303: Based on the set of time ranges of the trend change intervals, screen the monitoring data segments with overlapping time ranges from the emission nodes, analyze whether the direction of numerical change of the differentiated nodes in the corresponding time period is consistent with the trend of toxicity parameter change, and define the time range of toxicity disturbance according to the magnitude of toxicity parameter change to obtain a list of toxicity disturbance response periods.

8. The multi-level linkage method for real-time monitoring and emergency response of comprehensive toxicity of industrial wastewater according to claim 7, characterized in that, In the process of comparing the direction of numerical differences at consecutive time points within the data sequence: identify data pairs that show consistency in both magnitude and direction of change between adjacent time points, and connect them sequentially into a continuous sequence according to the chronological order; In the process of identifying continuous segments with the same direction of numerical change: analyze the distribution of data pairs with the same direction on the time axis, filter out mutually continuous and continuous sequence segments, and divide the time limits at the beginning and end of the sequence segments; In the process of filtering data segments with overlapping time ranges from nodes: by comparing the time range of sequence segments in emission nodes, data segments with overlapping time ranges are extracted, and the direction of change of emission nodes in overlapping time ranges is analyzed to see if they are consistent.

9. The multi-level linkage method for real-time monitoring and emergency response of comprehensive toxicity of industrial wastewater according to claim 1, characterized in that, The specific steps of S4 are as follows: S401: Based on the time range in the list of toxic disturbance response periods, extract the action identifiers associated with the time periods from the response actions, and compare them according to the chronological order to obtain the action sequence matching set; S402: Based on the node number sequence in the pollution chain path, match each group of action identifiers in the action sequence matching set with the node number, locate the action content to which the node belongs, and obtain the action node docking result; S403: Based on the nodes with missing positions in the action node docking results, extract the corresponding action identifiers of adjacent nodes and fill in the missing points to obtain the linkage response structure action sequence.

10. The multi-level linkage method for real-time monitoring and emergency response of comprehensive toxicity of industrial wastewater according to claim 1, characterized in that, The specific steps of S5 are as follows: S501: Based on the action identifier in the action sequence of the linkage response structure, the action instructions are sequentially assigned to the control execution nodes corresponding to the emission control link, and the action execution process of the control execution nodes is monitored, the corresponding action status data is collected, and the action response status set is obtained after the control execution nodes return the corresponding status information. S502: Based on the action state content in the action response state set, compare the action identifier text with the state field by referring to the action identifier content in the action sequence of the linkage response structure, analyze whether there are any incomplete action items in the response process, and obtain the missing action response information. S503: Based on the missing action response information, extract the corresponding action identifier and state feedback content, trace each state according to the action sequence, and obtain the closed-loop state set of toxic linkage response.

Images