A water environment online monitoring system

Abstract

This invention relates to the field of water quality analysis technology, specifically to an online water environment monitoring system. This system includes acquiring and aggregating the directional indicators of ammonia nitrogen, total phosphorus, and pH values; identifying trend convergence zones; dividing transition segments and extracting trend features; screening cross-trend combinations; summarizing jump nodes and identifying interference points; marking directional mismatches and generating off-track combinations. In this invention, by continuously extracting changes in key indicators such as ammonia nitrogen, total phosphorus, and pH values, the direction of load changes and synergistic trends are identified. Further comparison of indicator fluctuation rhythms and jump nodes is conducted to extract pollution trend structures and directional mismatch combinations, effectively revealing potential abnormal segments and interference points. This enhances the depth of identification and temporal analysis capabilities of pollution dynamics during reclaimed water transportation. Consequently, urban residential water, after treatment to form reclaimed water, is delivered to industrial parks with stable and qualified water quality, providing more accurate and structured monitoring support for the compliance of industrial water quality.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water quality analysis, in particular to a water environment online monitoring system. BACKGROUND

[0002] The technical field of water quality analysis includes quantitative detection and identification of physical, chemical and biological parameters in water bodies, aiming to evaluate and monitor water environment quality, and to ensure the stability of the ecological system and the safety of human water use. The core content of this technical field is to identify and measure the main suspended substances in water through multiple types of sensors, analytical instruments and detection processes. Commonly detected items include dissolved oxygen, chemical oxygen demand, total phosphorus, total nitrogen, ammonia nitrogen, heavy metal ions, pH value and turbidity, etc. Water quality analysis technology usually combines online sampling, data collection, result analysis and remote communication systems to form an integrated monitoring process, and is widely used in real-time environmental monitoring of various water bodies such as surface water, groundwater, industrial wastewater and domestic sewage.

[0003] Among them, the water environment online monitoring system refers to a data collection and analysis system for real-time water quality detection of urban reclaimed water and other water bodies. This system mainly covers water sampling, water sample pretreatment, parameter determination, and transmission and storage of monitoring data. It uses multiple detection methods based on electrochemistry, spectrophotometry, ion selective electrode method, etc. to complete the quantitative monitoring of ammonia nitrogen, total phosphorus, total nitrogen, chemical oxygen demand, dissolved oxygen and other indicators in water, and transmits the results to the data center using remote communication methods to realize online analysis and recording. The system is usually independently deployed with solar power supply units, self-cleaning structures, data recording devices, etc. to complete the continuous observation and data collection of dynamic environmental parameters of surface water.

[0004] In the existing water environment online monitoring process, although it has the ability to quantitatively detect ammonia nitrogen, total phosphorus, dissolved oxygen and other indicators in reclaimed water, it can realize real-time uploading and centralized storage of data, but its analysis and processing method mainly relies on static threshold judgment, lacks in-depth analysis of data trend structure, and it is difficult to effectively identify potential pollution fluctuation risks in long-distance transportation. In the process of urban reclaimed water being transported from the sewage treatment plant to the industrial park, the water body may cause short-term coordinated fluctuations or mutations of pollutant indicators due to pipe residues, pressure mutations or environmental disturbances, but the existing system does not analyze the direction consistency and trend period between indicators, and cannot identify pollution trend turning points or abnormal change points across indicators. For example, under high temperature conditions in summer, water quality fluctuations in long-distance water transportation sections may be intensified, and the single-point determination mechanism of the existing system is difficult to capture trend changes in time, which is prone to false positives or false negatives. The above shortcomings weaken the practical value of the system in dynamic supervision, early warning judgment and pollution tracking, and reduce the reliability of water quality protection for industrial users in the process of reclaimed water reuse. SUMMARY

[0005] The present application aims at solving the problems existing in the prior art and provides a water environment online monitoring system.

[0006] In order to achieve the above-mentioned purpose, the present application adopts the following technical scheme: a water environment online monitoring system comprises:

[0007] The pollution load calculation module obtains continuous data of ammonia nitrogen, total phosphorus and pH value, calculates the difference between adjacent time periods and marks the direction, combines the multi-index direction identification, and outputs the suspended matter water load convergence section label;

[0008] The abnormal section division module extracts the index change amplitude according to the suspended matter water load convergence section label, identifies the direction turning point and segments according to the continuous direction, records the trend intensity and duration, and forms an abnormal pollution paragraph trend structure table;

[0009] The rate trajectory screening module extracts the direction and rhythm of ammonia nitrogen and chemical oxygen demand based on the abnormal pollution paragraph trend structure table, and screens the cross-trend pollution index group segment with close starting point and consistent direction;

[0010] The jump coincidence determination module extracts the high-frequency coincidence time point based on the index jump node in the cross-trend pollution index group segment, judges whether it is an interference influence point, and generates a multi-index common jump node set;

[0011] The index off-track identification module analyzes the direction of total phosphorus, ammonia nitrogen, chemical oxygen demand and dissolved oxygen at the three time points before and after according to the multi-index common jump node set, and if two or more indexes are continuously mismatched, it is determined as a off-track combination, and an environmental monitoring result is generated.

[0012] As a further scheme of the present application, the suspended matter water load convergence section label comprises the index difference value of adjacent time periods, the unified direction identification, the consistency frequency of change direction, and the trend time period range; the abnormal pollution paragraph trend structure table comprises the direction turning point position, the direction fragment sequence, the index fluctuation amplitude, and the trend duration; the cross-trend pollution index group segment comprises the combination relationship of ammonia nitrogen and chemical oxygen demand, the index trend direction, the trend period proximity, and the starting point position consistency; the multi-index common jump node set comprises the index direction change time point, the jump node coincidence frequency, and the cross-index jump characteristics; and the off-track combination comprises the direction mismatching situation of total phosphorus, ammonia nitrogen, chemical oxygen demand and dissolved oxygen, the mismatching time period, and the continuous mismatching index group.

[0013] As a further scheme of the present application, the pollution load calculation module comprises:

[0014] The index difference calculation submodule calculates the numerical difference of each index in adjacent time periods based on the continuous detection records of ammonia nitrogen, total phosphorus and pH value, sorts the numerical difference in chronological order, obtains the direction identifier of the index at the differentiated time point, and generates an index direction identifier sequence;

[0015] The trend direction classification submodule calls the index direction identifier sequence, extracts the direction identifiers of multiple indexes at the same time point, classifies the time points with consistent direction identifiers by comparing the consistency of the index direction, accumulates the frequency of time points with consistent direction, and generates a consistency direction frequency distribution;

[0016] The convergence section identification submodule sets a frequency threshold and performs screening according to the consistency direction frequency distribution, extracts a time section with a consistency direction frequency exceeding the threshold, and uses the formula:

[0017] ;

[0018] The operation obtains a convergence degree coefficient, judges whether the convergence degree coefficient exceeds a section classification threshold, and obtains a suspended matter water load convergence section label;

[0019] wherein, represents the suspended matter water direction identifier of the i-th period, represents the ammonia nitrogen direction identifier, represents the total phosphorus direction identifier, represents the pH value direction identifier, is the direction consistency frequency in the i-th period, is the number of periods with consistent direction, is the interval time between every two time points, is the convergence degree coefficient. As a further scheme of the present application, the abnormal section division module comprises:

[0020] The operation acquisition submodule extracts the original value data sequence of the suspended matter water index in the time section based on the time section defined in the suspended matter water load convergence section label, detects the difference sequence of the index original value between adjacent time nodes, calculates the change amplitude of the difference, and obtains the index change amplitude value;

[0021] The direction identification submodule judges the sign change trend according to the index change amplitude value, continuously judges the direction identifier sequence, compares the difference between the current direction identifier and the adjacent period direction identifier, identifies whether there is a positive-negative value conversion phenomenon as a direction turning point, and obtains a direction turning position point set;

[0022]

[0023] ​​The structure construction submodule calls the set of directional turning points, re-divides the segments of directional turning points into units of each directional segment, and calculates the fluctuation range of suspended solids water index changes and the duration within each segment using the following formula:

[0024] ;

[0025] The fluctuation trend intensity is integrated with the length value to obtain the trend structure table of abnormal pollution paragraphs;

[0026] in, This indicates the intensity value of the abnormal pollution trend in the paragraph. Indicates the first The fluctuation range of suspended solids water index in each directional segment This represents the average fluctuation range value across all directional segments. Indicates the first The duration of a segment in each direction. Indicates the total number of directional segments. Representing the The duration of each segment, This represents the average fluctuation range.

[0027] As a further aspect of the present invention, the rate trajectory filtering module includes:

[0028] The trend extraction submodule, based on the segment division in the abnormal pollution segment trend structure table, obtains the start and end time information of each segment, and collects the original concentration value sequence of ammonia nitrogen and chemical oxygen demand within the corresponding time period. It calculates the change amplitude of adjacent values ​​every hour. That is, if the current value minus the previous value is greater than zero, it is marked as positive; otherwise, it is marked as negative. Time periods with the same continuous direction marking are divided into a rhythm cycle segment. The module constructs the direction sequence and rhythm cycle length set of each indicator within the segment and obtains the indicator trend cycle sequence group.

[0029] The cycle comparison submodule calls the indicator trend cycle sequence group, extracts the rhythm cycle length values ​​of the three indicators in each segment, and uses a multi-indicator pairing method to perform normalized difference analysis on the cycle value sequences of any two indicators. It also introduces cycle sequence perturbation to smooth and correct the cycle difference using the following formula:

[0030] ;

[0031] The calculation obtains the rhythmic cycle convergence value of the indicator in the current paragraph. If the period is less than 0.5 and close to the threshold, it is marked as a period close pair and a combination of period close pairs is generated.

[0032] in, a rhythm cycle convergence value, 、 a first segment, 、 a rhythm cycle value of the index, a first segment, a rhythm cycle value of the index an average cycle of all segments, a total number of segments;

[0033] The cross identification submodule extracts the starting value difference and calculates the absolute difference in each segment according to the cycle convergence of the index combination, and if the first direction identifiers of the two indexes are consistent, the cross trend segment is identified, and the corresponding index combination and segment number are recorded to obtain the cross trend pollution index segment.

[0034] As a further scheme of the present application, the jump coincidence determination module comprises:

[0035] The jump node statistics submodule calls all index combinations in the cross trend pollution index segment, extracts the trend direction sequence in the corresponding segment for each index combination, labels the jump nodes of the time sequence, and repeats the operation to extract the ammonia nitrogen and chemical oxygen demand, counts the jump node positions of the indexes in the segment, and obtains a direction jump time point set;

[0036] The coincidence frequency induction submodule extracts the nodes where two or three indexes simultaneously change direction at the same time point according to the direction jump time point set, counts and sorts the coincidence jump node frequencies appearing in all segments, and uses the formula:

[0037] ;

[0038] The operation obtains a jump coincidence measurement value, and determines whether the coincidence degree exceeds a set coincidence frequency threshold to obtain a high-frequency coincidence jump node;

[0039] wherein, a node coincidence measurement value, a node coincidence frequency of a first segment, a node coincidence frequency average of all segments, a jump coincidence number of the node in each index in the segment, a number of segments with consistent directions;

[0040] The interference point determination sub-module determines whether there is a case of consistent direction and change amplitude greater than the corresponding index jump reference value at the same time point according to the high-frequency coincidence jump node, the trend consistency and the change amplitude of the corresponding time point in three indexes, if the determination result is that both conditions are met, the time point is classified as an interference influence point, and a multi-index common jump node set is established.

[0041] As a further scheme of the application, the index derailment identification module comprises:

[0042] The index trend extraction sub-module obtains the ammonia nitrogen and chemical oxygen demand original index sequences in each paragraph based on the segment start and end records in the multi-index common jump node set, extracts the sequences in each paragraph according to time sequence and calculates the adjacent value direction identifier, judges the positive and negative changes according to the difference value, constructs the direction sequence, counts the time length of the same direction continuous appearance in the paragraph and divides the rhythm period sequence, and generates the three index trend direction sequence and rhythm sequence set;

[0043] The period proximity judgment sub-module calls the three index trend direction sequence and rhythm sequence set, respectively judges the rhythm period values of ammonia nitrogen and chemical oxygen demand in each paragraph in pairs, adopts the formula:

[0044] ;

[0045] The operation obtains the rhythm period difference degree of any two index combinations in the paragraph, and compares the difference degree with the set period proximity threshold value, if it is less than 0.2, it is judged that the period is approaching, and then the period approaching combination group is obtained;

[0046] Wherein, represents the rhythm period difference degree, and respectively represent the period values of the first segment in the first , index, represents the period disturbance harmonic coefficient of the first segment, is the number of segments with consistent direction;

[0047] The cross segment identification sub-module judges whether the start data difference value of the corresponding index combination in each segment is less than the set proximity value based on the period approaching combination group, and compares the first value of the direction sequence for consistency, if both conditions are met, it is marked as a cross trend index segment, all index combinations and paragraph index values that meet the conditions are output, and the environmental monitoring result is obtained.

[0048] Compared with the prior art, the application has the advantages and positive effects that:

[0049] In the present application, by continuously extracting the changes of key indicators such as ammonia nitrogen, total phosphorus, pH value, the load change direction and the synergistic trend are recognized, the index fluctuation rhythm and the jump node are further compared, the pollution trend structure and direction mismatch combination are extracted, the potential abnormal paragraph and interference point are effectively revealed, and the identification depth and time sequence analysis ability of the pollution dynamics in the reclaimed water transportation process are strengthened, so that the treated water for urban residents is formed into reclaimed water, and the water quality is stable and qualified after being transported to the industrial park, and more accurate and structured monitoring support is provided for the water quality compliance of industrial water. BRIEF DESCRIPTION OF DRAWINGS

[0050] Figure 1 The system flowchart of the present application is shown in the figure;

[0051] Figure 2 The pollution load calculation module flowchart of the present application is shown in the figure;

[0052] Figure 3 The abnormal section division module flowchart of the present application is shown in the figure;

[0053] Figure 4 The rate trajectory screening module flowchart of the present application is shown in the figure;

[0054] Figure 5 The jump coincidence determination module flowchart of the present application is shown in the figure;

[0055] Figure 6 The index off-track identification module flowchart of the present application is shown in the figure. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical scheme and advantages of the present application more clear, the present application is further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application, and are not used to limit the present application.

[0057] In the description of the present application, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only used to facilitate the description of the present application and simplify the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, in the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise specifically limited.

[0058] Referring to Figure 1 A water environment online monitoring system comprises:

[0059] The pollution load calculation module obtains continuous detection records of ammonia nitrogen, total phosphorus and pH value, sorts difference values of each index in adjacent time periods according to a time line, marks directions, performs merging processing on direction identifications of multiple indexes at the same time point, judges whether the variation directions of the suspended solids are consistent, sets a frequency threshold, classifies time periods with high consistency frequency as a load convergence stage, and outputs a suspended solids water load convergence section label.

[0060] The abnormal section division module extracts a variation amplitude of an original value of an index in a time period based on the time period defined in the suspended solids water load convergence section label, compares direction identification continuity, identifies whether there is a direction turning point as a section structure division basis, segments the section according to a direction segment in the presence of the turning point, records an index fluctuation trend strength and a continuous length section by section, and generates an abnormal pollution paragraph trend structure table.

[0061] The rate trajectory screening module extracts trend directions and variation rhythms of ammonia nitrogen and chemical oxygen demand according to start and end records of the section in the abnormal pollution paragraph trend structure table, judges whether trend periods of the three indexes in each section are close, identifies whether there is a cross section with a close start point and consistent direction, and outputs an index combination and a section that meet the conditions as a cross trend pollution index group section.

[0062] The jump coincidence judgment module calls the combination appearing in the cross trend pollution index group section, counts time points at which the direction changes in the time period, generalizes a coincidence jump node with the highest appearance frequency, judges whether there is a cross-index simultaneous jump feature, classifies the node as an interference affecting point if the feature is present, and forms a multi-index common jump node set.

[0063] The index off-track identification module extracts variation directions of total phosphorus, ammonia nitrogen, chemical oxygen demand and dissolved oxygen in three time points before and after the time point in the multi-index common jump node set, marks whether there is a trend of inconsistent directions between multiple continuous time points, classifies as an off-track combination if there are two or more indexes that appear continuous direction mismatch, and generates an environmental monitoring result.

[0064] The suspended solids water load convergence section label includes index difference values in adjacent time periods, unified direction identifications, variation direction consistency frequencies and convergence time period time ranges. The abnormal pollution paragraph trend structure table includes direction turning point positions, direction segment sequences, index fluctuation amplitudes and trend continuous lengths. The cross trend pollution index group section includes combination relationships of ammonia nitrogen and chemical oxygen demand, index trend directions, trend period closeness and start point position consistency. The multi-index common jump node set includes index direction change time points, jump node coincidence frequencies and cross-index jump features. The off-track combination includes direction mismatch conditions of total phosphorus, ammonia nitrogen, chemical oxygen demand and dissolved oxygen, mismatch time periods and continuous mismatch index groups.

[0065] Referring to Figure 2 , the pollution load calculation module comprises:

[0066] The index difference calculation submodule calculates the numerical difference of each index in adjacent time periods based on the continuous detection records of ammonia nitrogen, total phosphorus and pH value, sorts them in chronological order, obtains the direction identifier of the index at the differentiated time point, and generates the index direction identifier sequence.

[0067] First, read the hourly water quality data from January 1 to January 2, 2025 in the set monitoring system, set the collection frequency of each type of index to 1 hour, i.e. 24 records per day, select the fixed monitoring point set in the test river section, and obtain the values of the four types of indexes at each time point, for example, at 08:00, the ammonia nitrogen is 1.6 mg / L, the total phosphorus is 0.25 mg / L, and the pH value is 7.10, at 09:00, they are 5.6 mg / L, 1.8 mg / L, 0.30 mg / L, and 7.15 respectively, then calculate the numerical difference of each index between 08:00 and 09:00, the difference values are 0.4 mg / L, 0.2 mg / L, 0.05 mg / L, and 0.05 respectively, wherein each difference value represents the numerical variation of the index in the adjacent time period, and the unified direction mark is made for the group of difference values, if the difference value is greater than 0, the direction is marked as +1, if it is equal to 0, it is marked as 0, and if it is less than 0, it is marked as -1, in this example, the direction identifiers of the four indexes are all +1, forming a direction identifier array [+1, +1, +1, +1], the direction array is arranged in chronological order, and the index change at each time period is recorded in turn, the index values at the subsequent time points such as 10:00 are calculated in the same way, and a complete time sequence direction array is formed, each time point saves the direction information of the four indexes, which serves as the basis for subsequent consistency direction judgment, and finally the index direction identifier sequence is obtained.

[0068] The trend direction classification submodule calls the index direction identifier sequence, extracts the direction identifiers of multiple indexes at the same time point, compares whether the index directions are consistent, classifies the time points with consistent direction identifiers, and accumulates the frequency of time points with consistent direction, to generate the consistency direction frequency distribution.

[0069] Read four index direction arrays under each time node from it, count the number of the same direction value in the group by traversing each time point direction identification group, for example, in a certain time point direction group is [+1, +1, 0, +1], in which +1 direction accounts for 75%, which meets the direction consistency judgment standard, set the consistency judgment threshold as the number of the same direction is greater than or equal to 3, that is consistent, by traversing to judge whether each time point meets the standard, if it meets, it is recorded as consistent, otherwise it is inconsistent, for example, 10:00 direction group is [+1, +1, +1, +1], recorded as consistent, 11:00 direction group is [+1, -1, 0, +1], recorded as inconsistent, and the operation is carried out to form a group of Boolean value sequence such as [1, 1, 0, 1, 0, 0, 1], which represents the direction consistency state of the corresponding time point, and the cumulative frequency record mode is constructed, if the direction consistency mark is 1 in the continuous time point, the count is accumulated, for example, the first and second positions in the above sequence are consistent, so the continuous frequency is 2, when the third position is inconsistent, the count is cleared, and the accumulation is restarted from the next consistent point, and the obtained frequency is used to judge the strength of the continuous convergence section, for example, if the frequency reaches 4 in a certain time period, it is considered that the convergence is strong, and the consistency direction frequency distribution is generated by the operation.

[0070] The convergence section identification submodule sets the frequency threshold and filters according to the consistency direction frequency distribution, extracts the time section with consistency direction frequency exceeding the threshold, and uses the formula:

[0071] ;

[0072] The operation obtains the convergence degree coefficient, judges whether the convergence degree coefficient exceeds the section classification threshold, and obtains the suspended matter water load convergence section label;

[0073] Wherein, represents the direction identification of the th period, represents the ammonia nitrogen direction identification, represents the total phosphorus direction identification, represents the pH value direction identification, is the direction consistency frequency in the th period, is the number of direction consistent periods, is the interval time between each two time points, is the convergence degree coefficient;

[0074] The time section with high continuous frequency is extracted, and the frequency threshold is set to 3, that is, only when the direction is consistent for at least 3 times continuously in a certain time section, it is considered as a convergence section. The continuous consistent time points in a certain time section are set to the 5th to 9th hour, the sum of the directions of the four indicators corresponding to each hour in this section is read and the absolute value is taken, if the sum of the directions of the five time points is 4, 4, 3, 4, and 4 respectively, the corresponding direction consistency frequency is 4, 4, 3, 4, and 4, the time interval Δt is 1 hour, and the total number of time points n=5, then the formula is calculated:

[0075] ;

[0076] First, calculate each product:

[0077] i=1: |1+1+1+1|× =4×2=8;

[0078] i=2: |1+1+1+1|× =4×2=8;

[0079] i=3: |1+1+1+0|× ≈3×1.732≈5.196;

[0080] i=4: |1+1+1+1|× =4×2=8;

[0081] i=5: |1+1+1+1|× =4×2=8;

[0082] Second, calculate the sum: 8+8+5.196+8+8=37.196;

[0083] Third, bring in the denominator: 5×1=5;

[0084] Fourth, calculate the convergence coefficient:

[0085] ;

[0086] The threshold is set to ≥7.0 as the standard for judging the convergence section, and the calculation result =7.4392, which meets the threshold condition, which indicates that the variation of the direction of the four types of suspended solids in water in this time section shows a consistent trend, so it can be identified as a suspended solids water load convergence section label. By weighting the sum of the absolute values of the direction of the suspended solids and the square root of the consistency frequency, the strength of the consistent direction is effectively reflected, and the normalized index is constructed by combining the number of directions and the time distribution, so as to accurately divide the convergence section and improve the responsiveness and robustness of the load trend judgment.

[0087] Referring to Figure 3 The abnormal section division module comprises:

[0088] The operation acquisition submodule extracts the original value data sequence of the suspended matter water index in the time period defined in the convergence section label of the suspended matter water load, detects the difference value sequence of the original value of the index between adjacent time nodes, calculates the change amplitude of the difference value, and obtains the index change amplitude value;

[0089] First, the specific time range is extracted from the label, for example, the convergence section corresponding to a certain label is from 0 o'clock on July 1, 2024 to 24 o'clock on July 5, 2024, the system reads the start and end time and calls the complete record data in the original data source of the suspended matter water index according to the same, in the record situation of data frequency of once per hour, 120 original index values of time points in the 5 days are extracted, each time point corresponds to a suspended matter concentration value, for example, the suspended matter concentration at t1=2024-07-01-00:00 is 116.4 mg / L, the suspended matter concentration at t2=2024-07-01-01:00 is 117.9 mg / L, and so on to form an original concentration sequence, then the concentration values of adjacent two time points are difference calculated to form a difference value sequence, for example, the first group of difference values is 117.9-116.4=1.5 mg / L, and the difference value between 117.9 and 119.6 is 1.7 mg / L, the whole sequence forms a difference value sequence with a length of 119, in order to further identify the fluctuation of the difference value sequence, the change amplitude division threshold is set, the low amplitude threshold is set to 1.0 mg / L, which represents the natural fluctuation or weak change range, and the high amplitude threshold is set to 4.0 mg / L, which represents the strong fluctuation and violent change situation, the setting of the threshold is based on the statistical results of the variation amplitude of the data in the past three months, wherein 90% of the fluctuations are concentrated between 1.0 and 4.0, less than 1.0 mg / L is regarded as weak change, and more than 4.0 mg / L is regarded as violent change, on this basis, all the difference values are classified according to this interval, the fluctuation intensity level label corresponding to each time point is obtained and a change amplitude value sequence is formed.

[0090] The direction recognition submodule judges the sign change trend according to the index change amplitude value, judges the continuity of the direction identification sequence, compares the difference between the current direction identification and the adjacent period direction identification, identifies whether there is a positive-negative value conversion phenomenon as a direction turning point, and obtains a set of direction turning position points;

[0091] First, a directional comparison is performed on each pair of adjacent original suspended solids concentration values. If the value at the later time point is higher than that at the previous time point, it is marked as positive (i.e., an upward trend); if it is lower, it is marked as negative (i.e., a downward trend); if the two are equal, it is marked as unchanged. This operation covers the entire sequence to form a directional identifier sequence. For example, if the sequence contains five values: 116.4, 117.9, 119.6, 119.6, and 118.3, the directional identifiers are, in order, rising, rising, unchanged, and falling. Next, a continuity analysis is performed on this directional identifier sequence, comparing each item to see if the current position is consistent with the previous item in direction. For example, if the current position is rising and the previous item was also rising, it is considered continuous; otherwise, it is considered a direction reversal point. During the traversal, if the direction changes from upward to downward, or from downward to upward, or from having a direction to having no direction, it is recorded as a direction reversal point. For example, if a change from upward to no change is detected at t4=2024-07-01-03:00, it is recorded as a potential reversal point. Then, the fluctuation range before and after the reversal point is judged. If the difference before and after exceeds the set significant change threshold, such as 2.5mg / L, the reversal point is included in the set of valid reversal points. This threshold is derived from the 30% maximum change range in actual detection as the benchmark for defining the validity of the reversal. By tracing the consistency of the direction of the three time points before and the two time points after the reversal point, it is ensured that the point is not a short-term fluctuation that is mistakenly judged as a trend reversal point.

[0092] The structure construction submodule calls the set of directional turning point locations, re-divides the segments of directional turning points into units of each directional segment, and calculates the fluctuation range of suspended solids water index changes and the duration within each segment using the following formula:

[0093] ;

[0094] The fluctuation trend intensity is integrated with the length value to obtain the trend structure table of abnormal pollution paragraphs;

[0095] in, This indicates the intensity value of the abnormal pollution trend in the paragraph. Indicates the first The fluctuation range of suspended solids water index in each directional segment This represents the average fluctuation range value across all directional segments. Indicates the first The duration of a segment in each direction. Indicates the total number of directional segments. Representing the The duration of each segment, This represents the average fluctuation range.

[0096] Parameter definition and acquisition method description:

[0097] : represents the fluctuation interval value of suspended solids (SS) change in the i-th direction segment, unit: mg / L. The difference between the maximum and minimum values in each direction segment is calculated as the fluctuation interval by the daily average sequence obtained by the water quality detection standard of China's ecological environment department HJ91.1. The recommended daily collection frequency is 1 time / 2 hours, continuously for 7 days;

[0098] : represents the average fluctuation interval value of all direction segments, that is, the arithmetic mean of all ;

[0099] : represents the duration of the i-th direction segment, expressed in the number of days of continuous monitoring within the segment. The segment is divided by the turning point of the water flow direction, for example, by determining the direction mutation point through the change of flow velocity vector;

[0100] : represents the total number of direction segments. It is determined by the number of direction turning points in data analysis.

[0101] The numerical setting is based on the acquisition method:

[0102] Based on the public standard (HJ91.1-2020) and research investigation, the following values are set:

[0103] Direction segment number ;

[0104] Suspended solids fluctuation value in the first segment mg / L (data from the maximum and minimum daily value difference within 7 days);

[0105] Second segment mg / L;

[0106] Third segment mg / L;

[0107] Average fluctuation value mg / L;

[0108] The duration of the segment is respectively: days, days, days;

[0109] Formula step-by-step operation:

[0110] (a) Numerator calculation:

[0111] ;

[0112] (b) Denominator calculation:

[0113] ;

[0114] (c) Final value calculation:

[0115] ;

[0116] Calculation results:

[0117] Result 4.03 indicates that the pollution trend fluctuation intensity of the direction turning area in this section of water body is high. This value integrates the fluctuation amplitude and the duration of the section, and the larger the value, the more intense and persistent the pollution change. This result shows that although the fluctuation amplitude of the second section of the direction section is low, it deviates significantly from the average value, resulting in a strong contribution in the weight calculation.

[0118] See Figure 4 , the rate trajectory screening module includes:

[0119] The trend extraction submodule obtains the start and end time information within each section based on the section division in the abnormal pollution paragraph trend structure table, collects the original concentration value sequence of ammonia nitrogen and chemical oxygen demand in the corresponding time period, calculates the change amplitude of adjacent values every hour, i.e., if the current value minus the previous value is greater than zero, it is marked as positive, otherwise as negative, and the time period with the same continuous direction marker is divided as a rhythm period section, to construct the direction sequence and rhythm period length set of each index within the paragraph, and obtain the index trend period sequence group;

[0120] The start and end times of each paragraph are extracted in turn, and the original concentration data of ammonia nitrogen and chemical oxygen demand in the corresponding time period are obtained, and the hourly change trend is recorded, for example, the start and end times of a paragraph are July 5, 2024 08:00 to 12:00, and within this section, the hourly data of suspended solids water is [4.2, 4.5, 4.7, 4.3, 4.1] mg / L, the ammonia nitrogen is [1.1, 1.2, 1.3, 1.1, 1.0] mg / L, and the chemical oxygen demand is [18, 19, 19.5, 19, 18.5] mg / L. First, calculate the difference between adjacent hours for each group of data to obtain the ammonia nitrogen change sequence [+0.1, +0.1, -0.2, -0.1] and the chemical oxygen demand change sequence [+1, +0.5, -0.5, -0.5]. Based on these difference values, determine the direction identifier for each hour, and construct the direction identifier sequence such as [+, +, -, -]. Further identify the length of the time period with the same continuous direction, if the change is positive for two consecutive hours, then the two hours are a rhythm period, and the rhythm period sequence of each index within the paragraph is divided in this way, for example, the suspended solids water in this section is divided into two periods, positive for 2 hours and negative for 2 hours. Similarly, ammonia nitrogen and chemical oxygen demand are also divided into positive and negative rhythm sections, and the number and length of the rhythm periods of each index in this section are recorded as trend information. Finally, the index trend period sequence group is obtained.

[0121] The period comparison sub-module calls the index trend period sequence group, sequentially extracts the rhythm period length value of each of the three indexes in each paragraph, adopts a multi-index pairing combination method, performs normalized difference value analysis on the period value sequence of any two indexes, introduces period sequence disturbance to smooth correct the period difference value, and uses the formula:

[0122] ;

[0123] The operation obtains the rhythm period convergence value of the index pair under the current paragraph, and if the value is less than the period proximity threshold value of 0.5, the period is marked as close, and a period approaching combination is generated.

[0124] wherein, represents the rhythm period convergence value, , is the rhythm period value of the i-th index in the j-th paragraph, is the period disturbance of the index pair in the j-th paragraph, is the average period of the index, is the total number of paragraphs. The rhythm period values of the indexes in the paragraph are extracted, the period of the ammonia nitrogen and chemical oxygen demand, the ammonia nitrogen and chemical oxygen demand, and the three index pairs are compared, the rhythm period convergence formula is used to calculate the value, and it is determined whether the period is close. Taking suspended solids (SS) and ammonia nitrogen (NH3-N) as an example, the rhythm period length sequence in this paragraph is

[0125] , , , ; The period disturbance value is assumed to be , , the average period of the index is , and the total number of paragraphs is

[0126] . The formula is: ;

[0127] ;

[0128] The above results are compared with the period proximity threshold value of 0.5, , it is determined that the rhythm period of suspended solids and ammonia nitrogen in this paragraph is close, and is included in the period approaching pair combination.

[0129] ​​​​​​​​The cross identification submodule extracts the starting value difference in each paragraph and calculates the absolute difference according to the periodic approach to the combination of indicators, and if the first direction identifier of the two indicators is consistent, it is identified as a cross trend segment, and the corresponding indicator combination and paragraph number are recorded to obtain the cross trend pollution indicator segment;

[0130] The starting original values of the two indicators in the paragraph determined by each pair of combinations are extracted, and the difference value is calculated, for example, for suspended solids and ammonia nitrogen, the values at the start time of the paragraph are 4.2 mg / L and 1.1 mg / L, and the difference is 3.1 mg / L. If the approach threshold is set to 3, the starting point approach condition of the pair of indicators is not met, and it is not output. The starting values of chemical oxygen demand and ammonia nitrogen are 18 mg / L and 1.1 mg / L, and the difference is 16.9 mg / L, which does not meet the condition. The first labels of the direction sequence are compared, and if both directions are "+", it is considered that the directions are consistent. Only when the starting value difference is less than the threshold value and the direction identifier is consistent, it is determined as a cross trend paragraph. The indicator pair and its corresponding paragraph number that meet the condition are recorded and output, and finally the cross trend pollution indicator segment is obtained.

[0131] Please refer to Figure 5 The jump coincidence determination module includes:

[0132] The jump node statistics submodule calls all indicator combinations in the cross trend pollution indicator segment, extracts the trend direction sequence for each group of indicators in the corresponding paragraph, labels the jump nodes for the time sequence, and repeats the operation to extract ammonia nitrogen and chemical oxygen demand. The jump node position of the indicator in the paragraph is counted to obtain the direction jump time point set;

[0133] First, the trend direction sequences of ammonia nitrogen and chemical oxygen demand (COD) within each segment are obtained. The trend direction is determined by comparing the values ​​of the indicators at two adjacent time points. If the later value is greater than the earlier value, the direction is recorded as positive; otherwise, it is recorded as negative. If they are equal, it is considered stable. Taking the segment from 08:00 to 08:05 on July 3, 2024 as an example, the values ​​of the three indicators are as follows: suspended solids (water): [1.2, 1.0, 1.4, 1.6, 1.1] mg / L, ammonia nitrogen: [0.8, 0.9, 1.0, 0.9, 1.1] mg / L, and chemical oxygen demand (COD): [3.5, 3.4, 3.7, 3.2, 3.6] mg / L. The corresponding trend sequence is: permanganate: [–, + [+,+,–], ammonia nitrogen is [+,+,–,+], and chemical oxygen demand is [–,+,–,+]. Based on this, the position points of directional change are extracted, that is, the position points in the sequence that change from positive to negative or from negative to positive, and are marked as jump nodes. For example, the directional change of permanganate occurs at time points 1 and 4, ammonia nitrogen at time point 3, and chemical oxygen demand at time points 1, 2, and 3. The time points corresponding to these jump nodes in each segment are counted to form the set of all jump nodes in that segment. For example, permanganate and chemical oxygen demand jump together at minute 1, and ammonia nitrogen and chemical oxygen demand jump together at minute 3. Finally, all segments and combinations are traversed to obtain the set of directional jump time points under all index combinations.

[0134] The overlap frequency summarization submodule extracts nodes where two or three indicators change direction simultaneously at the same time point from the set of direction change time points. It then counts and sorts the frequency of overlapping change nodes in all paragraphs using the following formula:

[0135] ;

[0136] The calculation obtains the jump overlap metric value, and determines whether the overlap degree exceeds the set overlap frequency threshold to obtain high-frequency overlapping jump nodes;

[0137] in, This represents the node overlap metric. For the first The frequency of node overlap in a segment. This represents the average frequency of node overlap across all paragraphs. This represents the number of overlapping jumps in various indicators within the segment for that node. The number of time periods in which the direction is consistent;

[0138] The time node frequencies with multi-index simultaneous jump behavior are counted section by section, and the node coincidence of each paragraph is measured by difference. The node coincidence measurement formula is used for calculation. Taking paragraph 1 to paragraph 3 as an example, the jump nodes in paragraph 1 are [1, 3], which correspond to 2 and 2 index jumps respectively, and the frequencies are 2 and 2. The jump nodes in paragraph 2 are [2, 4], and the frequencies are 3 and 2. The jump nodes in paragraph 3 are [1, 2], and the frequencies are 2 and 3. The frequency values of each paragraph are set as , , , the corresponding number of coincident indexes is , , , and the average value of each frequency is:

[0139] ;

[0140] Substitute the formula:

[0141] ;

[0142] Wherein, the numerator is the sum of the absolute values of the difference degree and the number of coincident indexes after weighting, and the denominator is the sum of the square root of the total frequency and the square sum of the coincident items. The calculation result is This value is greater than the set threshold value 0.25, so it can be judged that there is a high-frequency coincident jump node in the paragraph group.

[0143] The interference point determination submodule compares the trend consistency and variation amplitude of the corresponding time point in the three indexes according to the high-frequency coincident jump node, judges whether there is a situation of consistent direction and variation amplitude greater than the corresponding index jump reference value at the same time point, and if the judgment result is satisfied at the same time, the time point is classified as an interference affecting point, and a multi-index common jump node set is established;

[0144] The direction mark and variation amplitude of the corresponding multiple indexes at the point are compared synchronously. If the trend direction of the three indexes is consistent at a certain node, and the variation amplitude of each index exceeds the set jump reference value, the node can be judged as an interference affecting point. For example, at the 3rd minute period, permanganate changes from 1.4 to 1.6, the variation amplitude is 0.2 mg / L, ammonia nitrogen changes from 1.0 to 0.9, the amplitude is 0.1 mg / L, and chemical oxygen demand changes from 3.7 to 3.2, the amplitude is 0.5 mg / L. The corresponding index jump reference values are permanganate 0.15 mg / L, ammonia nitrogen 0.08 mg / L, and chemical oxygen demand 0.3 mg / L. The variation amplitudes of the three indexes all exceed the reference values, and the directions all show a downward trend, which meets the judgment standard. The node is classified as an interference affecting point. According to this standard, whether the coincident nodes in all paragraphs meet the conditions is judged, and the nodes meeting the conditions are integrated to establish a multi-index common jump node set.

[0145] Please refer toFigure 6 The index derailment identification module comprises:

[0146] The index trend extraction submodule obtains the raw index sequences of ammonia nitrogen and chemical oxygen demand in each paragraph based on the segmented start and end records in the multi-index common jump node set, sequentially extracts the sequences according to the time sequence in each paragraph, and calculates the adjacent value direction identifier. According to the difference, the positive and negative changes are judged, and the direction sequence is constructed. The length of time of the same direction continuously appearing in the paragraph is counted and the rhythm period sequence is divided, and the three index trend direction sequence and the rhythm sequence set are generated;

[0147] The total phosphorus, ammonia nitrogen, chemical oxygen demand and dissolved oxygen original monitoring values of each three time points before and after each jump node are sequentially extracted. The difference between the adjacent two time points is calculated by difference operation on the original data. The positive and negative of each segment difference value are judged. If the difference is positive, it is recorded as positive change. If it is negative, it is recorded as negative change. If the difference is zero, it is marked as stable state. The direction sequence composed of two direction identifiers is constructed in each index, and it is stored as a list structure in the form of index classification. In the data calling process, the data source must be continuous and not missing, and the influence of abnormal value points on the difference direction judgment is excluded. Data retrieval and processing are performed for each index respectively, and finally a data collection is formed with the index as the key value and the direction sequence in the time period as the value. This collection is used as the basis for participation in the subsequent trend judgment process. Through data standardization processing, the original unit is converted into direction identifier form to avoid numerical interference, and the index direction sequence collection is obtained.

[0148] The period proximity judgment submodule calls the three index trend direction sequence and the rhythm sequence set, respectively judges the rhythm period values of ammonia nitrogen and chemical oxygen demand in each paragraph in pairs, and uses the formula:

[0149] ;

[0150] The difference degree of the rhythm period of any two index combinations in the paragraph is obtained by operation, and the difference degree is compared with the set period proximity threshold value. If it is less than 0.2, it is judged that the period is approaching, and then the period approaching combination group is obtained.

[0151] Wherein, represents the rhythm period difference degree, and represent the period values of the first , , index in the first paragraph, represents the period disturbance harmonic coefficient of the first paragraph,

[0152] Call the recorded index data in the index direction sequence set, judge whether each group of indexes at two adjacent directions formed by continuous three time points is consistent, if there is a conversion from "positive direction" to "negative direction" or "negative direction" to "positive direction", it is judged as direction mismatch, direction continuous consistency or direction is zero is marked as matching state, the number of mismatch in each group of index direction sequence is counted, and the trend mismatch ratio is determined by the following innovative formula:

[0153] ;

[0154] Taking the data at a certain time point as an example, the direction sequence of total phosphorus, ammonia nitrogen, chemical oxygen demand and dissolved oxygen is [-, +], [+, +], [-, -] and [+,-] respectively. The direction change amplitude of each index is: total phosphorus Δ1^ξ=0.05, ammonia nitrogen Δ2^ξ=0.02, chemical oxygen demand Δ3^ξ=0.03, and dissolved oxygen Δ4^ξ=0.04. The consistency determination value θ corresponding to each index is θ1^η=0 (mismatch), θ2^η=1 (match), θ3^η=1 (match), and θ4^η=0 (mismatch). The calculation process is as follows:

[0155] ;

[0156] ;

[0157] ;

[0158] If the threshold value is set to 0.02, the S value is greater than the threshold value, and it is determined that the trend is mismatched. The result is marked in the trend mismatch index value ratio set. This value can be used in the subsequent off-track judgment module for direction inconsistency judgment.

[0159] The cross-fragment recognition sub-module is based on the periodic approach combination group. It is judged whether the start data difference of the corresponding index combination in each section is less than the set approach value, and the first value of the direction sequence is compared for consistency. If both conditions are met, it is marked as a cross-trend index section. All index combinations that meet the conditions are output corresponding to the paragraph index value, and the generated environmental monitoring result is obtained.

[0160] The threshold is compared, combinations in which two or more indicators are simultaneously mismatched at three time points are screened, combinations meeting the conditions are classified into derailment combinations, the system reads the indicator mismatch ratio records at each jump time point in turn, extracts the time, monitoring point number and indicator name of the node meeting the mismatch condition, constructs a structured derailment combination list, adds a combination number to each record for calling, the list is sorted in ascending order of time and keeps consistent with the original data index to avoid node attribution conflict or overlap, the binding process of nodes and combination structure is completed, and finally the environmental monitoring result dataset is established. The result set can be returned to the main processing logic flow as an output port to obtain the environmental monitoring result.

[0161] The above is only a preferred embodiment of the present application, and does not limit the present application in other forms. Any skilled person in the art can modify or change the above disclosed technical content to equivalent embodiments applied to other fields, but any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application without departing from the technical solution content of the present application still belongs to the protection scope of the technical solution of the present application.

Claims

1. An online monitoring system for water environment, characterized in that, The system comprises: The pollution load calculation module obtains ammonia nitrogen, total phosphorus, pH value continuous data, calculates adjacent period difference and marks direction, combines multiple index direction identification, and outputs suspended matter water load convergence section label; The abnormal section division module extracts index change amplitude, identifies direction turning point and segments according to continuous direction according to the suspended matter water load convergence section label, records trend intensity and duration, and forms an abnormal pollution paragraph trend structure table; The rate trajectory screening module extracts the direction and rhythm of ammonia nitrogen and chemical oxygen demand based on the abnormal pollution paragraph trend structure table, screens the cross trend pollution index group segment with close starting point and consistent direction, and generates a multi-index common jump node set based on the index jump node in the cross trend pollution index group segment. The index off-track identification module analyzes the direction of total phosphorus, ammonia nitrogen, chemical oxygen demand and dissolved oxygen at the three time points before and after according to the multi-index common jump node set, determines the off-track combination if two or more indexes are continuously mismatched, and generates an environmental monitoring result. The rate trajectory screening module comprises a trend extraction submodule which obtains the start and end time information in each segment based on the paragraph division in the abnormal pollution paragraph trend structure table, collects the original concentration value sequence of ammonia nitrogen and chemical oxygen demand in the corresponding time period, calculates the change amplitude of adjacent values per hour, marks as positive if the current value minus the previous value is greater than zero, otherwise as negative, divides the time period with the same continuous direction mark as a rhythm period segment, constructs the direction sequence and rhythm period length set of each index in the paragraph, and obtains the index trend period sequence group. The suspended matter water load convergence section label comprises adjacent period index difference value, unified direction identification, change direction consistency frequency, and trend period time range. The abnormal pollution paragraph trend structure table comprises direction turning point position, direction fragment sequence, index fluctuation amplitude, and trend duration. The cross trend pollution index group segment comprises the combination relationship of ammonia nitrogen and chemical oxygen demand, index trend direction, trend period proximity, and starting point position consistency. The multi-index common jump node set comprises index direction change time point, jump node coincidence frequency, and cross-index jump characteristics. The off-track combination comprises the direction mismatching situation of total phosphorus, ammonia nitrogen, chemical oxygen demand and dissolved oxygen, mismatching time period, and continuous mismatching index group.

2. The water environment on-line monitoring system according to claim 1, characterized in that, The pollution load calculation module comprises:

3. The water environment on-line monitoring system according to claim 2, characterized in that, An index difference calculation submodule calculates the numerical difference of each index in adjacent time periods based on the continuous detection records of ammonia nitrogen, total phosphorus and pH value, sorts the index difference calculation submodule according to the time sequence, obtains the direction identification of the index at the difference point, and generates an index direction identification sequence. A trend direction classification submodule calls the index direction identification sequence, extracts the direction identification of multiple indexes at the same time point, classifies the time points with consistent direction identification by comparing the index direction, accumulates the frequency of time points with consistent direction, and generates a consistency direction frequency distribution. ​ The convergence section identification submodule sets a frequency threshold and performs screening according to the consistency direction frequency distribution, extracts a time section in which the consistency direction frequency continuously appears and exceeds the threshold, and obtains a convergence degree coefficient by using a formula: The operation obtains a convergence degree coefficient, judges whether the convergence degree coefficient exceeds a section classification boundary value, and obtains a suspended matter water load convergence section label; wherein, represents the direction of the suspended matter water index in the i th time period, represents the direction of the ammonia nitrogen, represents the direction of the total phosphorus, represents the direction of the pH value, F i is the consistency frequency of the direction in the i th time period, n is the number of consistent directions, Δt is the interval time of each two time points, T c is the degree of convergence coefficient.

4. The water environment on-line monitoring system according to claim 3, characterized in that, The abnormal section division module includes: The operation obtaining submodule extracts a sequence of original value data of the suspended matter water index in the time section, detects a sequence of differences between adjacent time nodes, calculates a change amplitude of the differences, and obtains an index change amplitude value based on the suspended matter water load convergence section label; The direction identification submodule judges a sign change trend according to the index change amplitude value, performs continuity judgment on the direction identification sequence, compares differences between a current direction identification and adjacent time section direction identifications, identifies whether there is a positive-negative value conversion phenomenon as a direction turning point, and obtains a direction turning position point set; The structure construction submodule calls the direction turning position point set, re-divides the section of the direction turning point in units of each direction segment, and calculates a fluctuation interval of the suspended matter water index change and a duration in each direction segment.

5. The water environment on-line monitoring system according to claim 4, characterized in that, The rate trajectory screening module includes: The trend extraction submodule obtains start and end time information in each section based on section division in the abnormal pollution paragraph trend structure table, collects a sequence of original concentration values of ammonia nitrogen and chemical oxygen demand in the corresponding time section, respectively calculates a change amplitude of adjacent values per hour, that is, if a current value minus a previous value is greater than zero, the current value is marked as positive, otherwise, the current value is marked as negative, a time section with a same continuous direction mark is divided as a rhythm cycle section, a direction sequence of each index in the section and a rhythm cycle length set are constructed, and an index trend cycle sequence group is obtained; The cycle comparison submodule calls the index trend cycle sequence group, sequentially extracts rhythm cycle length values of three indexes in each paragraph, performs normalized difference analysis on a cycle value sequence of any two indexes in a multi-index pairing combination mode, and introduces cycle sequence disturbance to smooth and correct the cycle difference value; The cross identification submodule extracts a start value difference and calculates an absolute difference in each section according to the cycle trend of the index combination, if the first direction identifications of the two indexes are consistent, the cross trend section is identified, and the corresponding index combination and paragraph number are recorded, and a cross trend pollution index section is obtained.

6. The water environment on-line monitoring system according to claim 5, characterized in that, The jump coincidence judgment module includes: The jump node statistical submodule calls all index combinations in the cross trend pollution index section, extracts a trend direction sequence of each group of indexes in the corresponding paragraph, labels a jump node of the time sequence, repeatedly performs the extraction on ammonia nitrogen and chemical oxygen demand, and counts jump node positions of the indexes in the paragraph to obtain a direction jump time point set; The coincidence frequency induction submodule extracts nodes in which two or three indexes simultaneously change directions at the same time point according to the direction jump time point set, counts and sorts coincidence jump node frequencies appearing in all paragraphs, and obtains a formula: The operation obtains a jump coincidence degree measurement value, and determines whether the coincidence degree exceeds a set coincidence frequency threshold value to obtain a high-frequency coincidence jump node; wherein R φ represents the node coincidence metric value, Z i μ is the node coincidence frequency of the i-th segment, and is the average node coincidence frequency of all segments, and δ i is the number of index jump coincidences of the node in each segment, and n is the number of segments with consistent directions. The interference point determination submodule compares the trend consistency and variation amplitude of the three indexes at the corresponding time point according to the high-frequency coincidence jump node, determines whether there is a case that the directions are consistent and the variation amplitude is greater than the corresponding index jump reference value at the same time point, and if the determination result is that both conditions are met, the time point is classified as an interference influence point, and a multi-index common jump node set is established.

7. The water environment on-line monitoring system according to claim 6, characterized in that, The index deviation identification module includes: An index trend extraction submodule obtains the ammonia nitrogen and chemical oxygen demand original index sequences in each paragraph based on the segment start and end records in the multi-index common jump node set, extracts the sequences in each paragraph according to the time sequence, calculates the adjacent value direction identifier, judges the positive and negative variation according to the difference, constructs the direction sequence, counts the length of time that the same direction continuously appears in the paragraph, divides the rhythm period sequence, and generates the three index trend direction sequence and rhythm sequence set; A period proximity determination submodule calls the three index trend direction sequence and rhythm sequence set, respectively judges the rhythm period values of ammonia nitrogen and chemical oxygen demand in each paragraph in pairs, and uses the formula: The operation obtains the rhythm period difference of any two index combinations in the paragraph, compares the difference with a set period proximity threshold value, and if the difference is less than 0.2, it is determined that the periods are approaching, and then the period approaching combination group is obtained; where Δ θ represents the rhythm cycle difference degree, C i α and C i β respectively represent the cycle values of the first α, β indexes in the first paragraph, γ i represents the i-th period disturbance harmonic coefficient, and n is the number of direction consistent paragraphs. A cross segment identification submodule judges whether the start point data difference of the corresponding index combination in each paragraph is less than a set proximity value based on the period approaching combination group, and compares the first value of the direction sequence for consistency, and if both conditions are met, it is marked as a cross trend index segment, and all index combinations and paragraph index values that meet the conditions are output to generate an environmental monitoring result.

Images