SCADA-based remote monitoring system for sewage treatment
Patent Information
- Application Number
- CN202611209085.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-11
- Publication Date
- 2026-11-03
AI Technical Summary
The existing SCADA system for wastewater treatment plants cannot adjust the data acquisition frequency of monitoring points in a way that is not tailored to local nodes with controllable value in the process. This results in data redundancy and insufficient accuracy in describing the process status, making it impossible to accurately identify key control nodes.
The data acquisition module acquires real-time sensor data, the topology mapping module maps the data onto a one-dimensional topology link, the node calibration module marks key control nodes, the differential conversion module performs differential and symbolization processing, and the frequency control module dynamically adjusts the acquisition frequency according to the data fluctuation pattern.
It enables accurate identification of key control nodes in wastewater treatment plants and dynamically adjusts the acquisition frequency of the SCADA system according to data fluctuation patterns, reducing data redundancy and improving the accuracy of process status characterization.
Smart Images

Figure CN123026257A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of remote monitoring technology for wastewater treatment, specifically a wastewater treatment remote monitoring system based on SCADA. Background Technology
[0002] Multiple monitoring points are distributed within the wastewater treatment plant. The SCADA system collects real-time sensor data such as dissolved oxygen concentration, mixed liquor suspended solids concentration, oxidation-reduction potential, and influent and effluent flow rates from remote terminal units at a fixed frequency. This fixed-frequency acquisition method generates a large amount of redundant data when the process is stable, increasing the system's storage and transmission burden. Furthermore, during periods of rapid change in process parameters, the fixed sampling interval fails to capture the detailed characteristics of data fluctuations, resulting in insufficient accuracy in characterizing the process state.
[0003] Existing monitoring solutions typically adjust the sampling frequency uniformly across the entire plant, failing to generate differentiated frequency adjustment strategies for local nodes in the process that truly have controllable value. Furthermore, because the monitoring points are spatially connected along the water flow direction, directly describing the state differences between points in three-dimensional spatial coordinates easily overlooks process drift caused by cumulative effects along the process chain, thus failing to accurately identify key locations with nodal impacts on overall operation.
[0004] The problem this application aims to solve is how to quantitatively characterize the cumulative change of state vectors in a one-dimensional process link to locate key control nodes, and how to dynamically adjust the acquisition frequency of the SCADA system based on the jump characteristics of data fluctuation patterns. Summary of the Invention
[0005] This application provides a SCADA-based remote monitoring system for wastewater treatment, which is used to quantitatively characterize the cumulative change of state vectors in a one-dimensional process link direction to locate key control nodes, and dynamically adjust the acquisition frequency of the SCADA system based on the jump characteristics of data fluctuation patterns.
[0006] To achieve the above objectives, the present invention provides the following technical solution: The present invention provides a SCADA-based remote monitoring system for wastewater treatment, the system comprising a data acquisition module, a topology mapping module, a node calibration module, a differential conversion module, and a frequency control module. The data acquisition module acquires real-time sensor data sequences from each monitoring point in the wastewater treatment plant within a continuous sampling period through the SCADA system. These real-time sensor data sequences include multi-dimensional parameters such as dissolved oxygen concentration, mixed liquor suspended solids concentration, oxidation-reduction potential, and influent and effluent flow rates, comprehensively reflecting the state changes of the treatment process. The topology mapping module maps the real-time sensor data sequences onto a one-dimensional topology link constructed based on the main pipeline flow direction, according to the spatial layout of each monitoring point, forming a multi-dimensional state vector sequence arranged along the flow direction. By projecting the three-dimensional spatial position onto the one-dimensional coordinate axis of the flow direction, the relative position of the monitoring point on the process chain is directly correlated with the hydraulic retention characteristics, providing a structured basis for subsequent analysis. The node calibration module uses the Euclidean distance between two adjacent state vectors in the multidimensional state vector sequence as a metric, calculates the cumulative distance point by point along the one-dimensional topology link, and marks the corresponding monitoring point as a key control node when the cumulative distance exceeds a preset threshold. By using the cumulative distance to measure the cumulative effect of the state differences between adjacent points, it can accurately identify link segments that have a significant impact on the downstream process status, thereby automatically selecting the key nodes with the most control value from numerous monitoring points and avoiding the computational overhead caused by indiscriminate processing of all points. The differential conversion module performs differential operations on the real-time sensor data sequence at the key control node according to the acquisition time to obtain a differential sequence, and performs symbolic conversion on the differential sequence to generate a binary string for representing the data fluctuation pattern. By performing differential symbolic processing on the sensor data at the key control node, the continuous numerical form of fluctuation characteristics can be compressed into discrete binary pattern codes, which facilitates rapid comparison of data stability at different time periods. The frequency control module uses the XOR value of two adjacent bits in the binary string as a trigger condition. When the XOR value is logically true, it extracts the real-time sensor data at the corresponding moment and stores it in the buffer. Simultaneously, it dynamically adjusts the data acquisition frequency of the SCADA system based on the time interval of the data in the buffer. By detecting field transitions in the binary string through XOR logic, it can keenly capture abrupt changes in data fluctuation patterns, triggering high-value data extraction only at such moments. Furthermore, it adjusts the acquisition frequency inversely based on the time interval distribution characteristics of the buffered data. This reduces data redundancy during normal operation while ensuring complete recording of abnormal events, and dynamically tilts acquisition resources towards periods of intensive operational changes.
[0007] As a technical solution of the present invention, the data acquisition module acquires analog signals from each monitoring point by the remote terminal unit of the SCADA system according to the initial sampling frequency, and generates the real-time sensing data sequence after analog-to-digital conversion; the monitoring points include inlet monitoring point, aeration tank monitoring point, secondary sedimentation tank monitoring point and outlet monitoring point, and each monitoring point is arranged sequentially from front to back on the one-dimensional topology link according to the flow direction of the main pipeline; the real-time sensing data sequence is subjected to sliding window filtering to remove outliers that exceed the sensor range, ensuring the data quality for subsequent analysis.
[0008] In the above technical solution, the topology mapping module uses the main axis of the main pipeline as the coordinate axis direction of the one-dimensional topology link, and vertically projects the coordinates of each monitoring point in three-dimensional space onto the coordinate axis to obtain the one-dimensional coordinate value of each monitoring point on the one-dimensional topology link; sorts all monitoring points in ascending order of the one-dimensional coordinate values to generate a monitoring point sequence; and arranges the real-time sensing data sequence of each monitoring point according to its position in the monitoring point sequence to form the multi-dimensional state vector sequence. This mapping method preserves the sequential relationship of the process flow, making the data organization along the water flow direction naturally conform to the biochemical reaction process.
[0009] In a preferred embodiment of the present invention, the node calibration module calculates the first Euclidean distance between the i-th state vector and the (i+1)-th state vector in the multidimensional state vector sequence, and uses the first Euclidean distance as the segment length of the i-th link segment. Starting from the first state vector of the one-dimensional topology link, the segment lengths of each link segment along the path are accumulated to obtain the cumulative distance at the current monitoring point. For each calculated cumulative distance at a monitoring point, the cumulative distance is compared with a preset threshold, and when the cumulative distance first exceeds the preset threshold, the corresponding monitoring point is marked as the key control node. This process achieves parameter-free automatic calibration of key control nodes, avoiding the limitations of relying on manual experience to set fixed monitoring sections.
[0010] Furthermore, the differential conversion module reads real-time sensor data values from the buffer, arranged chronologically by acquisition time, at the key control nodes; calculates the difference between real-time sensor data values at adjacent acquisition times to obtain an original differential value sequence; and takes the absolute value of each original differential value in the original differential value sequence to generate the differential sequence. Through absolute value processing, the differential sequence focuses on reflecting the fluctuation amplitude, ignoring the fluctuation direction, and uniformly incorporating rising and falling edges into the measurement of fluctuation intensity. Based on this, the mean of the differential sequence is calculated, and this mean is used as a symbolization threshold; each differential value in the differential sequence is judged one by one, and when the differential value is greater than or equal to the symbolization threshold, it is converted to a binary symbol 1; when the differential value is less than the symbolization threshold, it is converted to a binary symbol 0; all the converted binary symbols are arranged sequentially according to the order of their corresponding differential values in the differential sequence to generate the binary string. Using its own mean as the binarization threshold makes the symbolization result adaptive to the statistical characteristics of the current data, enhancing the robustness of the fluctuation pattern representation.
[0011] In the frequency control module described above, starting from the first bit of the binary string, the j-th bit and the (j+1)-th bit are sequentially XORed to obtain the j-th XOR value. When the j-th XOR value is 1, it is determined that there is a data fluctuation mode transition between the acquisition time corresponding to the j-th bit and the acquisition time corresponding to the (j+1)-th bit in the binary string. In the case of the data fluctuation mode transition, a transition trigger signal is generated, which is used to indicate the extraction of real-time sensor data at the corresponding time. The edge detection mechanism based on XOR is triggered only when the fluctuation mode changes, effectively filtering out redundant records during steady-state fluctuations. When the j-th XOR value is logically true, the acquisition time corresponding to the j-th bit and the acquisition time corresponding to the (j+1)-th bit are determined. The real-time sensor data stored in the SCADA system that matches the acquisition time corresponding to the j-th bit is used as the first extraction data, and the real-time sensor data that matches the acquisition time corresponding to the (j+1)-th bit is used as the second extraction data. The first extraction data and the second extraction data are combined in the order of acquisition time and written into the buffer. In this way, the buffer only retains key sensor data before and after the mode transition, providing a direct basis for frequency regulation.
[0012] Preferably, the frequency control module reads the acquisition times corresponding to two adjacent data writes from the buffer, calculates the time interval between adjacent acquisition times, and obtains an actual time interval sequence; calculates the mode of the actual time interval sequence and compares the mode with a reference time interval; determines the frequency adjustment direction based on the difference between the mode and the reference time interval, generates a frequency adjustment command according to the frequency adjustment direction, and sends it to the remote terminal unit of the SCADA system. Using the mode of the time interval as the adjustment basis, the concentrated occurrence rhythm of abnormal events under the current operating conditions can be extracted from historical jump records. If jumps are dense, the acquisition frequency is increased to capture more process details; if jumps are sparse, the acquisition frequency is decreased to save communication and storage resources, enabling the SCADA system's data acquisition strategy to autonomously optimize according to the actual operating status of the wastewater treatment plant.
[0013] The technical effects and advantages provided by the present invention in the above technical solution are as follows:
[0014] Real-time sensor data sequences from each monitoring point are mapped onto a one-dimensional topological link constructed along the main pipeline flow direction, resulting in a multi-dimensional state vector sequence arranged along the flow direction. The cumulative distance is calculated point-by-point along the link, using the Euclidean distance between adjacent state vectors as a metric. The Euclidean distance of the state vectors comprehensively reflects the synchronous changes in multiple parameters such as dissolved oxygen, sludge concentration, potential, and flow rate. Cumulative calculation along the flow direction captures the gradual drift effect of process states during spatial transmission. When the cumulative distance exceeds a preset threshold, the corresponding monitoring point is marked as a critical control node, thereby automatically identifying links in the continuous spatial link where significant cumulative changes in state occur, providing a clear spatial location basis for differentiated frequency regulation.
[0015] At key control nodes, differential operations are performed on the real-time sensor data sequence to obtain a differential sequence, which is then symbolically converted to generate a binary string representing the data fluctuation pattern. The XOR value of adjacent bits in the binary string is used as a trigger condition; when the XOR value is logically true, a data fluctuation pattern transition is determined between corresponding acquisition times, and the real-time sensor data at the transition time is extracted and stored in a buffer. The XOR operation directly captures the boundary of fluctuation sign changes, eliminating the need for fixed thresholds to determine fluctuation strength and reducing dependence on noise and parameter sensitivity. Based on the time interval between adjacent data write times in the buffer, the mode of the actual time interval sequence is calculated and compared with a reference time interval to determine the frequency adjustment direction, thereby generating a frequency adjustment command and sending it to the remote terminal unit of the SCADA system. The frequency adjustment strategy is directly driven by the temporal distribution characteristics of data fluctuation pattern transitions, making the acquisition frequency sparser during periods of stable process conditions and denser during periods of frequent fluctuation state transitions, achieving adaptive sampling that dynamically matches the processing process. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.
[0017] Figure 1 This is a schematic diagram of a wastewater treatment remote monitoring system based on SCADA.
[0018] Figure 2 This is a flowchart of the frequency control module.
[0019] Figure 3 This is a flowchart illustrating how the frequency control module adjusts the acquisition frequency based on the actual time interval sequence.
[0020] Figure 4 This is a schematic diagram of real-time sensor data on dissolved oxygen concentration at the inlet monitoring point and outlier filtering.
[0021] Figure 5 It is a curve showing the cumulative distance of monitoring points at wastewater treatment plants under operating conditions and a calibration diagram of key control nodes;
[0022] Figure 6 It is a frequency distribution diagram of the time interval between adjacent data collection times. Detailed Implementation
[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0024] See Figure 1This invention provides a SCADA-based remote monitoring system for wastewater treatment, comprising a data acquisition module, a topology mapping module, a node calibration module, a differential conversion module, and a frequency control module. The data acquisition module acquires real-time sensor data sequences from each monitoring point in the wastewater treatment plant within a continuous sampling period through the SCADA system. The topology mapping module maps the real-time sensor data sequences onto a one-dimensional topology link constructed based on the main pipeline flow direction, forming a multi-dimensional state vector sequence arranged along the flow direction, according to the spatial layout of each monitoring point. The node calibration module calculates the cumulative distance point by point along the one-dimensional topology link, using the Euclidean distance between two adjacent state vectors in the multi-dimensional state vector sequence as a metric, and marks the corresponding monitoring point as a key control node when the cumulative distance exceeds a preset threshold. The differential conversion module performs differential operations on the real-time sensor data sequences at the key control nodes according to the acquisition time to obtain a differential sequence, and performs symbolic conversion on the differential sequence to generate a binary string representing the data fluctuation pattern. The frequency control module uses the XOR value of two adjacent bits in a binary string as the trigger condition. When the XOR value is logically true, it extracts the real-time sensor data at the corresponding moment and stores it in the buffer. At the same time, it dynamically adjusts the data acquisition frequency of the SCADA system according to the time interval of the data in the buffer.
[0025] Example 1:
[0026] In practical implementation, the data acquisition module obtains real-time sensor data sequences from each monitoring point in the wastewater treatment plant within a continuous sampling period through the SCADA system. The remote terminal unit of the SCADA system collects analog signals from each monitoring point according to an initial sampling frequency. The initial sampling frequency is set based on the typical timescale of parameter changes in the wastewater treatment process, with a set value of once every 5 seconds. This time interval can capture fluctuations in the concentration of suspended solids and dissolved oxygen in the mixed liquor. The remote terminal unit connects to the sensors at each monitoring point. The sensors output analog signals, which are current signals ranging from 4 mA to 20 mA. The remote terminal unit has a built-in analog-to-digital converter with a resolution of 12 bits. 12 bits of resolution corresponds to 4096 quantization levels, linearly mapping the 4 mA to 20 mA current signals to digital values from 0 to 4095, thus converting the analog signal into a digital signal. Within the continuous sampling period, a digital signal is obtained at each sampling moment. The digital signals from each sampling moment are arranged in chronological order to generate a real-time sensor data sequence.
[0027] A sliding window filter is applied to the real-time sensor data sequence to remove outliers exceeding the sensor's measurement range. During the sliding window filtering process, a sliding window length of 7 sampling points is set. The 7 sampling points are chosen to cover data changes within 35 seconds, smoothing short-term disturbances without sacrificing responsiveness to medium-term trends. Starting from the initial sampling time of the real-time sensor data sequence, seven consecutive sampling points are taken to form a filtering window. The median of the seven data values within the filtering window is calculated. Each data value within the filtering window is compared to the median. If the absolute difference between a data value and the median exceeds a preset deviation threshold, and this data value is greater than the upper limit of the corresponding sensor range or less than the lower limit, then this data value is determined to be an outlier exceeding the sensor's measurement range. The preset deviation threshold is set to 5% of the sensor's measurement range, which enables the identification of outliers that significantly exceed normal fluctuations. Data values determined to be outliers are discarded, and the replacement is made up of the data value from the immediately preceding valid sampling time. After completing the processing of the current filtering window, slide the filtering window forward by one sampling time and repeat the above judgment and filling operations until all sampling times of the real-time sensing data sequence have been traversed.
[0028] The monitoring points include an inlet monitoring point, an aeration tank monitoring point, a secondary sedimentation tank monitoring point, and an outlet monitoring point. These four monitoring points are arranged sequentially from front to back along the main pipeline flow direction on a one-dimensional topology link. The inlet monitoring point is located at the very beginning of the one-dimensional topology link, followed by the aeration tank monitoring point, then the secondary sedimentation tank monitoring point, and finally the outlet monitoring point. The real-time sensor data sequence includes dissolved oxygen concentration, mixed liquor suspended solids concentration, oxidation-reduction potential, and influent and effluent flow rates. The real-time sensor data sequence for the inlet monitoring point consists of influent flow rate data. The real-time sensor data sequence for the aeration tank monitoring point consists of dissolved oxygen concentration data, mixed liquor suspended solids concentration data, and oxidation-reduction potential data. The real-time sensor data sequence for the secondary sedimentation tank monitoring point consists of mixed liquor suspended solids concentration data. The real-time sensor data sequence for the outlet monitoring point consists of effluent flow rate data and dissolved oxygen concentration data. The real-time sensor data sequence of each monitoring point is processed by the above analog-to-digital conversion and sliding window filtering to form a clean sensor data sequence for subsequent mapping.
[0029] See Figure 4 In the figure, the horizontal axis represents the sampling time in seconds, ranging from 0 to 2500 seconds, and the vertical axis represents the dissolved oxygen concentration in milligrams per liter (mg / L), ranging from 0 to 7. The legend shows three types of data: the dashed line represents the unfiltered real-time sensor data sequence, the blue solid line represents the real-time sensor data sequence after sliding window filtering, and the red cross indicates outliers that exceed the sensor's measurement range.
[0030] The solid blue line in the figure shows a periodic fluctuation trend, with the dissolved oxygen concentration maintained between approximately 2.5 and 3.5 mg / L, reflecting the normal range of dissolved oxygen concentration variation during wastewater treatment. The curve is smooth and without significant abrupt changes, indicating that the sliding window filtering effectively removes instantaneous noise and abnormal disturbances, ensuring the continuity and stability of the data.
[0031] The outliers marked with red crosses are distributed throughout the entire time period. These outlier values significantly deviate from the normal concentration range, with some exceeding 6 mg / L and others falling below 0.5 mg / L, clearly exceeding the sensor's measurement range. A large discrepancy exists between the corresponding times of these outliers and the blue filtered curves, indicating that these outliers were identified and removed during the sliding window filtering process. The presence of these outliers suggests the existence of sporadic measurement anomalies or sensor malfunction signals in the real-time sensing data sequence.
[0032] Example 2:
[0033] In practice, the topology mapping module acquires the three-dimensional spatial coordinates of each monitoring point. The three-dimensional spatial coordinates are established using the overall layout plan of the wastewater treatment plant as a reference, forming a three-dimensional Cartesian coordinate system. The origin of the coordinate system is selected at the geometric center of the inlet monitoring point. The positive direction of the X-axis is consistent with the flow direction of the main pipeline, the Y-axis is perpendicular to the X-axis and lies in the horizontal plane pointing due north, and the Z-axis is perpendicular to the XOY plane pointing upwards. The coordinates of each monitoring point in three-dimensional space are obtained through on-site measurement using a total station. The coordinates of the inlet monitoring point are (0,0,0), the aeration tank monitoring point are (15.2,2.1,-0.5), the secondary sedimentation tank monitoring point are (28.6,1.8,-0.3), and the outlet monitoring point are (42.3,2.5,0.1). All coordinate units are meters.
[0034] Determine the principal axis direction of the main pipeline. The main pipeline axis is the line connecting the inlet monitoring point to the outlet monitoring point along the pipeline centerline. The principal axis direction is represented by the principal axis direction vector pointing from the inlet monitoring point to the outlet monitoring point. The principal axis direction vector from the inlet monitoring point (0,0,0) to the outlet monitoring point (42.3,2.5,0.1) is (42.3,2.5,0.1). Normalize the principal axis direction vector to obtain the principal axis direction unit vector. The normalization formula is to divide each component of the vector by the magnitude of the vector. The unit vector is located along the principal axis. The three components are labeled as follows: , , .
[0035] The coordinates of each monitoring point in three-dimensional space are vertically projected onto a coordinate axis based on the flow direction of the main pipeline. For any monitoring point, its three-dimensional coordinates are represented as follows: ,in To monitor the coordinates of the points on the X-axis, To monitor the coordinates of the point on the Y-axis, This refers to the coordinates of the monitoring point on the Z-axis. It also refers to the one-dimensional coordinates of the monitoring point on the one-dimensional topology link. By monitoring the three-dimensional coordinate vector of the point location and the unit vector of the principal axis direction The dot product is calculated using the following formula:
[0036]
[0037] in, , , These represent the coordinate components of the monitoring point on the X-axis, Y-axis, and Z-axis in a three-dimensional Cartesian coordinate system, all in meters; Represents the unit vector along the principal axis. Components on the X-axis Represents the unit vector along the principal axis. Components on the Y-axis, Represents the unit vector along the principal axis. The components on the Z-axis are all dimensionless values; This represents the one-dimensional coordinates of the monitoring point after vertical projection, in meters.
[0038] In the specific calculation, the coordinates of the inlet monitoring point (0,0,0) are substituted into the formula to obtain the one-dimensional coordinate value 0. The coordinates of the aeration tank monitoring point (15.2,2.1,-0.5) are substituted into the formula, and a dot product is taken with the unit vector component along the principal axis to obtain the one-dimensional coordinate value of the aeration tank monitoring point. The coordinates of the secondary sedimentation tank monitoring point (28.6,1.8,-0.3) are substituted into the formula, and a dot product is taken with the unit vector component along the principal axis to obtain the one-dimensional coordinate value of the secondary sedimentation tank monitoring point. The coordinates of the outlet monitoring point (42.3,2.5,0.1) are substituted into the formula, and a dot product is taken with the unit vector component along the principal axis to obtain the one-dimensional coordinate value of the outlet monitoring point.
[0039] All monitoring points are sorted in ascending order of their calculated one-dimensional coordinate values to generate a monitoring point sequence. The monitoring point with the smallest one-dimensional coordinate value is placed first in the sequence, and the monitoring point with the largest one-dimensional coordinate value is placed last. Since the main pipeline flows from the inlet monitoring point to the outlet monitoring point, the sorted monitoring point sequence follows the same flow direction as the main pipeline, and is as follows: inlet monitoring point, aeration tank monitoring point, secondary sedimentation tank monitoring point, and outlet monitoring point.
[0040] In practice, the real-time sensor data sequences of each monitoring point are arranged according to their positions in the monitoring point sequence to form a multi-dimensional state vector sequence. For any given sampling time, the real-time sensor data values of each monitoring point at that time are acquired. The inlet monitoring point provides influent flow rate data, the aeration tank monitoring point provides dissolved oxygen concentration data, mixed liquor suspended solids concentration data, and oxidation-reduction potential data, the secondary sedimentation tank monitoring point provides mixed liquor suspended solids concentration data, and the outlet monitoring point provides effluent flow rate data and dissolved oxygen concentration data. Following the order of the inlet monitoring point, aeration tank monitoring point, secondary sedimentation tank monitoring point, and outlet monitoring point in the monitoring point sequence, the above real-time sensor data values are sequentially concatenated into a one-dimensional multi-dimensional state vector. The parameters corresponding to each dimension of the vector maintain a fixed correspondence with their positions in the monitoring point sequence. This concatenation operation is performed for each sampling time to generate the corresponding multi-dimensional state vector. All multi-dimensional state vectors from all sampling times are arranged in chronological order of sampling time to form a multi-dimensional state vector sequence arranged along the water flow direction. Each multidimensional state vector in this multidimensional state vector sequence contains real-time sensor data from all monitoring points along the main pipeline from upstream to downstream, directly reflecting the spatial distribution characteristics of the entire process state.
[0041] Example 3:
[0042] In practical implementation, the node calibration module acquires a multidimensional state vector sequence. This sequence consists of multiple multidimensional state vectors arranged along the main pipeline flow direction within the one-dimensional topology link. The number of multidimensional state vectors equals the number of monitoring points, and the order of the multidimensional state vectors is the same as the order of the monitoring point sequence. The node calibration module groups adjacent state vectors in the multidimensional state vector sequence and processes each group of adjacent state vectors sequentially, starting from the beginning of the one-dimensional topology link.
[0043] Calculate the first Euclidean distance between the a-th state vector and the (a+1)-th state vector in a multidimensional state vector sequence. The a-th state vector is represented as... The (a+1)th state vector is represented as ,in This represents the total number of sensor parameter dimensions contained in the multidimensional state vector. The value is equal to the sum of the real-time sensor data types of all monitoring points. This represents the component value of the a-th state vector in the k-th sensing parameter dimension. This represents the component value of the (a+1)th state vector in the kth sensing parameter dimension. First Euclidean distance. The calculation formula is:
[0044]
[0045] in, This represents the state difference metric for the a-th link segment; The range of values for is natural numbers starting from 1. The maximum value is the total number of multidimensional state vectors in the multidimensional state vector sequence minus 1; This is the index number for the sensing parameter dimension. The value range is from 1 to All integer values; Let be the specific value of the k-th sensing parameter dimension in the a-th state vector. This represents the specific value of the k-th sensing parameter dimension in the (a+1)-th state vector. Each value is taken from the real-time sensing data sequence of the corresponding monitoring point after analog-to-digital conversion and sliding window filtering.
[0046] In practical implementation, the calculated state Euclidean distance is used as the state difference metric for the *a*-th link segment. This metric characterizes the magnitude of the state difference in sensor data between two adjacent monitoring points. The segment length accumulation operation is performed starting from the first state vector of the one-dimensional topology link. The first state vector corresponds to the monitoring point at the beginning of the one-dimensional topology link, i.e., the inlet monitoring point, and the initial value of the distance accumulation is set to 0. The segment length of the first link segment is added to the initial value of the distance accumulation to obtain the distance accumulation after passing through the first link segment. This distance accumulation corresponds to the location of the second monitoring point, i.e., the aeration tank monitoring point. The segment length of the second link segment is added to the distance accumulation at the aeration tank monitoring point location to obtain the distance accumulation after passing through the second link segment. This distance accumulation corresponds to the location of the third monitoring point, i.e., the secondary sedimentation tank monitoring point. The segment length of the third link segment is added to the distance accumulation at the secondary sedimentation tank monitoring point location to obtain the distance accumulation after passing through the third link segment. This distance accumulation corresponds to the location of the fourth monitoring point, i.e., the outlet monitoring point. Each time the road segment length is accumulated, the cumulative distance at the current monitoring point is obtained.
[0047] For each monitoring point, the accumulated distance is calculated and compared with a preset threshold. The preset threshold is a pre-defined value, set based on the upper limit of the allowable difference in sensor data state between adjacent monitoring points under normal operating conditions of the wastewater treatment plant. The preset threshold is 3.2, determined by calculating the 95th percentile of the daily average Euclidean distance of the multidimensional state vector from the aeration tank to the secondary sedimentation tank using three months of historical normal operating data from the wastewater treatment plant. The calculated value was 3.17, rounded up to one decimal place. The accumulated distance is compared with the preset threshold of 3.2, starting from the first monitoring point. If the accumulated distance at the current monitoring point is less than or equal to the preset threshold of 3.2, the current monitoring point is not marked, and the calculation and comparison of the accumulated distance at the next monitoring point continues. If the accumulated distance exceeds the preset threshold of 3.2 for the first time, the corresponding monitoring point is marked as a critical control node, and the calculation and comparison of subsequent accumulated distances are stopped. After the key control nodes are marked, the node calibration module outputs the monitoring point identification information and the corresponding cumulative distance value of the key control nodes.
[0048] In practice, the differential conversion module reads real-time sensor data values from the key control nodes, arranged chronologically by acquisition time, from the buffer. The data in the buffer originates from the real-time sensor data sequences corresponding to the key control nodes stored in the SCADA system. Each data value in the real-time sensor data sequence is timestamped, recording the acquisition time. The differential conversion module reads the real-time sensor data values from two adjacent acquisition times sequentially, according to the acquisition time sequence indicated by the timestamps. It subtracts the real-time sensor data value from the previous acquisition time from the value at the later acquisition time, calculating the difference between the adjacent acquisition times. All differences calculated from adjacent acquisition times are arranged chronologically to obtain the original differential value sequence. The absolute value of each original differential value in the original differential value sequence is taken; if the original differential value is negative, the sign is reversed to positive; if the original differential value is positive, it remains unchanged. All differential values after taking the absolute value are arranged chronologically according to the original acquisition time to generate the differential sequence.
[0049] In practice, the difference sequence is symbolized to generate a binary string representing the data fluctuation pattern. The mean of the difference sequence is calculated by summing all difference values and dividing by the total number of difference values. This mean is used as the symbolization threshold. Each difference value in the difference sequence is compared to the symbolization threshold. If the difference value is greater than or equal to the threshold, it is converted to a binary symbol 1. If the difference value is less than the threshold, it is converted to a binary symbol 0. The converted binary symbols are then arranged sequentially according to their order in the difference sequence: the first difference value's binary symbol is in the first position, the second difference value's binary symbol is in the second position, and so on until the last difference value's binary symbol is in the last position. Each bit in the binary string corresponds to a difference value in the difference sequence, and the correspondence is determined by the order of the difference values.
[0050] See Figure 5 The horizontal axis of the graph represents the monitoring points of the wastewater treatment plant, in the order of inlet monitoring point, aeration tank monitoring point, secondary sedimentation tank monitoring point, and outlet monitoring point, consistent with the flow direction of the main pipeline. The vertical axis represents the cumulative operating distance, expressed as the cumulative value of Euclidean distance. The blue curve shows the trend of the cumulative operating distance between adjacent state vectors in a multidimensional state vector sequence corresponding to multiple different sampling times. The curve starts from the inlet monitoring point and increases sequentially along the main pipeline, reflecting the cumulative change of differences in real-time sensor data states along the water flow direction. The red dashed line in the graph represents the preset threshold of 3.2, which is the upper limit of the allowable difference in sensor data states between adjacent monitoring points under normal operating conditions of the wastewater treatment plant. The red pentagram in the graph marks the key control nodes, corresponding to the positions where the cumulative distance first exceeds the preset threshold of 3.2.
[0051] As shown in the figure, the cumulative distance under most operating conditions shows a continuous upward trend with the sequence of monitoring points, reflecting the gradual evolution of the sensor data status from the inlet to the outlet. Some curves show jumps in cumulative distance exceeding the preset threshold of 3.2 at the secondary sedimentation tank and outlet monitoring points, corresponding to the marking of key control nodes. This indicates that the sensor status at these locations differs significantly, reaching or exceeding the allowable range under normal operating conditions, requiring close monitoring and control. The distribution of key control nodes is concentrated in the midstream to downstream section of the main pipeline, consistent with typical characteristics of changes in the wastewater treatment process status.
[0052] Example 4:
[0053] In specific implementation, please refer to Figure 2The frequency control module acquires a binary string. The binary string is a symbolic sequence output by the differential conversion module, used to characterize the data fluctuation pattern. Each bit in the binary string corresponds to the symbolic conversion result of the difference value between two adjacent acquisition times in the real-time sensing data sequence at the key control node. The bit order in the binary string strictly corresponds to the chronological order of the acquisition times.
[0054] The frequency control module starts from the first bit of the binary string and sequentially performs an XOR operation between the p-th bit and the (p+1)-th bit. This represents the binary value of the p-th bit in the binary string. The value can be either 0 or 1. This represents the binary value of the (p+1)th bit in the binary string. The value can be 0 or 1. The p-th XOR value. The calculation formula is:
[0055]
[0056] in, This represents the XOR logical operator. The rule for XOR logical operation is: when... and When the values are the same Equal to 0, when and When the values are different It equals 1. This is the position index value of the binary string. The value range is from 1 to all integer values from the total number of bits in the binary string minus 1. This represents the binary value of the p-th bit in the binary string. This represents the binary value of the (p+1)th bit in the binary string. This represents the p-th XOR value, which can be either 0 or 1.
[0057] The frequency control module performs the XOR operation on the binary values of each adjacent bit sequence, starting from 1 and incrementing sequentially, to obtain a series of XOR values. When the p-th XOR value... When the value equals 1, the frequency control module determines that there is a data fluctuation pattern transition between the acquisition time corresponding to the p-th bit and the acquisition time corresponding to the (p+1)-th bit in the binary string. A data fluctuation pattern transition refers to a change in the amplitude of real-time sensor data variation between two adjacent acquisition times, either from below the symbol threshold to above the symbol threshold, or from above the symbol threshold to below the symbol threshold. Both transitions are manifested as a flip of the binary sign of each difference value in the difference sequence. In the presence of a data fluctuation pattern transition, the frequency control module generates a transition trigger signal. The transition trigger signal is a level pulse signal with a pulse width of one system clock cycle. The transition trigger signal is used to indicate the extraction of real-time sensor data at the corresponding time.
[0058] When the p-th XOR value When the value is 0, the frequency control module determines that there is no data fluctuation mode transition between the acquisition time corresponding to the p-th bit and the acquisition time corresponding to the p+1-th bit in the binary string, and does not generate a transition trigger signal. The frequency control module continues to perform XOR logic operation and judgment operation on the binary value of the next group of adjacent bits.
[0059] In practical implementation, when the p-th XOR value represents logical true, the frequency control module determines the two acquisition times for which real-time sensor data needs to be extracted. The acquisition time corresponding to the p-th bit is determined as follows: the p-th bit in the binary string is obtained by symbolic conversion of the p-th difference value in the difference sequence. The p-th difference value is obtained by taking the absolute value of the difference between the real-time sensor data value at the p-th acquisition time and the real-time sensor data value at the (p+1)-th acquisition time in the real-time sensor data sequence. The p-th bit corresponds to the p-th difference value in the difference sequence, and thus corresponds to the previous acquisition time in the pair of acquisition times, p-th acquisition time and (p+1)-th acquisition time. The acquisition time corresponding to the (p+1)th bit is determined as follows: the (p+1)th bit is obtained by symbolic conversion of the (p+1)th difference value in the difference sequence. The (p+1)th difference value is obtained by taking the absolute value of the difference between the real-time sensor data value at the (p+1)th acquisition time and the real-time sensor data value at the (p+2)th acquisition time in the real-time sensor data sequence. The (p+1)th bit corresponds to the (p+1)th difference value in the difference sequence, and thus corresponds to the preceding acquisition time in the pair of acquisition times, namely, the (p+1)th acquisition time and the (p+2)th acquisition time. When the p-th XOR value represents logical true, the frequency control module determines that the extraction target is the acquisition time corresponding to the p-th bit and the acquisition time corresponding to the (p+1)th bit. These two acquisition times are the adjacent acquisition times that generate the data fluctuation mode jump.
[0060] The frequency control module uses the real-time sensor data stored in the SCADA system that matches the acquisition time corresponding to the p-th bit as the first extracted data. The historical database in the SCADA system stores real-time sensor data using the acquisition time as the index key. The frequency control module performs an exact match query in the historical database using the acquisition time corresponding to the p-th bit as the retrieval key, and reads the returned real-time sensor data value, which is the first extracted data. The frequency control module also uses the real-time sensor data stored in the SCADA system that matches the acquisition time corresponding to the (p+1)-th bit as the second extracted data. It performs the same exact match query operation in the historical database using the acquisition time corresponding to the (p+1)-th bit as the retrieval key, and reads the returned real-time sensor data value, which is the second extracted data.
[0061] The frequency control module combines the first and second extracted data in chronological order of acquisition time and writes them into the buffer. The acquisition time corresponding to the first extracted data is earlier than that corresponding to the second extracted data; therefore, the first extracted data is listed first, and the second extracted data is listed later. During combination, the acquisition time and timestamp of the first extracted data are concatenated with its numerical value to form one data record, and the acquisition time and timestamp of the second extracted data are concatenated with its numerical value to form another data record. These two data records are then stored in the buffer in chronological order of acquisition time. The buffer is a fixed-length, first-in-first-out (FIFO) data storage area pre-allocated in system memory. The write pointer of the buffer automatically increments after each write operation, pointing to the next available storage unit.
[0062] Example 5:
[0063] In specific implementation, please refer to Figure 3 The frequency control module reads the acquisition times corresponding to two consecutive data writes from the buffer. The data records stored in the buffer contain the values of real-time sensor data and corresponding timestamps, with the timestamps recording the acquisition time of the real-time sensor data. The frequency control module maintains a buffer read pointer, initially pointing to the earliest written data record in the buffer. The frequency control module reads the timestamp of the data record pointed to by the buffer read pointer to obtain the acquisition time corresponding to the first data write, denoted as the first acquisition time. The frequency control module increments the buffer read pointer by one storage unit, reads the timestamp of the data record pointed to by the incremented pointer, and obtains the acquisition time corresponding to the second data write, denoted as the second acquisition time. The order of the first and second acquisition times is determined by the time order in which the data is written to the buffer; the first acquisition time is earlier than the second acquisition time.
[0064] The frequency control module calculates the time interval between the first and second acquisition moments. The time interval is calculated by subtracting the second count corresponding to the first acquisition moment from the second acquisition moment's second count; the resulting time difference is the time interval between one adjacent acquisition moment. The frequency control module iterates through all adjacent data record pairs in the buffer, performing the above read and calculation operations on each pair of data records with an adjacent write order, obtaining multiple time interval values. All calculated time interval values are arranged in the extraction order to form the actual time interval sequence. Each element in the actual time interval sequence corresponds to the acquisition moment difference of a pair of adjacent written data records in the buffer.
[0065] In practical implementation, the frequency control module calculates the mode of the actual time interval sequence. The process of calculating the mode involves establishing a correspondence table between time interval values and their frequencies. Each time interval value in the actual time interval sequence is traversed. For each traversed time interval value, the correspondence table is checked to see if a record for that time interval value already exists. If the record for that time interval value does not exist in the table, it is added as a new key, and its frequency is initialized to 1. If the record for that time interval value already exists in the table, its frequency is incremented by 1. After the traversal is complete, the value with the highest frequency is found in the table; this value is the mode of the actual time interval sequence. If multiple time interval values have the same highest frequency in the table, the one with the lowest frequency is selected as the calculated mode.
[0066] The frequency control module acquires the reference time interval. The reference time interval is a preset time value, which is set to 5 seconds. The basis for setting the reference time interval to 5 seconds is that the initial sampling frequency of the remote terminal unit of the SCADA system corresponds to a sampling period of once every 5 seconds. During the system initialization phase, the remote terminal unit acquires analog signals at a fixed frequency of once every 5 seconds. Therefore, the reference time interval reflects the default data acquisition time interval under stable operating conditions.
[0067] The frequency control module compares the calculated mode with the reference time interval. Both the mode and the reference time interval are time values in seconds. The frequency control module calculates the difference between the mode and the reference time interval using the following formula:
[0068]
[0069] in, The mode of the actual time interval sequence, in seconds; This represents the reference time interval, in seconds. The value is 5 seconds; This represents the difference between the mode and the reference time interval, in seconds.
[0070] The frequency control module determines the frequency adjustment direction based on the difference between the mode and the reference time interval. When the difference... A value greater than 0 indicates that the actual acquisition interval is mostly greater than the baseline time interval, suggesting a low frequency of data fluctuation pattern jumps. In this case, the frequency adjustment direction is determined to be to reduce the acquisition frequency. When the difference... When the value is less than 0, it indicates that the actual acquisition interval is mostly shorter than the baseline time interval, suggesting a high frequency of data fluctuation pattern jumps. In this case, the frequency adjustment direction is determined to be increasing the acquisition frequency. When the difference... When the value is 0, it means that the actual acquisition interval is consistent with the reference time interval in most cases. In this case, the frequency adjustment direction is determined to keep the current acquisition frequency unchanged.
[0071] The frequency control module generates frequency adjustment commands according to the frequency adjustment direction. The structure of the frequency adjustment command includes a command header, a frequency adjustment direction field, and a frequency adjustment step size field. The command header identifies the command type as a sampling frequency adjustment command; it is a fixed-length binary code with a value of 0x1A. The frequency adjustment direction field has different values depending on the determined frequency adjustment direction: decreasing the sampling frequency corresponds to a binary value of 00, increasing the sampling frequency corresponds to a binary value of 01, and maintaining the current sampling frequency unchanged corresponds to a binary value of 11. The frequency adjustment step size field is set to 1 second, meaning that each frequency adjustment increases or decreases by 1 second based on the current sampling period of the remote terminal unit.
[0072] The frequency control module sends the generated frequency adjustment commands to the remote terminal unit via the SCADA system's communication network. The SCADA system's communication network is a master-slave RS-485 bus network, using the Modbus RTU protocol. The frequency control module acts as the master device, and the remote terminal unit acts as the slave device, with the slave address being a pre-configured address number. The frequency control module encapsulates the frequency adjustment commands into a Modbus RTU protocol frame format, which includes the slave address, function code, register address, data, and CRC checksum. The function code is set to 0x06, corresponding to a single register write operation. The register address is set to the address of the register storing the sampling period in the remote terminal unit, with register address number 40002. The data field contains the combined value of the frequency adjustment direction field and the frequency adjustment step size field. The CRC checksum is calculated based on all bytes from the slave address to the data field. After the Modbus RTU protocol frame is encapsulated, it is sent to the remote terminal unit via the RS-485 bus. After receiving the frequency adjustment command, the remote terminal unit parses the command content and modifies the value of the sampling period register according to the value of the frequency adjustment direction field. After the response is completed, the next sampling period begins to perform analog signal acquisition operation at the adjusted sampling frequency.
[0073] See Figure 6 In the graph, the horizontal axis represents the time interval between adjacent data acquisition moments, in seconds, and the vertical axis represents the frequency of occurrence of the corresponding time interval. The blue bars illustrate the frequency distribution characteristics of the actual time interval sequence, reflecting the statistics of multiple time intervals calculated by the frequency control module after reading data from the buffer corresponding to two adjacent write moments. The red dashed line represents the baseline time interval. The second is a reference value set for the initial sampling period of the SCADA system's remote terminal unit. The orange solid line represents the mode of the actual time interval sequence. The mode is the time interval with the highest frequency of occurrence, which is calculated by establishing a table that corresponds to the time interval and the number of occurrences.
[0074] The bar chart shows that time intervals are mainly concentrated between 2 and 10 seconds, with the most frequent interval being 3 to 6 seconds, and the highest frequency occurring around 4 seconds. This is significantly lower than the baseline interval of 5 seconds, indicating that the time intervals between adjacent data collection moments are mostly concentrated within a short time span. The frequency of time intervals exceeding 10 seconds has decreased significantly, suggesting that extended collection intervals are less common. The mode of 4.2 seconds is less than the baseline interval of 5 seconds, indicating a higher frequency of data fluctuation pattern jumps. The frequency control module, by calculating the mode index of the actual time interval sequence, can effectively reflect the temporal characteristics of real-time sensor data collection at key control nodes.
[0075] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.
Claims
1. A wastewater treatment remote monitoring system based on SCADA, characterized in that, The system includes: The data acquisition module obtains real-time sensor data sequences from each monitoring point in the wastewater treatment plant within a continuous sampling period through the SCADA system. The topology mapping module maps the real-time sensing data sequence onto a one-dimensional topology link constructed based on the main pipeline flow direction, according to the spatial layout relationship of each monitoring point, forming a multi-dimensional state vector sequence arranged along the water flow direction; The node calibration module uses the Euclidean distance between two adjacent state vectors in the multidimensional state vector sequence as a metric to calculate the cumulative distance point by point along the one-dimensional topology link, and marks the corresponding monitoring point as a key control node when the cumulative distance exceeds a preset threshold. The differential conversion module performs differential operations on the real-time sensing data sequence at the key control node according to the acquisition time to obtain a differential sequence, and performs symbolic conversion on the differential sequence to generate a binary string used to characterize the data fluctuation pattern. The frequency control module uses the XOR value of two adjacent bits in the binary string as a trigger condition. When the XOR value is logically true, it extracts the real-time sensing data at the corresponding time and stores it in the buffer. At the same time, it dynamically adjusts the data acquisition frequency of the SCADA system according to the time interval of the data in the buffer.
2. The SCADA-based remote monitoring system for wastewater treatment according to claim 1, characterized in that, The SCADA system acquires real-time sensor data sequences from various monitoring points at the wastewater treatment plant within a continuous sampling period, specifically including: The remote terminal unit of the SCADA system collects analog signals from each monitoring point according to the initial sampling frequency, and generates the real-time sensing data sequence after analog-to-digital conversion; The monitoring points include an inlet monitoring point, an aeration tank monitoring point, a secondary sedimentation tank monitoring point, and an outlet monitoring point. The monitoring points are arranged sequentially from front to back along the main pipeline flow direction on the one-dimensional topology link. The real-time sensing data sequence is subjected to sliding window filtering to remove outliers that are outside the sensor's measurement range.
3. The SCADA-based remote monitoring system for wastewater treatment according to claim 1, characterized in that, The real-time sensing data sequence includes dissolved oxygen concentration, mixed liquid suspended solids concentration, redox potential, and influent and effluent flow rates.
4. The SCADA-based remote monitoring system for wastewater treatment according to claim 1, characterized in that, Based on the spatial layout of each monitoring point, the real-time sensing data sequence is mapped onto a one-dimensional topology link constructed with the main pipeline flow direction as the reference, specifically including: Using the main axis direction of the main pipeline as the coordinate axis direction of the one-dimensional topology link, the coordinates of each monitoring point in three-dimensional space are vertically projected onto the coordinate axis to obtain the one-dimensional coordinate values of each monitoring point on the one-dimensional topology link. All monitoring points are sorted in ascending order of their one-dimensional coordinate values to generate a monitoring point sequence. The real-time sensor data sequences of each monitoring point are arranged according to their positions in the monitoring point sequence to form the multidimensional state vector sequence.
5. The SCADA-based remote monitoring system for wastewater treatment according to claim 1, characterized in that, Using the Euclidean distance between two adjacent state vectors in the multidimensional state vector sequence as a metric, the cumulative distance is calculated point-by-point along the one-dimensional topology link, specifically including: Calculate the first Euclidean distance between the i-th state vector and the (i+1)-th state vector in the multidimensional state vector sequence, and use the first Euclidean distance as the segment length of the i-th link segment; Starting from the first state vector of the one-dimensional topology link, the lengths of each link segment along the path are accumulated to obtain the cumulative distance at the current monitoring point. For each monitoring point where the cumulative distance is calculated, the cumulative distance is compared with the preset threshold, and when the cumulative distance exceeds the preset threshold for the first time, the corresponding monitoring point is marked as the key control node.
6. The SCADA-based remote monitoring system for wastewater treatment according to claim 5, characterized in that, The real-time sensor data sequence at the key control node is differentially processed according to the acquisition time to obtain a differential sequence, specifically including: Read the real-time sensor data values at the key control nodes, arranged in chronological order of acquisition time, from the buffer area; Calculate the difference between real-time sensor data values at adjacent acquisition times to obtain the original difference value sequence; The absolute value of each original difference value in the original difference value sequence is taken to generate the difference sequence.
7. The SCADA-based remote monitoring system for wastewater treatment according to claim 6, characterized in that, The differential sequence is symbolically transformed to generate a binary string representing the data fluctuation pattern, specifically including: Calculate the mean of the difference sequence, and use the mean as a symbolization threshold; Each difference value in the difference sequence is judged one by one. When the difference value is greater than or equal to the symbolization threshold, it is converted into binary symbol 1. When the difference value is less than the symbolization threshold, it is converted into binary symbol 0. All the converted binary symbols are arranged sequentially according to their corresponding difference values in the difference sequence to generate the binary string.
8. The SCADA-based remote monitoring system for wastewater treatment according to claim 1, characterized in that, The trigger condition is the XOR value of two adjacent bits in the binary string, specifically including: Starting from the first bit of the binary string, the j-th bit is sequentially XORed with the (j+1)-th bit to obtain the j-th XOR value. When the j-th XOR value is 1, it is determined that there is a data fluctuation pattern jump between the acquisition time corresponding to the j-th bit and the acquisition time corresponding to the (j+1)-th bit in the binary string. In the event of a change in the data fluctuation pattern, a change trigger signal is generated, which is used to indicate the extraction of real-time sensor data at the corresponding moment.
9. The SCADA-based remote monitoring system for wastewater treatment according to claim 8, characterized in that, When the XOR value is logically true, the real-time sensing data at the corresponding moment is extracted and stored in the buffer, specifically including: When the j-th XOR value is logically true, the acquisition time corresponding to the j-th bit and the acquisition time corresponding to the (j+1)-th bit are determined. The real-time sensor data stored in the SCADA system that matches the acquisition time corresponding to the j-th bit is used as the first extracted data, and the real-time sensor data that matches the acquisition time corresponding to the (j+1)-th bit is used as the second extracted data. The first extracted data and the second extracted data are combined in chronological order of collection time and then written into the buffer area.
10. The SCADA-based remote monitoring system for wastewater treatment according to claim 9, characterized in that, The data acquisition frequency of the SCADA system is dynamically adjusted based on the time interval of the data in the buffer, specifically including: Read the acquisition time corresponding to two consecutive data writes from the buffer, calculate the time interval between adjacent acquisition times, and obtain the actual time interval sequence; Calculate the mode of the actual time interval sequence and compare the mode with the reference time interval; The frequency adjustment direction is determined based on the difference between the mode and the reference time interval, and a frequency adjustment command is generated according to the frequency adjustment direction and sent to the remote terminal unit of the SCADA system.
