A method and system for evaluating the quality of hydrological monitoring data in an alpine region

Abstract

The application discloses a kind of alpine region hydrological monitoring data quality evaluation method and system, the method is by collecting the air temperature of monitoring point, water temperature, ground temperature and the hydrological monitoring data to be evaluated, with the arithmetic mean of three temperatures as the comprehensive temperature to build freeze-thaw state index, and combined with its change rate, monitoring point is discriminated as thawing state, freezing transition state, frozen ice cover state or ablation transition state;Further, according to the freeze-thaw state, the corresponding quality control rules and dynamic threshold in the state quality control rule library are called to detect the abnormality of hydrological data and determine the rationality;The application adjusts the quality control rules and threshold according to the freeze-thaw state, so that normal phenomena such as constant water level during the frozen period, sudden drop of soil liquid water content after ground temperature exceeds the freezing point, and sudden change of flow under ice cover are not misjudged as abnormal, while effectively detecting real faults during the frozen period, significantly reducing the false positive rate, false negative rate and manual review workload.

Description

Technical Field

[0001] This invention relates to the field of hydrological monitoring data processing technology, and in particular to a method and system for assessing the quality of hydrological monitoring data in high-altitude and cold regions. Background Technology

[0002] High-altitude and cold regions (such as high mountains, high latitudes, and the Qinghai-Tibet Plateau) are the source of many rivers and important water conservation areas. Hydrological monitoring data from these regions is crucial for water resource management, flood and ice jam warnings, ecological protection, and climate change research. Due to the cold climate, water bodies and soils in these regions undergo significant freeze-thaw cycles throughout the year. Rivers experience unique hydrological processes such as freezing, thawing, and thawing, resulting in distinctly different variation patterns in water level, flow rate, and soil moisture content compared to temperate and warm-climate regions.

[0003] Existing hydrological monitoring data quality control methods are mostly designed for unobstructed flow conditions, typically employing fixed threshold range checks, rate of change checks, and persistence checks—general rules that fail to consider the impact of freeze-thaw cycles on the normal variation patterns of hydrological elements. When these general methods are directly applied to high-altitude and cold regions, many seemingly abnormal but actually normal freeze-thaw related phenomena are misjudged as anomalies or sensor malfunctions. These include: prolonged periods of constant water level during the freezing period; after a river freezes, the free water surface is fixed by ice, and the water level can remain unchanged for a long time, which the general persistence check will misjudge as sensor malfunction. A sudden drop in soil liquid water content after the ground temperature exceeds the freezing point; when the ground temperature drops below 0℃, soil pore water undergoes a phase change and freezes, causing a rapid decrease in liquid water content in a short period, which the general rate of change check will misjudge as anomaly. Sudden changes in flow rate under ice cover; after an ice cover forms in a river channel, the hydraulic conditions of the flow section change, and the measured flow rate deviates significantly from the water level-flow rate constant curve during the unobstructed flow period, which the general method will misjudge as anomalies. The aforementioned misjudgments, on the one hand, generate a large number of false alarms and significantly increase the workload of manual verification, and on the other hand, may also cover up the actual sensor failures that occurred during the freezing period, leading to missed reports.

[0004] Therefore, there is an urgent need for a method to assess the quality of hydrological monitoring data in high-altitude and cold regions that can identify the freeze-thaw state of monitoring points and adaptively adjust quality control rules accordingly. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for assessing the quality of hydrological monitoring data in high-altitude and cold regions, in order to solve the problem that existing general quality control methods in high-altitude and cold regions misjudge normal hydrological phenomena related to freeze-thaw cycles as abnormalities, and at the same time easily miss the actual faults during the freezing period.

[0006] To achieve the above objectives, in a first aspect, the present invention provides a method for assessing the quality of hydrological monitoring data in high-altitude and cold regions, comprising the following steps: Step S1: Collect air temperature, water temperature, ground temperature, and hydrological monitoring data to be evaluated at hydrological monitoring points in high-altitude and cold regions; Step S2: Calculate the comprehensive temperature representing the thermal state of the monitoring point based on the air temperature, water temperature, and ground temperature, and construct a freeze-thaw state index representing the current freeze-thaw or freezing degree of the monitoring point based on the comprehensive temperature and its rate of change over time. Step S3: Based on the freeze-thaw state index, determine the current freeze-thaw state of the monitoring point as thawing state, freezing transition state, frozen ice cover state, or melting transition state; Step S4: Based on the identified freeze-thaw state, retrieve the quality control rules and judgment thresholds corresponding to the freeze-thaw state from the pre-established sub-state quality control rule library; Step S5: Using the invoked quality control rules and judgment thresholds, perform anomaly detection and rationality judgment on the hydrological monitoring data, identify the hydrological phenomenon that conforms to physical laws under the freeze-thaw state as a normal phenomenon, otherwise judge it as a sensor malfunction. Step S6: Output the quality label of the hydrological monitoring data and issue an early warning for data identified as sensor malfunction.

[0007] Furthermore, the hydrological monitoring data to be evaluated includes water level, flow rate, and soil liquid water content; the air temperature, water temperature, and ground temperature are collected synchronously by corresponding sensors deployed at monitoring points at the same sampling interval, with the sampling interval not exceeding 1 hour.

[0008] Furthermore, the comprehensive temperature is taken as the arithmetic mean of air temperature, water temperature, and ground temperature. The freeze-thaw state index is calculated according to... Build, in The characteristic temperature scale is set to 5°C; the rate of change is the quotient of the difference between the comprehensive temperatures over a sampling interval. .

[0009] Furthermore, the range of the freeze-thaw state index is [ 1,1], set the first state threshold Second state threshold The index is not greater than If the state is classified as melting, and the state is between melting and freezing with a rate of change less than 0, then the state is frozen. If the state is between melting and freezing with a rate of change not less than 0, then the state is ablation. It was determined to be in a frozen ice cap state.

[0010] Furthermore, the determination threshold includes a dynamic threshold for anomaly detection. ,in , The threshold coefficient represents the mean and standard deviation of the historical residuals under the current state. Take 3 for the molten state, 4 for the frozen ice cap state, and 5 for the frozen transition state and the ablation transition state.

[0011] Furthermore, the quality control rules include state-related continuity check rules: when the water level remains unchanged within 1 cm for a duration exceeding the constant allowable duration of the current state, it is judged as a suspected sensor jamming, wherein the melting state is taken as 24 hours and the frozen ice cover state is taken as 240 hours.

[0012] Furthermore, the quality control rules include a directional exemption rule under the frozen transition state: when the local temperature drops from above the freezing point to below the freezing point, the change in the direction of the decrease in soil liquid water content is exempted and the change rate alarm is not triggered.

[0013] Furthermore, the quality control rules include the ice cover flow relationship rules under the frozen ice cover state: the expected value of the flow rate is taken as the product of the flow rate during the smooth flow period and the ice cover reduction coefficient, wherein the ice cover reduction coefficient is taken as 0.7.

[0014] Furthermore, in step S5, the direction and magnitude of the suspected abnormal changes are compared with the expected physical behavior under the current freeze-thaw state to determine its rationality. In step S6, the quality labels are marked according to four levels: correct, suspicious, incorrect, and missing.

[0015] Secondly, based on the same inventive concept, this invention provides a hydrological monitoring data quality assessment system for high-altitude and cold regions, used to implement the above-mentioned method, including a data acquisition module, a freeze-thaw state discrimination module, a quality control rule switching module, a data quality detection module, and a quality assessment output module.

[0016] Furthermore, the data acquisition module is used to simultaneously collect air temperature, water temperature, ground temperature, water level, flow rate, and soil liquid moisture content; the freeze-thaw state discrimination module is used to calculate the comprehensive temperature, construct the freeze-thaw state index, and determine the freeze-thaw state; the quality control rule switching module has a built-in sub-state quality control rule library and calls the corresponding rules and thresholds according to the state; the data quality detection module is used to perform anomaly detection and rationality judgment; and the quality assessment output module is used to output quality labels and provide early warning of sensor faults.

[0017] Compared with the prior art, the beneficial effects of the present invention are: This invention constructs a freeze-thaw state index based on air temperature, water temperature, and ground temperature to determine the freeze-thaw state of monitoring points. It then applies corresponding quality control rules and thresholds based on the state, preventing normal freeze-thaw-related hydrological phenomena such as constant water levels during the freezing period, a sharp drop in soil liquid water content after the ground temperature exceeds the freezing point, and sudden changes in flow rate under the ice cover from being misjudged as abnormal. This significantly reduces false alarm rates and the workload of manual verification. Simultaneously, through state-related dynamic thresholds and reasonableness judgments, while relaxing the criteria for normal freeze-thaw phenomena, it can still effectively detect sensor faults that actually occur during the freezing period, avoiding missed reports. This comprehensively improves the accuracy and reliability of hydrological monitoring data quality assessment in high-altitude and cold regions. The system parameters of this invention are few, the physical meaning is clear, and it is easy to promote and apply across different sites. Attached Figure Description

[0018] Figure 1 This is a flowchart of a method for assessing the quality of hydrological monitoring data in high-altitude and cold regions according to the present invention; Figure 2 This is a schematic diagram illustrating the relationship between the freeze-thaw state index and the freeze-thaw state discrimination of the present invention; Figure 3 This is a schematic diagram of the composition of a hydrological monitoring data quality assessment system for high-altitude and cold regions according to the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention; the parameter values ​​involved in the embodiments are self-consistent examples given for ease of explanation, and can be recalibrated based on historical data from monitoring stations in practical applications.

[0020] Example 1 This embodiment provides a method for assessing the quality of hydrological monitoring data in high-altitude and cold regions, such as... Figure 1 As shown, the method includes the following steps: Step S1: Collect air temperature, water temperature, ground temperature, and hydrological monitoring data to be evaluated at hydrological monitoring points in high-altitude and cold regions.

[0021] Air temperature sensors, water temperature sensors, and ground temperature sensors, as well as hydrological element sensors for water level, flow rate, and soil liquid moisture content, are deployed at hydrological monitoring points in high-altitude and cold regions. All sensors collect data synchronously at the same sampling interval, which is no greater than one hour (one hour in this embodiment). Air temperature is then obtained from these data. Water temperature Geothermal and the water level to be assessed ,flow and soil liquid water content .

[0022] The reason for collecting only the above three temperatures for state determination is that air temperature, water temperature, and ground temperature respectively characterize the thermal state of the near-surface atmosphere, river water, and soil. The three together determine the freeze-thaw / icing process at the monitoring point, which can cover the main freeze-thaw phenomena such as freezing, sealing, and soil freezing, while avoiding collecting too many parameters that are difficult to integrate and interpret.

[0023] Step S2: Calculate the comprehensive temperature representing the thermal state of the monitoring point based on the air temperature, water temperature, and ground temperature, and construct a freeze-thaw state index representing the current freeze-thaw or freezing degree of the monitoring point based on the comprehensive temperature and its rate of change over time.

[0024] Overall temperature Take the arithmetic mean of air temperature, water temperature, and ground temperature: .

[0025] In the absence of sufficient prior evidence to distinguish the primary and secondary values ​​of the three temperatures, using an equal-weighted average avoids introducing multiple weight parameters that require station-by-station calibration and could weaken the model's interpretability. Based on this, a freeze-thaw state index is constructed. : ; in The characteristic temperature scale is taken as 5℃. This index smooths and normalizes the overall temperature to [ [1,1]: The lower the overall temperature (more inclined towards freezing), the closer the index is to +1; the higher the temperature (more inclined towards melting), the closer the index is to... 1.

[0026] Pick =5℃, so that when the comprehensive temperature deviates from the freezing point by about 5℃, the index is about 0.76 and close to saturation (about 0.96) when it deviates by about 10℃. This matches the seasonal fluctuation range of temperature in high-altitude and cold regions, while ensuring that the index is approximately linear near the freezing point and most sensitive to temperature changes in the phase transition critical region.

[0027] Further calculate the rate of change of overall temperature Take the difference quotient within one sampling interval: ;in The sampling interval is specified. The sign of the rate of change reflects the current cooling and freezing trend at the monitoring point. <0) or a trend of warming and melting ( ≥0).

[0028] Step S3: Based on the freeze-thaw state index, determine the current freeze-thaw state of the monitoring point as thawing state, freezing transition state, ice cap state, or melting transition state.

[0029] Set the first state threshold Second state threshold The discrimination rule is as follows: when ≤ It is considered to be in a molten state at 0.5; when 0.5< <0.5 and When the value is less than 0, it is considered a frozen transition state; when 0.5< <0.5 and A value ≥0 is considered an ablation transition state; when A value ≥0.5 indicates a frozen ice cap state.

[0030] Solving tanh(x) = 0.5, we get x ≈ 0.55, corresponding to a comprehensive temperature of approximately ±2.7℃. This means that a comprehensive temperature approximately 2.7℃ above the freezing point is classified as the thawing state, and approximately 2.7℃ below the freezing point is classified as the frozen ice cap state; the state between these two is the transitional state. In high-altitude and cold regions, water freezing and shallow soil freezing mainly occur within a range of approximately ±3℃ near the freezing point. This bandwidth can accommodate temperature fluctuations during phase transitions, avoid frequent erroneous state switching, and clearly distinguish between stable thawing and stable frozen states. Furthermore, the transition zone is subdivided into freezing transitional states and ablation transitional states based on the sign of the rate of change. The relationship between the freeze-thaw state index and state discrimination is as follows: Figure 2 As shown.

[0031] Step S4: Based on the identified freeze-thaw state, retrieve the quality control rules and judgment thresholds corresponding to the freeze-thaw state from the pre-established sub-state quality control rule library.

[0032] Anomaly detection employs state-dependent dynamic thresholds: In the formula, the expected value of any hydrological monitoring quantity under the current state is taken. (t) (similar to a robust sliding estimate of the state history sample), the residual is the absolute value of the difference between the measured value and the expected value; , Let be the mean and standard deviation of the historical residual samples under the current state s, respectively. This is the threshold coefficient for this state.

[0033] The threshold coefficient is assigned a value based on the state: 3 for the thawing state, 4 for the frozen ice sheet state, and 5 for the freezing transition state and the thawing transition state. The value is based on the fact that the hydrological process is stable and the residuals are approximately normally distributed under the thawing state. =3 (the classic 3σ criterion, corresponding to a false alarm probability of approximately 0.3% under a normal distribution), which can effectively detect true anomalies; under the freezing and thawing transition states, physical processes such as soil water phase change and ice condition evolution will produce large but normal residuals, which are taken as... =5 (approximately 5σ, with a false alarm probability of less than one in a million under a normal distribution), avoiding misjudging drastic but normal phase transition processes as abnormal without missing real faults (whose residuals are much greater than 5σ); the frozen ice sheet state is generally stable with small residuals, but occasional factors such as ice sheet compression and cracking can cause step jumps, so a value between the two is taken. =4.

[0034] In addition to dynamic thresholds, the rule base also includes the following state-related rules: (1) Continuous inspection rule: When the water level remains unchanged within the constant judgment threshold ε = 1 cm for a duration exceeding the constant allowable duration of the current state. If the sensor is suspected to be stuck, it is considered to be ε. The typical resolution and noise level of the water level sensor are taken as ε, and the fluctuation is considered constant if it does not exceed 1 cm. The allowable constant time is determined according to the state: 24 hours for the thawing state and 240 hours (about 10 days) for the frozen ice cover state.

[0035] In the melting state (flowing period), the water level of the river channel fluctuates continuously with the runoff. If there is no change for 24 consecutive hours, it is judged as stuck. In the frozen state, the river channel is frozen by the ice layer and the free water surface is fixed. It is normal for the water level to remain constant for a long time. Therefore, 10 days is taken as a conservative lower limit. Only if there is no change for more than 10 consecutive days is it judged as a suspected fault.

[0036] (2) Directional Exemption Rule: Under the frozen transition state, when the local temperature From above the freezing point to below the freezing point (i.e. Crossing 0℃ and When <0), soil pore water freezes and liquid water content drops sharply according to the phase transition law. This is the normal physical response of the Stefan phase transition process. Therefore, the change in the direction of the decrease of soil liquid water content under this state is exempted and the change rate alarm is not triggered. Only the increase direction or the sudden change in the non-freezing period is checked.

[0037] (3) Ice cover flow rate relationship rule: Under the state of ice cover, the ice cover increases the flow resistance and reduces the effective flow area, causing the water level-flow rate relationship to deviate from the calibration curve during the free flow period; therefore, the expected value of the flow rate is changed to the calibration flow rate during the free flow period and the ice cover reduction factor. The product of We take 0.7 (ice caps typically reduce the effective flow rate at a given water level to 60%~80% of the free flow rate; 0.7 is taken as a representative value) to make the sudden drop in flow rate under ice cap relative to the free flow curve normal according to the ice cap relationship.

[0038] Step S5: Using the invoked quality control rules and judgment thresholds, perform anomaly detection and rationality judgment on the hydrological monitoring data. Identify hydrological phenomena that conform to physical laws under freeze-thaw conditions as normal phenomena; otherwise, judge them as sensor malfunctions.

[0039] For data deemed as potentially abnormal, their direction and magnitude of change are further compared with the expected physical behavior under the current freeze-thaw state: if they are consistent (e.g., water level changes slowly or remains constant under frozen conditions, soil liquid water content decreases under frozen transition conditions, flow rate conforms to ice cover relationship under frozen conditions), they are identified as normal hydrological phenomena under this state and are suppressed and warned; if they are inconsistent (e.g., water level jumps far exceeding the dynamic threshold under frozen conditions), they are identified as sensor malfunctions.

[0040] Step S6: Output the quality label of the hydrological monitoring data and issue an early warning for data identified as sensor malfunction.

[0041] The quality labels are marked in four levels: correct, doubtful, incorrect, and missing, and are coded as 1, 2, 3, and 9 respectively: data that is normal and passes the inspection is marked as correct (1); data that cannot be confirmed but does not affect the use is marked as doubtful (2); data identified as sensor failure is marked as incorrect (3) and an early warning is generated; data that has no output or outputs out of bounds is marked as missing (9).

[0042] The following are examples of four typical scenarios: Scenario 1 (Constant water level during freezing period, suppressing false alarms): At a certain moment in deep winter, the temperature... = 18℃, water temperature =0.1℃, ground temperature = 5℃. The overall temperature is: The freeze-thaw index is: FSI=0.91≥0.5, indicating a frozen ice cap state.

[0043] Water level during this period The sensor has remained constant at 1.253 meters (with a fluctuation of less than 1 centimeter) for 72 hours. A general continuous monitoring test with an allowable duration of 24 hours resulted in a false alarm indicating sensor jamming because 72 > 24. This method, using an allowable duration of 240 hours under frozen ice conditions, correctly identified 72 < 240, successfully suppressing false alarms.

[0044] Scenario 2 (Ground temperature exceeds freezing point, soil moisture content drops sharply, suppressing false alarms): At a certain moment in autumn, the air temperature... = 2℃, water temperature =0.3℃, ground temperature = 0.8℃, overall temperature = 0.83℃, freeze-thaw index =tanh(0.83 / 5)≈0.17; the combined temperature at the previous sampling time was approximately 0.2℃, therefore the rate of change =( 0.83 ( 0.2)) / 1≈ 0.63℃ / hour <0. (From) 0.5 < 0.17 < 0.5 and If the value is less than 0, it is considered a frozen transition state.

[0045] Soil liquid water content during this period The percentage decreased from 31.5% to 12.8% within 3 hours, with a maximum single-step decrease of 8.5%. A general rate of change check using μ=0.8%, σ=1.2%, and a coefficient of 3 yielded a threshold of 4.4%, which resulted in a false alarm because 8.5% > 4.4%. This method uses a threshold coefficient of 5 for the frozen transition state, and the residual statistics for this state are as follows: =3.5%, =2.8%, dynamic threshold Because 8.5% < 17.5%; at the same time, the ground temperature changed from positive to negative. If the value is less than 0, the directional exemption rule is triggered, exempting the descent direction and classifying it as normal.

[0046] Scenario 3 (Sudden Drop in Flow Rate Under Ice Sheet, Suppressing False Alarms): The scenario involves a frozen ice sheet. The water level-flow rate constant curve during the open flow period indicates a flow rate of 12.0 cubic meters per second at the current water level, while the measured flow rate is 8.3 cubic meters per second, a sudden drop of approximately 31% relative to the open flow curve. Common methods would falsely report this as abnormal. This method, however, uses an ice sheet reduction factor of 0.7 according to the ice sheet flow rate relationship rules, resulting in an expected flow rate of 12.0 × 0.7 = 8.4 cubic meters per second. The measured flow rate of 8.3 is close to the expected 8.4, with a residual difference of only 0.1, and is therefore judged as normal.

[0047] Scenario 4 (Real-world fault during freezing period, to avoid missed reports): Still under ice cover. Water level at a certain moment. The water level jumped from 1.253 meters to 1.850 meters (a jump of 0.597 meters) and then back down. The statistical value of the residual water level under this condition is as follows: =0.002 meters =0.008 meters, threshold coefficient is 4, dynamic threshold A jump of 0.597 meters, significantly greater than 0.034 meters, is considered a suspected anomaly. Furthermore, this positive jump contradicts the physical expectation of a gradual or constant water level under frozen conditions. Based on a reasonableness assessment, it is identified as a sensor malfunction, marked as an error, and an alert is issued. This demonstrates that even with relaxed continuous checks and threshold coefficients under frozen ice conditions, this method can still effectively detect genuine faults without false negatives.

[0048] In an exemplary application at a high-altitude hydrological station, the false alarm rate was approximately 18.4% and the missed alarm rate was approximately 7.1% when using the general quality control method; after adopting this method, the false alarm rate was reduced to approximately 3.5% and the missed alarm rate was reduced to approximately 2.0%, and the workload of manual review was reduced by approximately 80%.

[0049] Example 2 This embodiment provides a hydrological monitoring data quality assessment system for high-altitude and cold regions, used to implement the method described in Embodiment 1. Figure 3 As shown, the system includes a data acquisition module, a freeze-thaw state discrimination module, a quality control rule switching module, a data quality detection module, and a quality assessment output module. These five modules are connected in sequence. The system also includes a connected multi-parameter sensor array and a state-specific quality control rule library.

[0050] The data acquisition module includes air temperature sensors, water temperature sensors, and ground temperature sensors deployed at hydrological monitoring points in high-altitude and cold regions, as well as water level sensors, flow sensors, and soil liquid moisture content sensors; wherein the air temperature sensor, water temperature sensor, and ground temperature sensor are used to collect air temperature data. Water temperature Geothermal Water level sensors, flow sensors, and soil liquid moisture content sensors are used to collect water level data as hydrological monitoring data to be evaluated. ,flow and soil liquid water content Each sensor collects data synchronously at the same sampling interval, which is no more than 1 hour; in this embodiment, 1 hour is used. The data acquisition module is connected to the freeze-thaw state discrimination module and the data quality detection module, and outputs the collected data to the two modules respectively.

[0051] The freeze-thaw state discrimination module is connected to the data acquisition module and is used to perform the following processes in sequence: First, according to... The comprehensive temperature is calculated by taking the arithmetic mean of air temperature, water temperature, and ground temperature; then... Construct a freeze-thaw state index, with characteristic temperature scales. Take 5℃ and follow Calculate the rate of change of the overall temperature over a sampling interval; finally, based on the first state threshold... Second state threshold and rate of change The sign of the freeze-thaw index is determined by the following criteria: the freeze-thaw index is not greater than 1. Determined to be in a molten state, between and Between and <0 indicates a frozen transition state. ≥0 is considered an ablation transition state, and not less than The state is determined to be a frozen ice sheet; the determination result is output to the quality control rule switching module. The relationship between the freeze-thaw state index and the freeze-thaw state determination is as follows: Figure 3 As shown.

[0052] The quality control rule switching module is connected to the freeze-thaw state discrimination module, and it has a built-in state-specific quality control rule library; the rule library is configured with dynamic thresholds for anomaly detection according to the freeze-thaw state. and its threshold coefficient —The thawing state is set to 3, the frozen ice cover state to 4, and the freezing transition state and the thawing transition state to 5; the water level continuity check rules and their constant judgment threshold of 1 cm and constant allowable duration are set to 24 hours for the thawing state and 240 hours for the frozen ice cover state; the directional exemption rules for the direction of soil liquid water content decrease under the freezing transition state; and the ice cover flow relationship rules under the frozen ice cover state, with an ice cover reduction coefficient of 0.7; this module calls the corresponding quality control rules and judgment thresholds according to the identified freeze-thaw state and outputs them to the data quality detection module.

[0053] The data quality detection module is connected to both the data acquisition module and the quality control rule switching module. It is used to perform anomaly detection and rationality assessment of water level, flow rate, and soil liquid moisture content using the invoked quality control rules and judgment thresholds. Calculate the dynamic threshold under the current freeze-thaw state, and judge those with residuals greater than the dynamic threshold as suspected anomalies; then compare the change direction and amplitude of the suspected anomalies with the expected physical behavior under the current freeze-thaw state. If they match, they are identified as normal hydrological phenomena under this state, and if they do not match, they are identified as sensor malfunctions; the judgment results are output to the quality assessment output module.

[0054] The quality assessment output module is connected to the data quality detection module. It is used to distinguish between normal hydrological phenomena and sensor malfunctions under the freeze-thaw state based on the detection and judgment results. It outputs quality labels for each hydrological monitoring data according to four levels: correct, suspicious, incorrect, and missing, coded as 1, 2, 3, and 9 respectively, and generates early warnings for data identified as sensor malfunctions.

[0055] The above modules can be implemented by hydrological telemetry terminals, data processing servers, or corresponding software function modules with data acquisition, storage, and calculation functions. Each sensor is connected to the data processing unit via wired or wireless communication links.

[0056] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for assessing the quality of hydrological monitoring data in high-altitude and cold regions, characterized in that, Includes the following steps: Step S1: Collect air temperature, water temperature, ground temperature, and hydrological monitoring data to be evaluated at hydrological monitoring points in high-altitude and cold regions; Step S2: Calculate the comprehensive temperature representing the thermal state of the monitoring point based on the air temperature, water temperature, and ground temperature, and construct a freeze-thaw state index representing the current freeze-thaw or freezing degree of the monitoring point based on the comprehensive temperature and its rate of change over time. Step S3: Based on the freeze-thaw state index, determine the current freeze-thaw state of the monitoring point as thawing state, freezing transition state, frozen ice cover state, or melting transition state; Step S4: Based on the identified freeze-thaw state, retrieve the quality control rules and judgment thresholds corresponding to the freeze-thaw state from the pre-established sub-state quality control rule library; Step S5: Using the invoked quality control rules and judgment thresholds, perform anomaly detection and rationality judgment on the hydrological monitoring data, identify the hydrological phenomenon that conforms to physical laws under the freeze-thaw state as a normal phenomenon, otherwise judge it as a sensor malfunction. Step S6: Output the quality label of the hydrological monitoring data and issue an early warning for data identified as sensor malfunction.

2. The method for assessing the quality of hydrological monitoring data in high-altitude and cold regions according to claim 1, characterized in that, The hydrological monitoring data to be evaluated includes water level, flow rate, and soil liquid moisture content; The air temperature, water temperature, and ground temperature are collected by air temperature sensors, water temperature sensors, and ground temperature sensors deployed at the monitoring points, respectively. Each sensor collects data synchronously at the same sampling interval, which is no more than 1 hour.

3. The method for assessing the quality of hydrological monitoring data in high-altitude and cold regions according to claim 2, characterized in that, In step S2, the comprehensive temperature Take the arithmetic mean of air temperature, water temperature, and ground temperature, that is... The freeze-thaw state index ,in The characteristic temperature scale is taken as 5℃; the rate of change Take the quotient of the differences in the overall temperature over a sampling interval, i.e. ,in The sampling interval is denoted as .

4. The method for assessing the quality of hydrological monitoring data in high-altitude and cold regions according to claim 3, characterized in that, In step S3, the freeze-thaw state index The range of values ​​for is [ 1,1], set the first state threshold Second state threshold ; When the freeze-thaw index is not greater than It is determined to be in a molten state at that time; When the freeze-thaw index is between and When the rate of change is less than 0, it is determined to be a frozen transition state; When the freeze-thaw index is between and When the rate of change is not less than 0, it is determined to be an ablation transition state; When the freeze-thaw index is not less than It was determined to be in a frozen ice cap state.

5. The method for assessing the quality of hydrological monitoring data in high-altitude and cold regions according to claim 5, characterized in that, The quality control rules and judgment thresholds used are invoked to perform anomaly detection and reasonableness assessment on the hydrological monitoring data. The judgment thresholds are dynamic thresholds. according to Confirmed, among which and These are the mean and standard deviation of the historical residual samples under the current freeze-thaw conditions, respectively. The threshold coefficients for the freeze-thaw state are set as follows: 3 for the thawing state, 4 for the frozen ice cover state, and 5 for the freezing transition state and the thawing transition state. When the residual of the hydrological monitoring data is greater than the corresponding dynamic threshold, it is judged as a suspected anomaly.

6. The method for assessing the quality of hydrological monitoring data in high-altitude and cold regions according to claim 5, characterized in that, The quality control rules include: when the water level remains unchanged within a constant threshold of 1 cm for a duration exceeding the constant allowable duration corresponding to the current freeze-thaw state, it is judged as a suspected sensor jamming. The allowable duration for the thawing state is set at 24 hours, and the allowable duration for the frozen ice cover state is set at 240 hours, so that the water level remains constant for a long time during the freezing period and is not judged as a sensor failure.

7. The method for assessing the quality of hydrological monitoring data in high-altitude and cold regions according to claim 6, characterized in that, The quality control rules include: when the ground temperature drops from above the freezing point to below the freezing point, the change in the direction of the decrease in soil liquid water content is exempted and the change rate alarm is not triggered, so that the sudden drop in soil liquid water content after the ground temperature exceeds the freezing point is not judged as a sensor failure.

8. The method for assessing the quality of hydrological monitoring data in high-altitude and cold regions according to claim 7, characterized in that, The quality control rules include: taking the expected value of the flow rate as the product of the flow rate obtained from the water level-flow rate calibration curve during the free flow period and the ice cover reduction factor, wherein the ice cover reduction factor is 0.7; and ensuring that a sudden drop in the flow rate under the ice cover relative to the free flow period calibration curve is judged according to the ice cover flow rate relationship and is not judged as a sensor failure.

9. The method for assessing the quality of hydrological monitoring data in high-altitude and cold regions according to claim 8, characterized in that, The rationality determination in step S5 includes: comparing the direction and amplitude of suspected abnormal changes with the expected physical behavior under the current freeze-thaw state; if they match, they are identified as normal hydrological phenomena under the freeze-thaw state and the warning is suppressed; if they do not match, they are identified as sensor malfunctions.

10. A hydrological monitoring data quality assessment system for high-altitude and cold regions, used to implement the method described in any one of claims 1 to 9, characterized in that, The system includes: a data acquisition module, a freeze-thaw state discrimination module, a quality control rule switching module, a data quality detection module, and a quality assessment output module; The data acquisition module includes air temperature sensors, water temperature sensors, ground temperature sensors, and hydrological element sensors deployed at hydrological monitoring points in high-altitude and cold regions, used to simultaneously collect air temperature, water temperature, ground temperature, water level, flow rate, and soil liquid moisture content; The freeze-thaw state discrimination module is connected to the data acquisition module and is used to calculate the comprehensive temperature based on air temperature, water temperature and ground temperature, thereby constructing a freeze-thaw state index, and judging the current freeze-thaw state of the monitoring point based on the freeze-thaw state index and its rate of change. The quality control rule switching module is connected to the freeze-thaw state discrimination module. It has a built-in state-specific quality control rule library, which is used to call the quality control rules and judgment thresholds corresponding to the identified freeze-thaw state according to the identified freeze-thaw state. The data quality detection module is connected to the data acquisition module and the quality control rule switching module respectively, and is used to perform anomaly detection and rationality judgment on hydrological monitoring data using the invoked quality control rules and judgment thresholds. The quality assessment output module is connected to the data quality detection module and is used to distinguish between normal hydrological phenomena and sensor malfunctions under the freeze-thaw condition, output quality labels for hydrological monitoring data, and issue warnings for data identified as sensor malfunctions.

Images