Sewage transportation management method and electronic device

Abstract

The present application relates to sewage monitoring technical field, specifically disclose a kind of sewage transport management method and electronic equipment, including S1: data acquisition module carries out multi-source transport data acquisition, data synchronization and validity processing and steady-state transport benchmark determination operation;S2: working condition judgment module carries out transport working condition feature determination, attitude liquid level response calculation, inertia liquid level response calculation and vertical impact liquid level response calculation, and each kind of response is synthesized into normal liquid level change prediction value;S3: leakage judgment module calculates unexplained liquid level residual error, determines residual error allowable boundary, carries out disturbance recovery test and continuous anomaly accumulation;S4: transport comparison module carries out real-time water quality change judgment;S5: display alarm module determines abnormal type according to the received result, and corresponding alarm information or artificial review information is sent to management personnel, solve the technical problem of liquid level abnormality misjudgment caused by lack of differentiating judgment logic for different driving conditions in prior art.

Description

Technical Field

[0001] This invention relates to the field of wastewater monitoring technology, and specifically discloses a wastewater transportation management method and electronic equipment. Background Technology

[0002] In the current field of industrial wastewater transportation management, it is common practice to install level sensors in the storage tanks of transport vehicles and combine them with GPS or BeiDou positioning modules to remotely monitor changes in liquid level and vehicle trajectory during transportation in real time. Managers can view liquid level data and vehicle location through cloud platforms or monitoring centers to determine if there are any abnormalities in the transportation process. Some existing systems also record simple liquid level data during loading and unloading to verify the accuracy of the transport volume. These basic monitoring methods have improved the visibility of wastewater transportation to a certain extent and provided preliminary data support for post-event traceability.

[0003] However, the existing monitoring methods still have certain shortcomings in practical applications. First, transport vehicles frequently experience different operating conditions such as ramps, acceleration, and deceleration during operation, causing the liquid level in the storage tank to sway and shift violently, resulting in drastic fluctuations in the liquid level data read by the sensors. The existing system lacks differentiated judgment logic for different driving postures, which can easily misjudge normal liquid level swaying as leakage, thus generating a large number of invalid alarms and seriously interfering with the normal work of management personnel and emergency response efficiency. Second, the existing liquid level registration and comparison methods are relatively crude, and there is no correlation between the initial liquid level at loading and the liquid level at the end of loading. It is difficult to effectively prevent behaviors such as unauthorized discharge, substitution, or accidental leakage during transportation, and it is impossible to form a closed-loop control of the entire process from loading, transportation to unloading. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a sewage transportation management method and electronic device to solve the technical problem of misjudgment of abnormal liquid level caused by the lack of differentiated judgment logic for different driving conditions in the prior art.

[0005] To achieve the above objectives, one of the objectives of this invention is to provide a wastewater transportation management method, including S1: a data acquisition module performs multi-source transportation data acquisition, data synchronization and validity processing, and steady-state transportation benchmark determination operations;

[0006] S2: The operating condition judgment module determines the current driving condition based on the vehicle's pitch angle, roll angle, and longitudinal acceleration. It calculates the attitude level response based on the vehicle's pitch angle and roll angle at the current and previous sampling times, as well as the actual liquid level at the previous time. It calculates the inertial level response based on the vehicle's longitudinal and lateral acceleration within the preset time delay range at the current sampling time. It calculates the vertical impact intensity of the road surface based on the vehicle's vertical acceleration and calculates the vertical impact level response based on the impact that has not yet been fully recovered. The attitude level response, inertial level response, and vertical impact level response are combined into a predicted value for normal liquid level change.

[0007] S3: The leakage detection module compares the actual liquid level change with the predicted value of the normal liquid level change to obtain the unexplained liquid level residual. It determines the residual tolerance boundary based on the degree of unrecovered vertical impact liquid level response, the degree of prediction uncertainty, and the degree of liquid level noise. It calculates the working condition disturbance intensity and the degree of consistency of the residual direction based on the vehicle attitude change and acceleration change. Within the anomaly detection window, it performs time decay accumulation on residuals that exceed the residual tolerance boundary and point to the direction of liquid level decline. When the accumulated amount exceeds the threshold and the working condition disturbance intensity is continuously lower than the threshold, it generates a suspected abnormal liquid loss identification result.

[0008] S4: The transportation comparison module determines the water quality parameters at the time of liquid loading as the initial value, calculates the normalized deviation of the water quality parameters during transportation and takes the maximum value as the water quality deviation index. When the limit is continuously exceeded, a real-time water quality change anomaly result is generated. After arrival, the end liquid level and end water quality parameters are compared with the initial value. When the liquid level difference exceeds the preset range, a total abnormal result is generated. When the water quality parameter difference exceeds the preset range, an arrival water quality change anomaly result is generated. The total abnormal result is cross-checked with the suspected abnormal liquid loss identification result. When both are abnormal, a transportation anomaly alarm result is generated. When only one is abnormal, a manual review result is generated.

[0009] S5: The alarm display module determines the type of abnormality based on the received results and sends the corresponding alarm information or manual review information to the management personnel.

[0010] The second objective of this invention is to provide an electronic device for sewage transportation management, including a data acquisition module, a working condition judgment module, a leakage judgment module, a transportation comparison module, and a display and alarm module;

[0011] The data acquisition module is used to collect multi-source real-time data from the transport vehicle, generate an aligned data sequence with the sampling time of the liquid level sensor as the unified time axis, and establish steady-state transport reference parameters after the liquid loading is completed.

[0012] The operating condition judgment module is used to identify the driving conditions based on the vehicle attitude and acceleration information provided by the data acquisition module, and to calculate a dynamic alarm threshold for distinguishing between normal liquid level sloshing and leakage behavior based on the driving conditions.

[0013] The leakage detection module is used to compare the real-time liquid level with the dynamic alarm threshold to determine whether a leakage alarm is triggered, and to generate a suspected abnormal liquid loss identification result based on the continuous abnormal accumulation of unexplained liquid level residuals within a preset window.

[0014] The transport comparison module is used to compare the final liquid level and water quality parameters after arrival with the initial liquid level and water quality parameters when the liquid filling is completed, and to cross-verify the liquid level comparison result with the suspected abnormal liquid loss identification result, so as to output the transport abnormality alarm result or the manual verification result.

[0015] The display alarm module is used to show monitoring data to managers and, based on the judgment results of the leakage judgment module or the transportation comparison module, push alarm information containing the type of abnormality and the time of occurrence when an abnormality occurs.

[0016] The working principle and beneficial effects of this solution are as follows:

[0017] 1. Calculate the attitude level response, inertial level response, and road vertical impact level response according to the transportation conditions, and obtain the predicted value of normal level change through response coupling correction, so as to distinguish the explainable level change caused by vehicle movement from the abnormal level drop, and reduce the possibility of false alarms of abnormal level triggered by vehicle movement.

[0018] 2. Based on the degree of prediction uncertainty, the degree of liquid level noise and the degree of impact unrecovery, the residual tolerance boundary is dynamically determined, and combined with the disturbance recovery test, the residuals that continuously point to the direction of liquid level decline are accumulated to reduce the possibility of misidentification of suspected abnormal liquid loss under complex transportation conditions.

[0019] 3. The water quality deviation index is used to uniformly characterize the changes in pH, conductivity, turbidity, chemical oxygen demand and ammonia nitrogen content relative to the initial water quality parameters. The abnormal water quality changes during transportation are identified by continuous exceedance judgment, reducing the situation where a single instantaneous fluctuation directly triggers a water quality abnormality alarm.

[0020] 4. Cross-check the abnormal total liquid level results after transportation with the identification results of suspected abnormal liquid loss during transportation, and combine the abnormal real-time water quality changes and abnormal arrival water quality changes to distinguish between continuous abnormal liquid level, abnormal total arrival volume and abnormal water quality changes, so as to provide a basis for graded alarm, manual review and transportation traceability. Attached Figure Description

[0021] Figure 1 A flowchart of a wastewater transportation management method;

[0022] Figure 2 This is an architecture diagram of an electronic device for wastewater transportation management. Detailed Implementation

[0023] The following detailed description illustrates the specific implementation method:

[0024] like Figures 1 to 2 As shown, an electronic device for sewage transportation management is disclosed, including a data acquisition module, a working condition judgment module, a leakage judgment module, a transportation comparison module, and a display and alarm module;

[0025] The data acquisition module is used to collect various real-time data of the transport vehicle during its operation. The operating condition judgment module is used to identify the current operating condition based on the vehicle's posture and motion information and calculate the dynamic alarm threshold. The leakage judgment module is used to compare the real-time liquid level with the dynamic threshold to determine whether a leakage alarm has been triggered. The transport comparison module is used to compare the liquid level and water quality data during the loading and unloading processes to verify the integrity of the transport. The display alarm module is used to display the monitoring data to the management personnel and push alarm information when an anomaly occurs.

[0026] The data acquisition module includes a multi-source transportation data acquisition submodule, a data synchronization and validity processing submodule, and a steady-state transportation benchmark determination submodule;

[0027] The multi-source transportation data acquisition submodule includes a liquid level acquisition unit, an attitude acquisition unit, an acceleration acquisition unit, a positioning and speed measurement unit, a vehicle interface reading unit, and a water quality detection unit. The liquid level acquisition unit collects liquid level data at a fixed sampling frequency of once per second using a liquid level sensor installed inside the storage tank. The attitude acquisition unit detects the vehicle's pitch and roll angles using a tilt sensor. The acceleration acquisition unit detects the vehicle's instantaneous acceleration using an acceleration sensor. The positioning and speed measurement unit obtains real-time location and speed information using a GPS positioning component. The vehicle interface reading unit obtains the vehicle's driving status through an OBD interface, specifically including vehicle speed, braking status, throttle opening, and engine speed. The water quality detection unit uses pH, conductivity, turbidity, COD, and ammonia nitrogen sensors installed inside the storage tank to detect the pH value, conductivity, turbidity, chemical oxygen demand, and ammonia nitrogen content of the wastewater in real time, records the initial water quality data during filling, and continuously monitors water quality changes during transportation.

[0028] The data synchronization and validity processing submodule uses the sampling time of the liquid level sensor as a unified time axis to align the vehicle attitude and vehicle three-axis acceleration in time, and forms a transport data sequence with equal time intervals according to the preset sampling interval. The water quality parameters retain their original valid sampling time and are associated with the vehicle status record that is closest in time. No linear interpolation is performed on the water quality parameters.

[0029] When the OBD vehicle speed or GPS vehicle speed is successfully read at the current sampling time, is within the vehicle speed range of the corresponding data source, and does not trigger a speed change judgment, it is determined to be valid vehicle speed data. When no vehicle speed data is read, it is determined to be missing vehicle speed data. When vehicle speed data is read but does not meet the aforementioned conditions, it is determined to be invalid vehicle speed data. When the OBD vehicle speed is valid, the OBD vehicle speed is determined to be the adopted vehicle speed. When the OBD vehicle speed is missing or invalid and the GPS vehicle speed is valid, the GPS vehicle speed is determined to be the adopted vehicle speed. When both are valid but the difference exceeds the vehicle speed consistency threshold, the adopted vehicle speed is marked as low confidence. When the low confidence state reaches a preset number consecutively, the update of the time-delay response parameters dependent on the vehicle speed interval and the anomaly judgment are suspended. The vehicle speed change threshold and the vehicle speed consistency threshold are respectively taken as the preset upper quantile value of the absolute value of the corresponding vehicle speed difference in historical normal transportation data. The preset number is determined according to the minimum low confidence duration and the sampling interval.

[0030] For liquid level, attitude, acceleration, or vehicle speed data that exceed the corresponding sensor range, or whose difference from the preceding and following sampled values ​​simultaneously exceeds the preset abrupt change range, they are marked as invalid sampled values ​​and replaced with the median of adjacent valid sampled values. A single missing sampled value is obtained by linear interpolation of adjacent valid sampled values. When the number of consecutive invalid sampled values ​​exceeds the preset number, the anomaly determination is paused.

[0031] The steady-state transportation benchmark determination submodule is used to select a continuous sampling period after the vehicle is loaded with liquid, where the changes in vehicle pitch angle, vehicle roll angle, and triaxial acceleration are all lower than the corresponding steady-state thresholds. The average values ​​of the actual liquid level, vehicle pitch angle, and vehicle roll angle within the continuous sampling period are determined as the benchmark liquid level, benchmark pitch angle, and benchmark roll angle, respectively. The median of the absolute value of the actual liquid level change is determined as the base liquid level noise. Each steady-state threshold is taken as the preset upper quantile value of the corresponding change in historical normal transportation data.

[0032] The operating condition judgment module includes a transportation operating condition characteristic determination submodule, an attitude level response calculation submodule, an inertial level response calculation submodule, a vertical impact level response calculation submodule, and a normal level change fusion submodule.

[0033] The transportation condition feature determination submodule is used to determine the slope driving condition when the vehicle pitch angle or roll angle deviates from the corresponding reference angle by more than the slope recognition threshold, the acceleration driving condition when the vehicle longitudinal acceleration exceeds the acceleration recognition threshold, the deceleration driving condition when the vehicle longitudinal acceleration is lower than the deceleration recognition threshold, and the flat road steady-state driving condition when the changes in vehicle pitch angle, roll angle, and three-axis acceleration are all lower than the corresponding steady-state thresholds. When two or more types of conditions exist simultaneously, a composite condition label is formed. Based on the determined condition label, the corresponding attitude response parameters, time delay response parameters, impact recognition threshold, and basic allowable boundary are selected. Each recognition threshold is taken as the preset quantile value of the corresponding feature quantity in the historical normal transportation data.

[0034] The attitude level response calculation submodule is used to calculate the attitude level response based on the vehicle pitch angle, vehicle roll angle and actual liquid level at the previous sampling time when the vehicle passes over a slope or the vehicle body tilts.

[0035] The inertial liquid level response calculation submodule is used to calculate the inertial liquid level response when the vehicle accelerates, decelerates, or turns, as the liquid level response has a time lag relative to the vehicle's acceleration change. It reads the vehicle's longitudinal acceleration and lateral acceleration within the preset time lag range at the current sampling time and before, and calculates the inertial liquid level response.

[0036] The vertical impact level response calculation submodule calculates the road vertical impact intensity based on the deviation of the vehicle's vertical acceleration from the local stable component and the absolute value of the second-order difference of the vehicle's vertical acceleration. When the road vertical impact intensity exceeds the impact identification threshold, the corresponding sampling time is determined as the time when the road vertical impact occurs. After the road vertical impact occurs, the liquid surface forms a gradually decaying oscillation. The vertical impact level response is calculated based on the road vertical impact that has not yet completed recovery.

[0037] The normal liquid level change fusion submodule addresses the simultaneous existence of attitude liquid level response, inertial liquid level response, and vertical impact liquid level response when a vehicle accelerates, decelerates, or is subjected to vertical impact from the road surface on a slope. It calculates the response coupling correction based on the pairwise product of the three types of liquid level responses to correct the interactive effects caused by the simultaneous existence of multiple types of liquid level responses.

[0038] The leakage detection module includes a submodule for calculating unexplained liquid level residuals, a submodule for determining residual tolerance boundaries, a submodule for disturbance recovery verification, and a submodule for continuous anomaly accumulation.

[0039] The unexplained liquid level residual calculation submodule compares the actual liquid level change with the predicted normal liquid level change to obtain the liquid level residual where the actual liquid level change exceeds the predicted normal liquid level change.

[0040] The larger the absolute value of the vertical impact liquid level response, the less fully the liquid surface oscillation caused by the vertical impact on the road surface has recovered. The residual allowable boundary determination submodule calculates the degree of unrecovered vertical impact liquid level response based on the current vertical impact liquid level response and the reference vertical impact liquid level response in historical normal transportation data.

[0041] The disturbance recovery verification submodule calculates the disturbance intensity under operating conditions based on changes in vehicle attitude and three-axis acceleration.

[0042] The continuous anomaly accumulation submodule is used to accumulate unexplained liquid level residuals that exceed the residual tolerance boundary and point in the actual liquid level drop direction within the anomaly judgment window. The accumulation result is adjusted according to the consistency of the residual direction. When the liquid level anomaly accumulation exceeds the anomaly judgment threshold and the operating condition disturbance intensity is continuously lower than the disturbance identification threshold for a preset recovery time, a suspected abnormal liquid loss identification result is generated. The identification result indicates that the leakage judgment module has found a continuous liquid level drop that cannot be fully explained by vehicle movement. It is not directly used as a confirmation result of leakage. The identification result is simultaneously transmitted to the transportation comparison module and the display alarm module.

[0043] The length of the anomaly judgment window is the product of the preset minimum identifiable continuous liquid reduction duration and the sampling frequency. The time decay coefficient is determined according to the preset ratio of the weight corresponding to the starting point of the anomaly judgment window to the weight corresponding to the current sampling time. The preset reduction value corresponding to the interpolation sampling is determined according to the ratio of the interpolation sampling error to the effective sampling error in the historical normal transportation data. The anomaly judgment threshold is determined according to the preset upper quantile value of the corresponding liquid level anomaly accumulation in the historical normal transportation data.

[0044] The attitude response parameters, time delay response parameters, second-order change adjustment coefficient, vertical impact level response parameters, and coupling correction coefficient are calibrated with the goal of minimizing the sum of the squared differences between the actual level change and the predicted normal level change in historical normal transportation data. The coupling correction coefficient adopts a zero-center range constraint. The boundary adjustment coefficient is selected based on historical normal transportation data to ensure that the proportion of unexplained level residuals exceeding the residual tolerance boundary is not higher than the preset false alarm proportion. The disturbance adjustment coefficient is selected based on the marked disturbance period and steady-state period to maximize the difference in the operating condition disturbance intensity between the two types of periods. The calibration data does not include transportation records where leakage, discharge, or sensor failure has been confirmed.

[0045] The transportation comparison module includes a real-time water quality change judgment submodule and an arrival comparison and cross-verification submodule. The real-time water quality change judgment submodule determines the five water quality parameters collected when the liquid filling is completed as the initial water quality parameters. The five water quality parameters are pH value, conductivity, turbidity, chemical oxygen demand and ammonia nitrogen content. For each set of effective water quality parameters obtained during transportation, the normalized deviation degree relative to the corresponding initial water quality parameters is calculated, and the maximum value of the five normalized deviation degrees is taken as the water quality deviation index.

[0046] Upon arrival at the comparison and cross-verification submodule, after the liquid filling is completed, the initial liquid level, initial water quality parameters, liquid filling time, and vehicle number are recorded, and a QR code containing the data is generated. After the transport arrives at the target area, the arrival of the vehicle is confirmed by the GPS positioning component, and the ending liquid level and ending water quality parameters are read. When the difference between the ending liquid level and the initial liquid level exceeds the preset liquid level allowable range, a total abnormal result is generated. When the difference between any ending water quality parameter and the corresponding initial water quality parameter exceeds the corresponding preset allowable change amount, an arrival water quality change abnormal result is generated.

[0047] The total abnormality result is cross-checked with the suspected abnormal liquid loss identification result output by the leakage judgment module. If both indicate an abnormality, a transportation abnormality alarm result is generated. If only one indicates an abnormality, a manual verification result is generated. When a real-time water quality change abnormality result or a water quality change abnormality result is generated, a water quality abnormality alarm is triggered. If none of the above results are generated, a result indicating that no abnormality corresponding to the set detection conditions is found is output. Cross-checking is used to distinguish between continuous liquid level abnormality during transportation, total abnormality after transportation, and water quality change abnormality.

[0048] The alarm display module presents monitoring data and alarm information to management personnel. It displays the vehicle's real-time location and driving trajectory in the form of an electronic map, and shows the liquid level change trend and water quality parameter change trend in the entire transportation process in the form of a curve graph. The interface also marks the current driving condition of the vehicle and historical alarm records. When the transportation comparison module or the leakage judgment module determines that a transportation abnormality has occurred, the alarm display module will pop up an abnormality prompt window on the monitoring interface and send alarm information to management personnel via SMS or APP push. The alarm information includes the abnormality type, the time of the abnormality, the current location of the vehicle, and the corresponding liquid level and water quality data at the time of the abnormality.

[0049] This embodiment also discloses a wastewater transportation management method, including the following steps:

[0050] S1: The data acquisition module performs multi-source transportation data acquisition, data synchronization and validity processing, and steady-state transportation baseline determination operations;

[0051] S11: The multi-source transportation data acquisition submodule is used to collect multi-dimensional data such as liquid level, attitude, acceleration, position, vehicle speed, engine status and water quality parameters through various sensors and interfaces.

[0052] S12: The data synchronization and validity processing submodule uses the sampling time of the liquid level sensor as a unified time axis, aligns the data in time, and forms an equally spaced sequence according to a preset interval. The water quality parameters retain the original sampling time without interpolation. According to the validity judgment result of the aforementioned vehicle speed data, the valid OBD vehicle speed is preferentially determined as the adopted vehicle speed. If the OBD vehicle speed is missing or invalid, the valid GPS vehicle speed is determined as the adopted vehicle speed. If both are valid but the difference exceeds the vehicle speed consistency threshold, the adopted vehicle speed is marked as low confidence. If both are invalid or the low confidence state reaches a preset number consecutively, the abnormal judgment based on the adopted vehicle speed is suspended. For liquid level, attitude, and acceleration data, if the corresponding sensor range is exceeded or a data mutation judgment is triggered, it is marked as an invalid sampling value.

[0053] The actual liquid level change is calculated based on the actual liquid level at adjacent sampling times, as shown in the following calculation:

[0054] ,

[0055] in, Indicates time The actual change in liquid level. Indicates time The actual liquid level, This indicates the actual liquid level at the previous sampling time. This indicates that at the current sampling time, a negative value for the actual liquid level change indicates a decrease in the actual liquid level.

[0056] S13: Steady-state transportation benchmark determination submodule. After the liquid loading is completed, select a stable period when the vehicle pitch angle, roll angle and triaxial acceleration are all below the corresponding threshold. The average value of liquid level, pitch angle and roll angle during this period is determined as the benchmark value, and the median of the absolute value of liquid level change is determined as the basic liquid level noise.

[0057] S2: The working condition judgment module performs the following operations: determining the characteristics of the transportation working condition, calculating the attitude level response, calculating the inertial level response, calculating the vertical impact level response, and fusing the normal level change.

[0058] S21: Transportation Condition Characteristic Determination Submodule. When the pitch angle or roll angle deviates from the reference angle by more than the ramp threshold, it is determined to be a ramp condition. When the longitudinal acceleration exceeds the acceleration threshold or falls below the deceleration threshold, it is determined to be an acceleration / deceleration condition. When all attitude parameters are below the steady-state threshold, it is determined to be a level road steady-state condition. When multiple conditions coexist, they are combined into a composite condition label. Based on the condition label, the corresponding attitude level response parameters, time delay response parameters, vertical impact response parameters, and residual tolerance boundary parameters are selected. Each threshold is determined based on the distribution range of the corresponding feature in historical normal transportation data.

[0059] S22: Attitude and Liquid Level Response Calculation Submodule. Based on the vehicle pitch and roll angles at the current and previous sampling times, combined with the actual liquid level at the previous time, the specific calculation is as follows:

[0060] ,

[0061] in, Indicates time Attitude and liquid level response. and Representing time respectively The vehicle pitch angle at the previous sampling time, and Representing time respectively Compared to the vehicle roll angle at the previous sampling time, This indicates the actual liquid level at the previous sampling time. Indicates the base liquid level. This represents the normalized reference for the liquid level response, determined by a preset upper quantile value of the absolute value of the actual liquid level change in historical normal transportation data. It is used to convert the actual liquid level deviation into a dimensionless liquid level deviation ratio. Indicates and Zero-positive numbers with the same dimensions This represents the dimensionless deviation of the actual liquid level from the baseline liquid level at the previous sampling time. This represents the basic attitude fluid level response parameters corresponding to vehicle pitch. These represent the basic attitude fluid level response parameters corresponding to vehicle roll. This refers to the level difference adjustment parameter that indicates the vehicle's pitch response when the actual liquid level deviates from the baseline liquid level. These are the level difference adjustment parameters that represent the vehicle's roll response when the actual liquid level deviates from the baseline liquid level. All four parameters are obtained using the actual liquid level change as the calibration target and are used to convert dimensionless changes and dimensionless liquid level deviation ratios into liquid level changes. This represents the tangent function.

[0062] The basic attitude level response parameters and level difference adjustment parameters were obtained through attitude-dominant samples from historical normal transportation data. Attitude-dominant samples are sampling records where the vehicle's longitudinal acceleration, lateral acceleration, and road vertical impact intensity are all below their corresponding thresholds, and the vehicle's pitch angle change or roll angle change exceeds the attitude change threshold. These attitude-dominant samples are divided into multiple attitude calibration categories according to vehicle number, tank structure type, actual level range, and operating condition label. For the x-th type of attitude calibration sample, the parameter set after attitude calibration is determined according to the following formula:

[0063] ,

[0064] ,

[0065] in, Indicates the first The set of attitude parameters to be calibrated corresponding to the attitude calibration samples. Indicates the first The set of parameters corresponding to the pose calibration samples. This indicates the attitude calibration category, determined by the vehicle number, tank structure type, actual liquid level range, and operating condition label. This represents the basic attitude fluid level response parameters corresponding to vehicle pitch. These represent the basic attitude fluid level response parameters corresponding to vehicle roll. This refers to the level difference adjustment parameter that indicates the vehicle's pitch response when the actual liquid level deviates from the baseline liquid level. This refers to the level difference adjustment parameter that indicates the vehicle's roll response when the actual liquid level deviates from the baseline liquid level. Indicates the first A set of pose calibration samples, Represents a set Any historical sampling record in China Represents historical sampling records The corresponding actual change in liquid level. This indicates the set of attitude parameters to be calibrated. The calculated attitude and liquid level response. This represents the set of parameters corresponding to the minimum value of the objective function.

[0066] When calculating the attitude and liquid level response at the current sampling moment, the calculation is based on the current vehicle number, the liquid tank structure type, and the actual liquid level.

[0067] The position range and operating condition label determine the corresponding attitude calibration category. If the number of historical samples for a certain attitude calibration category is lower than the preset minimum number of samples, then adjacent categories will be merged and calibrated in the order of actual liquid level range, operating condition label, and storage tank structure type.

[0068] S23: The inertial liquid level response calculation submodule calculates the inertial liquid level response. The specific calculation is as follows:

[0069] ,

[0070] in, Indicates time The inertial liquid level response, Indicates the acceleration delay index. This indicates the maximum time delay of the effect of acceleration. Indicates time The vehicle's longitudinal acceleration, Indicates time The vehicle's lateral acceleration, and Representing time respectively Corresponding to the The time-delay response parameters of the vehicle's longitudinal acceleration and lateral acceleration are used to convert the vehicle acceleration into the corresponding liquid level change.

[0071] The time-delay response parameters are determined by acceleration-dominant samples from historical normal transportation data. Acceleration-dominant samples are sampling records where the vehicle's longitudinal or lateral acceleration changes, and the vehicle's attitude change and the vertical impact intensity of the road surface are both below the corresponding threshold. The maximum time delay is determined by the upper quantile of the sampling time required for the liquid level response to return to the basic liquid level noise range after the acceleration change ends. When calculating the inertial liquid level response at the current sampling time, the corresponding time-delay response parameters are called according to the current operating condition label, the actual liquid level range, and the vehicle speed range. If the number of samples is insufficient, adjacent vehicle speed ranges are merged first, and then adjacent actual liquid level ranges are merged.

[0072] S24: Vertical Impact Liquid Level Response Calculation Submodule. To ensure that the vertical acceleration deviation and the second-order difference of vertical acceleration can jointly characterize the vertical impact intensity of the road surface, the road surface vertical impact intensity calculation submodule sets a normalized benchmark for the vertical acceleration deviation. and vertical acceleration second-order difference normalized reference The specific calculations are as follows:

[0073] ,

[0074] in, Indicates time The vertical impact strength of the road surface, , and These represent the vertical acceleration of the vehicle at the corresponding sampling time. Indicates the end time The median of the vertical acceleration of the vehicle within the preset local window. The normalization benchmark for vertical acceleration deviation is determined by the preset upper quantile value of the absolute value of the deviation of vertical acceleration from the local median in historical normal transportation data. The reference for the second-order difference normalization of vertical acceleration is determined by the preset upper quantile value of the absolute value of the second-order difference of vertical acceleration in historical normal transportation data. Both are used to convert the corresponding vertical acceleration characteristics into dimensionless impact characteristics. and These represent zero-positive numbers that have the same dimensions as the corresponding normalized benchmark. This represents the dimensionless second-order difference adjustment coefficient. It represents the absolute value.

[0075] The second-order differential adjustment coefficient was determined using vertical acceleration samples from historical normal transportation data. Sampling records of marked road impacts were used as impact samples, while sampling records of unmarked road impacts where both vehicle attitude change and horizontal acceleration change were below the corresponding thresholds were used as non-impact samples. The value of the second-order differential adjustment coefficient was chosen to improve the distinction between impact and non-impact samples and was limited to a non-negative value. When the sample size was insufficient, calibration results from adjacent actual liquid level ranges under the same storage tank structure type were used.

[0076] When the vertical impact intensity of the road surface exceeds the impact identification threshold, the corresponding sampling time is determined as the time of occurrence of the vertical impact. After the vertical impact occurs, the liquid surface forms a gradually decaying oscillation. The vertical impact liquid level response is calculated based on the vertical impact of the road surface before it has fully recovered. The specific calculation is as follows:

[0077] ,

[0078] in, Indicates time Vertical impact liquid level response. Indicates the end time Vertical impact collection on the road surface where the liquid level has not yet been restored. Indicates the vertical impact index of the road surface. Indicates the first Vertical impact strength of the road surface during a secondary vertical impact. Indicates the first The actual liquid level at the time of the secondary vertical impact on the road surface. Indicates the first The moment of occurrence of the secondary vertical impact on the road surface Indicates the base liquid level. This represents the normalized reference for the liquid level response, used to convert the dimensionless impact response amplitude into a liquid level change. Indicates and Zero-positive numbers with the same dimensions and This represents the dimensionless initial response amplitude coefficient. Indicates the first Dimensionless attenuation parameters based on sampling intervals corresponding to the vertical impact on the road surface. Indicates the first Dimensionless angular frequency parameters corresponding to the vertical impact on the road surface, calculated based on the sampling interval. Indicates the first The dimensionless initial phase corresponding to the vertical impact on the road surface. Represents absolute value. This represents the natural exponential function. This represents the cosine function.

[0079] The initial response amplitude coefficient, attenuation coefficient, liquid surface oscillation frequency, and initial phase are determined by fitting the actual liquid level change sequence after a vertical impact on the road surface in historical normal transportation data. The fitting samples are divided according to vehicle number, storage tank structure type, actual liquid level range, and road surface vertical impact intensity range. The impact response fitting window length is taken as the upper quantile of the sampling time required for the liquid level change after a vertical impact on the road surface to return to the basic liquid level noise range. When calculating the vertical impact liquid level response at the current sampling time, the fitting parameters of the corresponding category are called. When the number of samples is insufficient, adjacent road surface vertical impact intensity ranges are merged first.

[0080] S25: Normal Liquid Level Change Fusion Submodule. When a vehicle accelerates, decelerates, or experiences a vertical impact from the road surface on a slope, attitude liquid level response, inertial liquid level response, and vertical impact liquid level response coexist. The response coupling correction is calculated based on the pairwise product of the three types of liquid level responses to correct the interactive effects caused by the simultaneous existence of multiple types of liquid level responses. The specific calculation is shown below:

[0081] ,

[0082] in, Indicates time The response coupling correction amount, Indicates time Attitude and liquid level response. Indicates time The inertial liquid level response, Indicates time Vertical impact liquid level response. , and These represent the dimensionless coupling correction coefficients for attitude level response and inertial level response, attitude level response and vertical impact level response, and inertial level response and vertical impact level response, respectively. This represents the normalized reference for the liquid level response, which is determined by the preset upper quantile value of the absolute value of the actual liquid level change in historical normal transportation data. Indicates and Zero positive numbers with the same dimensions.

[0083] The three types of liquid level responses, along with the response coupling correction, are combined to form a predicted value for normal liquid level changes, as shown below:

[0084] ,

[0085] in, Indicates time Predicted normal liquid level changes Indicates time Attitude and liquid level response. Indicates time The inertial liquid level response, Indicates time Vertical impact liquid level response. The response coupling correction amount represents the amount of liquid level change at any given time. The normal liquid level change prediction value is used to characterize the amount of liquid level change caused by the current vehicle movement under the assumption that no liquid loss has occurred.

[0086] The dimensionless coupling correction coefficients are not obtained directly from the sensors, but are calibrated based on the prediction error when multiple types of liquid level responses coexist in historical normal transportation data. For the χ-th type of fusion calibration sample, the set of coupling correction coefficients is determined according to the following formula:

[0087] ,

[0088] ,

[0089] ,

[0090] in, Indicates the first The set of coupling correction coefficients corresponding to the class fusion calibration samples. Indicates the first The set of calibration-corrected coupling coefficients corresponding to the class fusion calibration samples. Indicates the first Class fusion calibration sample set, Represents a set Any historical sampling record in China , and These represent the dimensionless coupling correction coefficients for attitude level response and inertial level response, attitude level response and vertical impact level response, and inertial level response and vertical impact level response, respectively. Represents historical sampling records The corresponding actual change in liquid level. Represents historical sampling records The corresponding predicted value of normal liquid level change. Used to convert prediction error into dimensionless error. Indicates the first The dimensionless coupling constraint coefficients corresponding to the class fusion calibration samples This represents the index of the dimensionless coupling correction coefficient. Indicates the first The first class fusion calibration sample A dimensionless coupling correction coefficient Indicates the first The upper limit of the dimensionless coupling correction corresponding to the fusion calibration sample is determined based on the allowable proportion of the coupling correction term to the predicted value of normal liquid level change in historical normal transportation data.

[0091] When performing normal liquid level change fusion at the current sampling time, the corresponding fusion calibration category is determined based on the current vehicle number, liquid tank structure type, actual liquid level range, vehicle speed range, and operating condition label. If the number of fusion calibration samples is insufficient, fix them first. , and The initial value is zero, and coupling correction will be enabled only after the number of historical samples reaches the preset minimum number of samples.

[0092] S3: The leakage detection module performs unexplained liquid level residual calculation, residual tolerance boundary determination, disturbance recovery verification, and continuous anomaly accumulation operation;

[0093] S31: The unexplained liquid level residual calculation submodule compares the actual liquid level change with the predicted normal liquid level change to obtain the liquid level residual where the actual liquid level change exceeds the predicted normal liquid level change. The specific calculation is as follows:

[0094] ,

[0095] in, Indicates time Unexplained liquid level residuals Indicates time The actual change in liquid level. Indicates time The predicted value of normal liquid level change. When the unexplained liquid level residual is negative, it means that the actual liquid level drop exceeds the liquid level drop that can be explained by vehicle movement.

[0096] S32: Residual tolerance boundary determination submodule. The residual tolerance boundary is used to determine whether the unexplained liquid level residual has exceeded the range that can be explained by normal transportation disturbance. Since the unexplained liquid level residual is simultaneously affected by the prediction error of normal liquid level change, the local noise of the liquid level sensor, and the unrecovered vertical impact, the residual tolerance boundary determination submodule calculates the uncertainty of the prediction of normal liquid level change, the level noise level, and the liquid level boundary increment of the unrecovered vertical impact respectively, and then synthesizes them into the residual tolerance boundary.

[0097] The residual tolerance boundary determination submodule first calculates the prediction error based on historical normal transportation samples, as shown in the following calculation:

[0098] ,

[0099] in, Represents historical sampling records The corresponding prediction error, Represents historical sampling records The corresponding actual change in liquid level. Represents historical sampling records The corresponding predicted value of normal liquid level change.

[0100] Based on the historical normal transport sample set corresponding to the current calibration category, the uncertainty of predicting normal liquid level changes is calculated as follows:

[0101] ,

[0102] in, Indicates time The degree of uncertainty in predicting normal liquid level changes Indicates the current calibration category The corresponding historical normal transportation sample set, Represents a set Number of historical normal transport samples This indicates the current calibration category, determined by the vehicle number, tank structure type, actual liquid level range, vehicle speed range, and operating condition label. Represents a set Any historical sampling record in China Represents historical sampling records The corresponding prediction error, Represents a set The average value of the prediction error.

[0103] The level noise level is used to characterize the effect of fluctuations in the level sensor readings within the current local noise window on unexplained level residuals.

[0104] The impact of the difference. The level of liquid level noise is calculated based on the actual liquid level change within the current local noise window, as shown in the following calculation:

[0105] ,

[0106] in, Indicates time The level of liquid level noise, Indicates the end time Local noise window, and These represent the sampling times within the local noise window. and These represent the actual liquid level changes at the corresponding sampling times. This indicates taking the median. This represents a dimensionless constant used to convert the absolute deviation of the median to the order of standard deviation.

[0107] Reference vertical impact level response This is only a reference scale for describing historical vertical impact level responses and cannot be multiplied by a dimensionless scale to obtain the level increment in the current residual tolerance boundary. To ensure that the residual tolerance boundary reflects the actual impact before the current vertical impact has recovered, the residual tolerance boundary determination submodule uses the absolute value of the current vertical impact level response as the increment of the unrecovered vertical impact level boundary, as shown below:

[0108] ,

[0109] in, Indicates time The vertical impact did not recover the liquid level boundary increment. Indicates time The vertical impact liquid level response.

[0110] Based on the basic allowable boundary, the uncertainty of normal liquid level change prediction, the level noise level, and the unrecovered liquid level boundary increment due to vertical impact, the residual allowable boundary is calculated as follows:

[0111] ,

[0112] in, Indicates time The residual tolerance boundary, This indicates the basic allowable boundary determined based on historical normal transportation data. Indicates time The degree of uncertainty in predicting normal liquid level changes Indicates time The level of liquid level noise, Indicates time The vertical impact did not recover the liquid level boundary increment. , and This represents a dimensionless adjustment coefficient, used to adjust the impact of the uncertainty in predicting normal liquid level changes, the level of liquid level noise, and the impact of the unrecovered liquid level boundary increment due to vertical impact on the residual allowable boundary.

[0113] When the number of historical samples is insufficient, adjacent categories are merged in the order of vehicle speed range, actual liquid level range, and operating condition label for calculation. When the number of effective samples within a local noise window is insufficient, The baseline liquid level noise output of the submodule is determined using a steady-state transport reference. Dimensionless adjustment coefficient. , and The selection is based on historical normal transport data. The selection principle is that the proportion of unexplained liquid level residuals in historical normal transport data that exceed the residual tolerance boundary is not higher than the preset false alarm ratio. Among the candidate coefficient combinations that meet the false alarm ratio constraint, the combination with the smaller average residual tolerance boundary value is selected. This set of coefficients is used to determine the boundary width and does not share the same calibration target with the prediction coefficients of attitude liquid level response, inertial liquid level response or vertical impact liquid level response.

[0114] S33: Disturbance Recovery Verification Submodule, calculates the disturbance intensity under operating conditions based on vehicle attitude changes and three-axis acceleration changes. The specific calculation is shown below:

[0115] ,

[0116] in, Indicates time The intensity of the disturbance under operating conditions, This represents the difference between the vehicle pitch angle at time t and the vehicle pitch angle at the previous sampling time. This represents the difference between the vehicle roll angle at time t and the vehicle roll angle at the previous sampling time. and Representing time respectively Compared to the vehicle's three-axis acceleration vector at the previous sampling time, This represents the normalized reference for vehicle pitch angle variation. This represents the normalized baseline for vehicle roll angle variation. The three normalized references represent the vehicle acceleration vector change normalization benchmarks. They are determined by the preset upper quantile values ​​of the corresponding changes in historical normal transportation data. , and This represents the non-negative dimensionless disturbance adjustment coefficient; the sum of the three is one. , and These represent positive numbers with the same dimensions as the changes in vehicle pitch angle, vehicle roll angle, and vehicle acceleration vector, respectively. Represents absolute value. The second normal form of a vector.

[0117] Non-negative dimensionless disturbance adjustment coefficient , and The selection criteria are based on the marked disturbance period and steady-state period. The selection principle is to ensure that the operating disturbance intensity during the disturbance period is highly distinguishable from that during the steady-state period, and to keep the three coefficients non-negative and their sum equal to one. This set of coefficients is used to determine whether the transportation disturbance has recovered and is not involved in fitting the predicted value of normal liquid level change.

[0118] When the intensity of the disturbance decreases from above the disturbance identification threshold to below the disturbance identification threshold, the liquid level is determined to have entered the recovery phase. A recovery observation window is formed by selecting consecutive sampling times within the recovery phase. The consistency of the residual direction is calculated based on the cancellation of unexplained liquid level residuals within the recovery observation window, as detailed below:

[0119] ,

[0120] in, Indicates time The degree of consistency of residual direction Indicates the sampling time within the observation window to be restored. Unexplained liquid level residuals This indicates the sampling time within the restored observation window. This indicates that the observation window length has been restored. Represents absolute value. This represents a positive number with the same dimensions as the unexplained liquid level residual.

[0121] When the unexplained liquid level residuals within the recovery observation window alternate between positive and negative and cancel each other out, the consistency of the residual direction approaches zero, indicating that the actual liquid level change is more consistent with the liquid surface sloshing caused by vehicle disturbance. When the unexplained liquid level residuals continue to change in the same direction, the consistency of the residual direction increases, indicating that there is a possibility of continuous liquid reduction in the actual liquid level change. The recovery observation window length is taken as the preset upper quantile value of the sampling time required for the unexplained liquid level residuals to return to the residual tolerance boundary after the intensity of the operating condition disturbance in the historical normal transportation data drops below the disturbance identification threshold. The disturbance identification threshold is taken as the preset upper quantile value of the intensity of the operating condition disturbance in the historical normal transportation data.

[0122] S34: Continuous Anomaly Accumulation Submodule. Within the anomaly judgment window, it performs time-decay accumulation on unexplained liquid level residuals that exceed the residual tolerance boundary and point in the actual direction of liquid level decline, and adjusts the accumulation result based on the consistency of the residual direction; the specific calculation is shown below:

[0123] ,

[0124] in, Indicates time The cumulative amount of abnormal liquid level; Indicates time The degree of consistency of residual direction; Indicates the sampling time within the anomaly detection window; Indicates the length of the exception detection window; Indicates the sampling time The reliability of the corresponding liquid level data is determined by taking one value for valid sampling and a preset reduction value for interpolated sampling. Indicates the time decay coefficient; Indicates the sampling time Distance from the current sampling time The number of sampling intervals; Indicates the sampling time Corresponding unexplained liquid level residuals; Indicates the sampling time The corresponding residual tolerance boundary; This indicates that the larger value among the listed values ​​will be selected. , and All are dimensionless quantities. This represents the unilateral level drop exceeding the residual tolerance boundary, therefore This represents the cumulative drop in liquid level within the anomaly detection window. The zero value represents the amount of liquid level drop, and and They have the same dimensions.

[0125] The anomaly detection window length, time decay coefficient, interpolation reduction value, and anomaly detection threshold are determined based on historical normal transportation data. The anomaly detection window length is the product of the preset minimum identifiable continuous liquid reduction duration and the sampling frequency. The time decay coefficient is determined based on the preset ratio of the weight corresponding to the starting point of the anomaly detection window to the weight corresponding to the current sampling time. The interpolation reduction value is determined based on the ratio of the interpolation sampling error to the effective sampling error in the historical normal transportation data. The anomaly detection threshold is the upper quantile value of the corresponding cumulative abnormal liquid level in the historical normal transportation data.

[0126] When the cumulative abnormal liquid level exceeds the judgment threshold and the intensity of the operating disturbance is continuously lower than the identification threshold for a preset recovery time, the leakage judgment module generates a suspected abnormal liquid loss identification result. This result indicates that a continuous drop in liquid level that cannot be fully explained by vehicle movement has been found and is not considered a confirmation of leakage. At the same time, it is sent to the transportation comparison module and the display alarm module. The abnormal judgment window length, time decay coefficient, interpolation reduction value and abnormal judgment threshold are all determined based on historical normal transportation data or preset parameters. The above suspected abnormal liquid loss identification result is sent to the transportation comparison module and the display alarm module simultaneously. The attitude level response parameters, time delay response parameters, vertical impact response parameters, coupling correction coefficient, boundary adjustment coefficient, and disturbance adjustment coefficient are determined according to the sample screening conditions and calibration rules recorded in the corresponding sub-modules. The calibration data does not include transportation records where leakage, discharge, or sensor failure has been confirmed. The coupling correction coefficient adopts a zero-center range constraint. The boundary adjustment coefficient ensures that the proportion of unexplained level residuals exceeding the allowable boundary is not higher than the preset false alarm proportion. The disturbance adjustment coefficient maximizes the difference in operating condition disturbance intensity between the disturbance period and the steady-state period. The calibration data does not include transportation records where leakage, discharge, or sensor failure has been confirmed.

[0127] S4: The transportation comparison module performs real-time water quality change judgment, arrival comparison and cross-verification operations;

[0128] S41: The real-time water quality change judgment submodule determines the five water quality parameters collected when the liquid filling is completed as the initial water quality parameters. For each set of valid water quality parameters obtained during transportation, it calculates the normalized deviation degree relative to the corresponding initial water quality parameters, and takes the maximum value among the five normalized deviation degrees as the water quality deviation index. The specific calculation is as follows:

[0129] ,

[0130] in, Indicates time Water quality deviation index, This represents a set of water quality parameters consisting of pH value, conductivity, turbidity, chemical oxygen demand, and ammonia nitrogen content. Indicates time The Water quality parameters, This indicates that the liquid collection is complete. Initial water quality parameters, Indicates the first The preset allowable variation of the water quality parameters, Indicates the relationship with the first All water quality parameters have zero-positive values ​​of the same dimension. This indicates that the larger value among the listed values ​​will be selected.

[0131] Water quality deviation index greater than When the corresponding valid water quality sampling time is marked as the water quality exceeding the limit time, when the water quality exceeding the limit state continuously reaches the preset valid sampling number, a real-time water quality change anomaly result is generated; isolated exceeding sampling values ​​that have not reached the preset valid sampling number are only used as records to be reviewed, and the preset allowable change amounts are determined based on historical normal transportation records and preset management requirements.

[0132] S42: Arrival Comparison and Cross-verification Submodule. After liquid loading is completed, the initial liquid level, initial water quality parameters, liquid loading time and vehicle number are recorded, and a QR code containing the data is generated. After transportation to the target area, the arrival of the vehicle is confirmed by the GPS positioning component. The ending liquid level and ending water quality parameters are read. When the difference between the ending liquid level and the initial liquid level exceeds the preset liquid level allowable range, a total abnormal result is generated. When the difference between any ending water quality parameter and the corresponding initial water quality parameter exceeds the corresponding preset allowable change amount, an arrival water quality change abnormal result is generated.

[0133] The total abnormality result is cross-checked with the suspected abnormal liquid loss identification result output by the leakage judgment module. If both indicate an abnormality, a transportation abnormality alarm result is generated. If only one indicates an abnormality, a manual verification result is generated. When a real-time water quality change abnormality result or a water quality change abnormality result is generated, a water quality abnormality alarm is triggered. If none of the above results are generated, a result indicating that no abnormality corresponding to the set detection conditions is found is output. Cross-checking is used to distinguish between continuous liquid level abnormality during transportation, total abnormality after transportation, and water quality change abnormality.

[0134] S5: The display alarm module receives the suspected abnormal liquid loss identification results output by the leakage judgment module, and displays in real time the time of the suspected abnormality, vehicle location, actual liquid level, unexplained liquid level residual, residual tolerance boundary, and cumulative amount of abnormal liquid level. The display alarm module also receives the transportation abnormality alarm results, water quality abnormality alarm results, or manual review results output by the transportation comparison module. When a water quality abnormality alarm occurs, the corresponding water quality parameters, water quality deviation index, and time of abnormality occurrence are displayed simultaneously. The display alarm module determines the abnormality type based on the source of the result and sends the corresponding alarm information or manual review information to the management personnel.

[0135] The above descriptions are merely embodiments of the present invention, and common knowledge regarding specific structures and characteristics in the solutions is not described in detail here. It should be noted that those skilled in the art can make various modifications and improvements without departing from the structure of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the implementation of the present invention or its practicality.

Claims

1. A method for managing wastewater transportation, characterized in that: Includes the following steps S1: The data acquisition module performs multi-source transportation data acquisition, data synchronization and validity processing, and steady-state transportation benchmark determination. S2: The operating condition judgment module determines the current driving condition based on the vehicle's pitch angle, roll angle, and longitudinal acceleration. It calculates the attitude level response based on the vehicle's pitch angle and roll angle at the current and previous sampling times, as well as the actual liquid level at the previous time. It calculates the inertial level response based on the vehicle's longitudinal and lateral acceleration within the preset time delay range at the current sampling time. It calculates the vertical impact intensity of the road surface based on the vehicle's vertical acceleration and calculates the vertical impact level response based on the impact that has not yet been fully recovered. The attitude level response, inertial level response, and vertical impact level response are combined into a predicted value for normal liquid level change. S3: The leakage detection module compares the actual liquid level change with the predicted value of the normal liquid level change to obtain the unexplained liquid level residual. It determines the residual tolerance boundary based on the degree of unrecovered vertical impact liquid level response, the degree of prediction uncertainty, and the degree of liquid level noise. It calculates the working condition disturbance intensity and the degree of consistency of the residual direction based on the vehicle attitude change and acceleration change. Within the anomaly detection window, it performs time decay accumulation on residuals that exceed the residual tolerance boundary and point to the direction of liquid level decline. When the accumulated amount exceeds the threshold and the working condition disturbance intensity is continuously lower than the threshold, it generates a suspected abnormal liquid loss identification result. S4: The transportation comparison module determines the water quality parameters at the time of liquid loading as the initial value, calculates the normalized deviation of the water quality parameters during transportation and takes the maximum value as the water quality deviation index. When the limit is continuously exceeded, a real-time water quality change anomaly result is generated. After arrival, the end liquid level and end water quality parameters are compared with the initial value. When the liquid level difference exceeds the preset range, a total abnormal result is generated. When the water quality parameter difference exceeds the preset range, an arrival water quality change anomaly result is generated. The total abnormal result is cross-checked with the suspected abnormal liquid loss identification result. When both are abnormal, a transportation anomaly alarm result is generated. When only one is abnormal, a manual review result is generated. S5: The alarm display module determines the type of abnormality based on the received results and sends the corresponding alarm information or manual review information to the management personnel.

2. The sewage transportation management method according to claim 1, characterized in that: S1 includes the following sub-steps: S11: Multi-source transportation data acquisition submodule collects liquid level, attitude, acceleration, position, vehicle speed, engine status and water quality parameters; S12: The data synchronization and validity processing submodule uses the sampling time of the liquid level sensor as a unified time axis to perform time alignment and form an equal time interval sequence. The water quality parameters retain the original sampling time. If the OBD vehicle speed or GPS vehicle speed is successfully read at the current sampling time, is within the range and has not triggered the vehicle speed change judgment, it is determined to be valid. The actual liquid level change is calculated based on the actual liquid level at adjacent sampling times, as shown below: ， in, Indicates time The actual change in liquid level. Indicates time The actual liquid level, This indicates the actual liquid level at the previous sampling time. At the current sampling time, a negative value for the actual liquid level change indicates that the actual liquid level has decreased. S13: The steady-state transportation benchmark determination submodule selects a stable period when the vehicle's attitude values ​​are all below the corresponding threshold after the vehicle is loaded with liquid. The average value of the liquid level and attitude angle during this period is determined as the benchmark value, and the median of the absolute value of the liquid level change is determined as the basic liquid level noise.

3. The sewage transportation management method according to claim 2, characterized in that: S2 includes the following sub-steps: S21: The transportation condition characteristic determination submodule determines the condition as a slope when the deviation of the pitch angle or roll angle from the corresponding reference angle exceeds the slope identification threshold, the condition as an acceleration / deceleration condition when the longitudinal acceleration exceeds the acceleration identification threshold or is lower than the deceleration identification threshold, and the condition as a level road steady-state condition when the changes in pitch angle, roll angle, and triaxial acceleration are all lower than the corresponding steady-state thresholds. S22: The attitude and liquid level response calculation submodule calculates the attitude and liquid level response based on the vehicle pitch angle, roll angle, and actual liquid level at the current and previous sampling times. The specific calculation is as follows: ， in, Indicates time Attitude and liquid level response. and Representing time respectively The vehicle pitch angle at the previous sampling time, and Representing time respectively Compared to the vehicle roll angle at the previous sampling time, This indicates the actual liquid level at the previous sampling time. Indicates the base liquid level. This represents the normalized reference for the liquid level response, determined by a preset upper quantile value of the absolute value of the actual liquid level change in historical normal transportation data. It is used to convert the actual liquid level deviation into a dimensionless liquid level deviation ratio. Indicates and Zero-positive numbers with the same dimensions This represents the dimensionless deviation of the actual liquid level from the baseline liquid level at the previous sampling time. This represents the basic attitude fluid level response parameters corresponding to vehicle pitch. These represent the basic attitude fluid level response parameters corresponding to vehicle roll. This refers to the level difference adjustment parameter that indicates the vehicle's pitch response when the actual liquid level deviates from the baseline liquid level. These are the level difference adjustment parameters that represent the vehicle's roll response when the actual liquid level deviates from the baseline liquid level. All four parameters are obtained using the actual liquid level change as the calibration target and are used to convert dimensionless changes and dimensionless liquid level deviation ratios into liquid level changes. Represents the tangent function; S23: The inertial liquid level response calculation submodule reads the vehicle's longitudinal and lateral accelerations within the current sampling time and the preset time delay range before that time, and calculates the inertial liquid level response; the specific calculation is as follows: ， in, Indicates time The inertial liquid level response, Indicates the acceleration delay index. This indicates the maximum time delay of the effect of acceleration. Indicates time The vehicle's longitudinal acceleration, Indicates time The vehicle's lateral acceleration, and Representing time respectively Corresponding to the The time-delay response parameters of the vehicle's longitudinal acceleration and lateral acceleration are used to convert the vehicle acceleration into the corresponding liquid level change.

4. The sewage transportation management method according to claim 3, characterized in that: S2 also includes the following sub-steps: S24: Vertical impact level response calculation sub-module. In order to enable the vertical acceleration deviation and the second-order difference of vertical acceleration to jointly characterize the vertical impact intensity of the road surface, the road surface vertical impact intensity calculation sub-module sets a normalized benchmark for the vertical acceleration deviation. and vertical acceleration second-order difference normalized reference The specific calculations are as follows: ， in, Indicates time The vertical impact strength of the road surface, , and These represent the vertical acceleration of the vehicle at the corresponding sampling time. Indicates the end time The median of the vertical acceleration of the vehicle within the preset local window. The normalization benchmark for vertical acceleration deviation is determined by the preset upper quantile value of the absolute value of the deviation of vertical acceleration from the local median in historical normal transportation data. The reference for the second-order difference normalization of vertical acceleration is determined by the preset upper quantile value of the absolute value of the second-order difference of vertical acceleration in historical normal transportation data. Both are used to convert the corresponding vertical acceleration characteristics into dimensionless impact characteristics. and These represent zero-positive numbers that have the same dimensions as the corresponding normalized benchmark. This represents the dimensionless second-order difference adjustment coefficient. Represents absolute value; When the vertical impact intensity of the road surface exceeds the impact identification threshold, the corresponding sampling time is determined as the time of occurrence of the vertical impact. After the vertical impact occurs, the liquid surface forms a gradually decaying oscillation. The vertical impact liquid level response is calculated based on the vertical impact of the road surface before it has fully recovered. The specific calculation is as follows: ， in, Indicates time Vertical impact liquid level response. Indicates the end time Vertical impact collection on the road surface where the liquid level has not yet been restored. Indicates the vertical impact index of the road surface. Indicates the first Vertical impact strength of the road surface during a secondary vertical impact. Indicates the first The actual liquid level at the time of the secondary vertical impact on the road surface. Indicates the first The moment of occurrence of the secondary vertical impact on the road surface Indicates the base liquid level. This represents the normalized reference for the liquid level response, used to convert the dimensionless impact response amplitude into a liquid level change. Indicates and Zero-positive numbers with the same dimensions and This represents the dimensionless initial response amplitude coefficient. Indicates the first Dimensionless attenuation parameters based on sampling intervals corresponding to the vertical impact on the road surface. Indicates the first Dimensionless angular frequency parameters corresponding to the vertical impact on the road surface, calculated based on the sampling interval. Indicates the first The dimensionless initial phase corresponding to the vertical impact on the road surface. Represents absolute value. This represents the natural exponential function. This represents the cosine function.

5. The sewage transportation management method according to claim 4, characterized in that: S2 also includes the following sub-steps: S25: The normal liquid level change fusion submodule calculates the response coupling correction based on the pairwise product of the three types of liquid level responses, and combines the three types of liquid level responses with the correction to form the predicted value of the normal liquid level change. The specific calculation is as follows: ， in, Indicates time The response coupling correction amount, Indicates time Attitude and liquid level response. Indicates time The inertial liquid level response, Indicates time Vertical impact liquid level response. , and These represent the dimensionless coupling correction coefficients for attitude level response and inertial level response, attitude level response and vertical impact level response, and inertial level response and vertical impact level response, respectively. This represents the normalized reference for the liquid level response, which is determined by the preset upper quantile value of the absolute value of the actual liquid level change in historical normal transportation data. Indicates and Positive numbers with the same dimensions that prevent zeroing; The three types of liquid level responses, along with the response coupling correction, are combined to form a predicted value for normal liquid level changes, as shown below: ， in, Indicates time Predicted normal liquid level changes Indicates time Attitude and liquid level response. Indicates time The inertial liquid level response, Indicates time Vertical impact liquid level response. The response coupling correction amount represents the amount of liquid level change at any given time. The normal liquid level change prediction value is used to characterize the amount of liquid level change caused by the current vehicle movement under the assumption that no liquid loss has occurred.

6. The sewage transportation management method according to claim 5, characterized in that: S3 includes the following sub-steps: S31: The unexplained liquid level residual calculation submodule compares the actual liquid level change with the predicted normal liquid level change to obtain the liquid level residual where the actual liquid level change exceeds the predicted normal liquid level change. The specific calculation is as follows: ， in, Indicates time Unexplained liquid level residuals Indicates time The actual change in liquid level. Indicates time The predicted value of normal liquid level change, when the unexplained liquid level residual is negative, indicates that the actual liquid level drop exceeds the liquid level drop that can be explained by vehicle movement; S32: Residual tolerance boundary determination submodule. The residual tolerance boundary is used to determine whether the unexplained liquid level residual has exceeded the range that can be explained by normal transportation disturbance. Since the unexplained liquid level residual is affected by the prediction error of normal liquid level change, the local noise of the liquid level sensor, and the unrecovered vertical impact, the residual tolerance boundary determination submodule calculates the uncertainty of the prediction of normal liquid level change, the level noise level, and the liquid level boundary increment of the unrecovered vertical impact respectively, and then synthesizes them into the residual tolerance boundary. The residual tolerance boundary determination submodule first calculates the prediction error based on historical normal transportation samples, as shown in the following calculation: ， in, Represents historical sampling records The corresponding prediction error, Represents historical sampling records The corresponding actual change in liquid level. Represents historical sampling records Corresponding predicted normal liquid level change value; Based on the historical normal transport sample set corresponding to the current calibration category, the uncertainty of predicting normal liquid level changes is calculated as follows: ， in, Indicates time The degree of uncertainty in predicting normal liquid level changes Indicates the current calibration category The corresponding historical normal transportation sample set, Represents a set Number of historical normal transport samples This indicates the current calibration category, determined by the vehicle number, tank structure type, actual liquid level range, vehicle speed range, and operating condition label. Represents a set Any historical sampling record in China Represents historical sampling records The corresponding prediction error, Represents a set The average value of the prediction error; The level noise level is used to characterize the effect of fluctuations in the level sensor readings within the current local noise window on unexplained level residuals. The impact of the difference is assessed by calculating the level noise level based on the actual liquid level change within the current local noise window, as shown in the following calculation: ， in, Indicates time The level of liquid level noise, Indicates the end time Local noise window, and These represent the sampling times within the local noise window. and These represent the actual liquid level changes at the corresponding sampling times. This indicates taking the median. This represents a dimensionless constant used to convert the median absolute deviation to the order of standard deviation. Reference vertical impact level response This is only a reference scale for describing historical vertical impact level responses and cannot be multiplied by a dimensionless scale to be used as the level increment in the current residual tolerance boundary. To ensure that the residual tolerance boundary reflects the actual impact before the current vertical impact has recovered, the residual tolerance boundary determination submodule uses the absolute value of the current vertical impact level response as the vertical impact unrecovered level boundary increment, as shown below: ， in, Indicates time The vertical impact did not recover the liquid level boundary increment. Indicates time Vertical impact liquid level response. Based on the basic allowable boundary, the uncertainty of normal liquid level change prediction, the level noise level, and the unrecovered liquid level boundary increment due to vertical impact, the residual allowable boundary is calculated as follows: ， in, Indicates time The residual tolerance boundary, This indicates the basic allowable boundary determined based on historical normal transportation data. Indicates time The degree of uncertainty in predicting normal liquid level changes Indicates time The level of liquid level noise, Indicates time The vertical impact did not recover the liquid level boundary increment. , and This represents a dimensionless adjustment coefficient, used to adjust the impact of the uncertainty in predicting normal liquid level changes, the level of liquid level noise, and the impact of the unrecovered liquid level boundary increment due to vertical impact on the residual allowable boundary.

7. A wastewater transportation management method according to claim 6, characterized in that: S3 also includes the following sub-steps: S33: The disturbance recovery verification submodule calculates the disturbance intensity based on vehicle attitude and acceleration changes. When the disturbance intensity drops below a threshold, a recovery observation window is formed to calculate the consistency of the residual direction. The specific calculation is as follows: ， in, Indicates time The intensity of the disturbance under operating conditions, This represents the difference between the vehicle pitch angle at time t and the vehicle pitch angle at the previous sampling time. This represents the difference between the vehicle roll angle at time t and the vehicle roll angle at the previous sampling time. and Representing time respectively Compared to the vehicle's three-axis acceleration vector at the previous sampling time, This represents the normalized reference for vehicle pitch angle variation. This represents the normalized baseline for vehicle roll angle variation. The three normalized references represent the vehicle acceleration vector change normalization benchmarks. They are determined by the preset upper quantile values ​​of the corresponding changes in historical normal transportation data. , and This represents the non-negative dimensionless disturbance adjustment coefficient; the sum of the three is one. , and These represent positive numbers with the same dimensions as the changes in vehicle pitch angle, vehicle roll angle, and vehicle acceleration vector, respectively. Represents absolute value. The second normal form of a vector; When the intensity of the disturbance decreases from above the disturbance identification threshold to below the disturbance identification threshold, the liquid level is determined to have entered the recovery phase. A recovery observation window is formed by selecting consecutive sampling times within this recovery phase. The consistency of the residual direction is calculated based on the cancellation of unexplained liquid level residuals within the recovery observation window, as detailed below: ， in, Indicates time The degree of consistency of residual direction Indicates the sampling time within the observation window to be restored. Unexplained liquid level residuals This indicates the sampling time within the restored observation window. This indicates that the observation window length has been restored. Represents absolute value. Represents a positive number that has the same dimensions as the unexplained liquid level residual; S34: The continuous anomaly accumulation submodule performs time decay accumulation on residuals that exceed the residual tolerance boundary and point in the direction of liquid level decline within the anomaly judgment window. When the accumulated amount exceeds the threshold and the intensity of the operating condition disturbance is continuously lower than the threshold, a suspected abnormal liquid loss identification result is generated. The specific calculation is as follows: ， in, Indicates time The cumulative amount of abnormal liquid level; Indicates time The degree of consistency of residual direction; Indicates the sampling time within the anomaly detection window; Indicates the length of the exception detection window; Indicates the sampling time The reliability of the corresponding liquid level data is determined by taking one value for valid sampling and a preset reduction value for interpolated sampling. Indicates the time decay coefficient; Indicates the sampling time Distance from the current sampling time The number of sampling intervals; Indicates the sampling time Corresponding unexplained liquid level residuals; Indicates the sampling time The corresponding residual tolerance boundary; This indicates that the larger value among the listed values ​​is selected. , and All are dimensionless quantities. This represents the unilateral level drop exceeding the residual tolerance boundary, therefore This represents the cumulative drop in liquid level within the anomaly detection window. The zero value represents the amount of liquid level drop, and and They have the same dimensions.

8. The sewage transportation management method according to claim 7, characterized in that: S4 includes the following sub-steps: S41: The real-time water quality change judgment submodule determines the water quality parameters at the time of liquid filling as the initial value, calculates the normalized deviation degree for each set of valid water quality parameters obtained during transportation, and takes the maximum value as the water quality deviation index. When the index exceeds 1 and continuously reaches the preset sampling number, a real-time water quality change anomaly result is generated. The specific calculation is as follows: ， in, Indicates time Water quality deviation index, This represents a set of water quality parameters consisting of pH value, conductivity, turbidity, chemical oxygen demand, and ammonia nitrogen content. Indicates time The Water quality parameters, This indicates that the liquid collection is complete. Initial water quality parameters, Indicates the first The preset allowable variation of the water quality parameters, Indicates the relationship with the first All water quality parameters have zero-positive values ​​of the same dimension. This indicates that the larger value among the listed values ​​will be selected; S42: The arrival comparison and cross-verification submodule generates a QR code containing the initial liquid level, initial water quality parameters, liquid filling time, and vehicle number after the liquid is filled. Upon arrival, it reads the ending liquid level and ending water quality parameters and compares them with the initial values. If the liquid level difference exceeds the preset range, it generates a total abnormal result. If the water quality parameter difference exceeds the preset range, it generates an arrival water quality change abnormal result. The total abnormal result is cross-verified with the suspected abnormal liquid loss identification result. If both are abnormal, a transportation abnormal alarm result is generated. If only one is abnormal, a manual verification result is generated.

9. An electronic device for sewage transport management, characterized in that: A wastewater transportation management method according to any one of claims 1-8 includes a data acquisition module, an operating condition judgment module, a leakage judgment module, a transportation comparison module, and a display and alarm module. The data acquisition module is used to collect multi-source real-time data from the transport vehicle, generate an aligned data sequence with the sampling time of the liquid level sensor as the unified time axis, and establish steady-state transport reference parameters after the liquid loading is completed. The operating condition judgment module is used to identify the driving conditions based on the vehicle attitude and acceleration information provided by the data acquisition module, and to calculate a dynamic alarm threshold for distinguishing between normal liquid surface sloshing and leakage behavior based on the driving conditions. The leakage detection module is used to compare the real-time liquid level with the dynamic alarm threshold to determine whether a leakage alarm is triggered, and to generate a suspected abnormal liquid loss identification result based on the continuous abnormal accumulation of unexplained liquid level residuals within a preset window. The transport comparison module is used to compare the final liquid level and water quality parameters after arrival with the initial liquid level and water quality parameters when the liquid filling is completed, and to cross-verify the liquid level comparison result with the suspected abnormal liquid loss identification result, so as to output the transport abnormality alarm result or the manual verification result. The display alarm module is used to show monitoring data to managers and, based on the judgment results of the leakage judgment module or the transportation comparison module, push alarm information containing the type of abnormality and the time of occurrence when an abnormality occurs.

10. The wastewater transport management electronic device according to claim 9, characterized in that: The data acquisition module includes a multi-source transportation data acquisition submodule, a data synchronization and validity processing submodule, and a steady-state transportation benchmark determination submodule; The multi-source transportation data acquisition submodule includes a liquid level acquisition unit for acquiring liquid level, an attitude acquisition unit for detecting attitude, an acceleration acquisition unit for detecting acceleration, a positioning and speed measurement unit for positioning and speed measurement, a vehicle interface reading unit for reading vehicle driving status, and a water quality detection unit for detecting water quality parameters. The operating condition judgment module includes a transportation operating condition characteristic determination submodule for judging driving operating conditions, an attitude liquid level response calculation submodule for calculating attitude liquid level response, an inertial liquid level response calculation submodule for calculating inertial liquid level response, a vertical impact liquid level response calculation submodule for calculating vertical impact liquid level response, and a normal liquid level change fusion submodule for synthesizing normal liquid level change prediction values. The leakage detection module includes an unexplained liquid level residual calculation submodule for calculating unexplained liquid level residuals, a residual allowable boundary determination submodule for determining residual allowable boundaries, a disturbance recovery verification submodule for calculating the intensity of operating condition disturbances and the degree of consistency of residual directions, and a continuous anomaly accumulation submodule for generating suspected abnormal liquid loss identification results. The transport comparison module includes a real-time water quality change judgment submodule for judging real-time water quality changes and an arrival comparison and cross-verification submodule for performing arrival comparison and cross-verification.

Images