Water network operation design domain state index real-time calculation and hierarchical control method and system

Abstract

The application provides a water network operation design domain state index real-time calculation and hierarchical control method and system, relates to the technical field of water conservancy project autonomous operation safety monitoring and control. The application takes water level state, change rate, sensing credibility, execution capability and operation environment as unified evaluation objects, constructs operation design domain state indexes oriented to real-time working conditions, and realizes cooperative discrimination of conventional deviation and high-risk events in combination with a red line triggering mechanism; state indexes are classified and recognized according to intervals, and corresponding control strategies are matched, so that orderly switching from normal adjustment to limited adjustment and safety backup control is realized. The application can improve the integrity of state cognition, the timeliness of risk response and the rationality of control switching under the condition of not relying on single static threshold determination, thereby enhancing the safety guarantee capability and engineering applicability of water network operation under complex working conditions.

Description

Technical Field

[0001] This invention relates to the field of autonomous operation safety monitoring and control technology for water conservancy projects, and in particular to a method and system for real-time calculation and hierarchical control of water network operation design domain state indicators. Background Technology

[0002] In autonomous water network operation scenarios, the control system needs to continuously determine whether the current operating conditions are still within the preset applicable range. Unlike single equipment or isolated processes, engineering water networks are typically composed of water source units, backbone canals, series and parallel control gates, pumping stations, valves, storage tanks, and receiving terminals according to a clear topological hierarchy. Operational objectives are simultaneously constrained by multiple dimensions, including water delivery capacity, target canal water level, upstream and downstream water level relationships, equipment linkage status, and communication link status. Any anomaly in any single dimension can propagate through the topology to the entire network, creating systemic risks.

[0003] In existing technologies, the assessment of water network operation safety mainly relies on univariate threshold alarm methods. These methods use whether a certain observation exceeds a preset upper or lower limit as the criterion, and the output is usually a binary conclusion of either exceeding or not exceeding the limit. While this method is effective in cases of significant univariate anomalies, it is difficult to provide a unified state assessment for systems like water networks, which involve upstream and downstream transmission, multi-node coupling, and differences in topological location.

[0004] Furthermore, the operational status of the water network is characterized by multiple variables, heterogeneous dimensions, strong coupling, and significant differences in topological location. Flow rate, water level, pressure, water level difference before and after the gate, rainfall intensity, equipment health, and link delay each correspond to different physical quantities, and their value ranges, directions of change, and safety sensitivities are not the same. At the same time, the degree of impact of the same variable on the overall safety of the system also varies when it appears in different topological locations.

[0005] Furthermore, the determination of water network operation is not solely determined by continuous variables. Whether key gates can reach their target opening, whether the main pump units are in operational condition, whether the main and backup communication links meet control delay requirements, whether the downstream receiving pool has remaining capacity, and whether control interlocking conditions have been lifted all directly affect the system's applicability to the current operating conditions. In a water network scenario, failure of key equipment can lead to abrupt changes in the control capability boundary. If static thresholds or single over-limit alarms are still used, it is difficult to promptly identify the contraction of the operational design domain caused by the combined effects of equipment status, communication status, and downstream receiving conditions, and it is also difficult to provide a consistent basis for subsequent degraded operation or emergency backup control.

[0006] The existing technologies have the following technical problems: 1) The evaluation of the water network operation status relies on single-variable binary judgment and lacks a unified quantitative framework for multi-dimensional heterogeneous variables; 2) Existing methods do not incorporate logical constraints such as equipment health status and communication link status into the same status judgment system; 3) Existing technologies are difficult to continuously and computably quantify the current operation status of the water network and cannot support the graded judgment requirements of stable operation, degraded operation and emergency operation.

[0007] For the reasons mentioned above, existing technologies struggle to provide a unified, continuous, and calculable quantitative characterization of the current operational status of a water network comprised of channels, pumping stations, gates, and valves. Furthermore, they cannot output consistent judgment results based on unified indicators that take into account channel distribution processes, equipment linkages, and downstream water constraints. To address this issue, it is necessary to propose a real-time calculation and hierarchical control method for status indicators oriented towards the characteristics of water network objects. Summary of the Invention

[0008] To address the challenges of safety assessment caused by multi-source disturbance coupling, time-varying equipment states, and dynamic contraction of operating boundaries during water network operation, this invention provides a method and system for real-time calculation and hierarchical control of water network operation design domain state indicators. This solves the problem in existing technologies where it is difficult to incorporate upstream and downstream transmission relationships, channel distribution constraints, controlled unit associated water level constraints, gate, pump, and valve linkage conditions, downstream water receiving capacity, and control link states into the same state assessment framework, thus making it impossible to uniformly quantify and implement hierarchical control of the current operating state.

[0009] To achieve the above objectives, the technical solution adopted by this invention is: a method for real-time calculation and hierarchical control of water network operation design domain state indicators, comprising the following steps: S1. Obtain multidimensional observation data characterizing the current operating status of the water network; among which, multidimensional observation data includes continuous variable data and logical state data; S2. Based on multi-dimensional observation data, and according to the water network topology, upstream and downstream transmission and distribution relationships, and downstream water constraints, establish a set of constraints for the water network operation design domain. S3. Using the set of constraints of the water network operation design domain, determine the upper and lower bounds of the operation design domain corresponding to each continuous variable, and perform dimensionless normalization on the continuous variable data to construct the state sub-indicators corresponding to each continuous variable. S4. Based on the constructed state sub-indicators, a weight function is constructed according to the coupling relationship between multiple dimension indicators, and the state indicators of the water network operation design domain are calculated in combination with logical state constraints. S5. Based on the comparison results between the water network operation design domain status indicators and the preset interval boundaries, the current water network operation status is divided into stable operation, degraded operation, or emergency operation, and the corresponding control strategies are executed.

[0010] The beneficial effects of this invention are as follows: Addressing the difficulties in safety assessment caused by multi-source disturbance coupling, time-varying equipment states, and dynamic contraction of operating boundaries during water network operation, this invention uses water level status, rate of change, perception reliability, execution capability, and operating environment as unified evaluation objects to construct a real-time operating condition-oriented operation design domain state index. It also incorporates a red-line triggering mechanism to achieve collaborative discrimination between routine deviations and high-risk events. Based on this, the water network operating conditions are classified and identified according to the interval of the state index, and corresponding control strategies are matched to achieve an orderly switch from normal regulation to restricted regulation and safety-based control. This method can improve the completeness of state cognition, the timeliness of risk response, and the rationality of control switching in autonomous water network operation scenarios without relying on a single static threshold judgment, thereby enhancing the safety assurance capability and engineering applicability of water network operation under complex conditions.

[0011] Furthermore, the continuous variable data includes one or more of the following: channel flow rate, water level associated with the controlled unit, water level change rate, gate opening, actuator control quantity corresponding to pump station output or valve opening, actuator change rate, downstream water receiving pool water level or remaining storage capacity, rainfall intensity, equipment health, and link delay. The logical status data includes one or more of the following: health status of key sensors, availability status of key actuators, downstream water supply conditions met, availability status of the main communication link, and control lockout status.

[0012] The beneficial effects of the above-mentioned further solutions are as follows: Since continuous variable data is used to construct sub-indicators of each continuous state, and logical state data is used to construct logical penalty terms and red-line triggering conditions, this invention can uniformly incorporate continuous state deviations, equipment availability, communication availability, downstream water supply conditions, and control interlocking states into the operational design domain state indicators and hierarchical judgment process. Compared with methods that rely solely on thresholds for single continuous variables such as water level and flow rate, this invention can identify operational design domain contraction in advance when equipment conditions deteriorate, communication links malfunction, or downstream water supply conditions are limited, and trigger downgraded operation or emergency operation according to the severity of the anomaly, thereby improving the completeness of the autonomous operation status judgment of the water network, the timeliness of anomaly response, and the accuracy of control strategy switching.

[0013] Furthermore, the expression for the set of constraints in the water network operation design domain is as follows:

[0014] in, Indicates time Water network operation design domain constraint set, Represents a continuous variable state vector. Indicates the first A continuous variable, and They represent the first Lower and upper bounds of the operational design domain for a continuous variable. n This represents the total number of continuous variables included in the set of constraints of the operational design domain. Represents the set of edges in the water network topology. Represents a node With nodes The transmission flow between them This indicates the transmission and distribution relationship determined by upstream and downstream water levels and actuator control parameters. Indicates the first A single actuator control quantity, r This represents the total number of actuator control variables included in the set of constraints of the operating design domain. , and These represent the lower limit, upper limit, and maximum allowable variation of the actuator control quantity, respectively. Indicates downstream acceptance capacity. This indicates downstream demand or constrained water supply. Indicates the first A number of logical state constraints are satisfied, where s represents the total number of logical state constraints included in the set of constraints of the running design domain.

[0015] Furthermore, S3 includes: Using the set of constraints in the water network operation design domain, determine the upper and lower bounds of the operation design domain for each continuous variable; Based on the i The current observation of the nth continuous variable and the nth i The continuous variables correspond to the lower bound of the water network operation design domain and the first continuous variable. i The upper bound of the water network operation design domain is corresponding to each continuous variable, and the dimensionless normalized value is obtained. Normalization is performed based on dimensionless normalized values; Based on the normalization results, according to the first... i The current observation value of the nth continuous variable, the nth i The corresponding continuous variable is the central state value of the water network operation design domain and the first continuous variable is the central state value of the water network operation design domain. i The half-interval width of the water network operation design domain corresponding to the continuous variables is determined. i The state sub-indicators corresponding to continuous variables.

[0016] The beneficial effect of the above further scheme is: by based on the first Dimensionless normalization and state sub-indices are constructed for the lower and upper bounds, central state values, and half-interval widths of the operation design domain corresponding to each continuous variable. This eliminates the influence of differences in dimensions, value ranges, and physical meanings among different continuous variables on the state evaluation results. It enables heterogeneous data such as channel flow, water level, water level change rate, actuator control quantities, and operating environment variables to participate in the calculation of operation design domain state indicators on a unified scale. At the same time, by using the central state value and half-interval width of the operation design domain to characterize the degree of variable deviation, it can continuously reflect the trend of the current observed value deviating from the normal operation zone to the warning zone or danger zone, thereby improving the sensitivity of the state indicators to the contraction of the operation boundary and the early warning capability.

[0017] Furthermore, the state sub-indicators include: The water level status sub-index is used to characterize the degree of deviation of the water level in key channels from the center of the normal operating area; The rate of change of water level is a state sub-index used to characterize the stability of the evolution trend of channel water level; Sensor status sub-indices are used to characterize the reliability of measurement information; Actuator status sub-indices are used to characterize the gate actuator's ability to track control commands; The operating environment status sub-index is used to characterize the degree of deviation between the upstream boundary inflow and the downstream receiving conditions relative to normal operating conditions.

[0018] The beneficial effects of the above-mentioned further scheme are as follows: by constructing sub-indices for water level status, water level change rate, sensor status, actuator status, and operating environment status, the applicable status of water network operation can be decomposed and evaluated from five aspects: hydraulic status, evolution trend, perception reliability, execution capability, and external operating environment. This avoids the problem that a single water level or flow threshold cannot reflect equipment anomalies, measurement anomalies, and changes in upstream and downstream receiving conditions. At the same time, each status sub-indicator serves as the input to the subsequent weighting function and the status index of the operation design domain, enabling different types of anomalies to participate in the comprehensive judgment in a quantifiable manner, thereby improving the completeness of the hierarchical identification results and the pertinence of control strategy switching.

[0019] Furthermore, during the sensor status assessment process, the actual sensor values ​​are compared with the Kalman filter predicted observations. Based on the comparison results, the sensor observation residuals are used to construct a normalized statistic to build a sensor status sub-index. When the observation residuals do not exceed the threshold, the sensor status is normal. If the observation residuals exceed the threshold, the sensor is abnormally warned, and the Kalman filter predicted observations are switched. During actuator status assessment, the actuator tracking error between the target opening and the actuator feedback opening is obtained to construct an execution status sub-indicator. When the actuator tracking error exceeds a preset threshold, an actuator fault alarm is triggered, and a backup actuator is switched to limit the scheduling amplitude. When the actuator tracking error does not exceed the preset threshold, the actuator status is normal. When the main actuator fails and the backup actuator is unavailable, the trigger value for actuator fault and backup unavailability is set to [value missing]. and make the red line trigger function value The trigger state is determined as an emergency operation state; when the actuator tracking error exceeds the preset threshold but the standby actuator can still be deployed, the red line is not triggered, and the actuator status sub-index is reduced. .

[0020] The beneficial effects of the above-mentioned further solutions are as follows: By comparing the actual sensor values ​​with the predicted observation values ​​obtained through Kalman filtering and constructing sensor state sub-indices based on the observation residuals, the measurement reliability evaluation results can be reduced in a timely manner when the sensor drifts, distorts, or becomes abnormal, thus avoiding abnormal measurement values ​​from directly affecting the determination of the water network operation status; By constructing actuator state sub-indices based on the tracking error between the target opening degree or target output and the actuator feedback value, problems such as jamming, hysteresis, saturation, or tracking mismatch of gate, pumping station, or valve actuators can be identified, so that both sensing anomalies and execution anomalies can be reflected in the state indicators of the operation design domain, thereby improving the reliability of fault identification and control switching in autonomous operation scenarios.

[0021] Furthermore, the expression for the water network operation design domain state index is as follows:

[0022]

[0023] in, Indicates time Water network operation design domain status indicators , , , and These represent the weighting coefficients corresponding to the water level status, water level change rate status, sensor status, actuator status, and operating environment status, respectively. , , , and Representing time respectively The sub-indicators include water level status, water level change rate status, sensor status, actuator status, and operating environment status. This represents the logical penalty term corresponding to the logical state constraint. , , , and These represent the health status of key sensors, the availability status of key actuators, the availability status of the main communication link, the downstream water supply conditions, and the control interlock release status, respectively. The value is set to [value] when the corresponding logical state is satisfied. If the condition is not met, the value is taken as follows: ; , , , and These represent the penalty coefficients corresponding to each logical state constraint. m The index representing the logical state constraint. Indicates time Key sensor health status indicators Indicates time The key actuators have available status indicators. Indicates time The main communication link availability status indicator. Indicates time The downstream water receiving conditions meet the status indication quantity. Indicates time The control interlock release status indication quantity;

[0024]

[0025]

[0026] in, Indicates the first State-like sub-indicators at time The weighting coefficients, This represents a set of status sub-indicators, including water level status, water level change rate status, sensor status, actuator status, and operating environment status. Indicates the first The base weights of the class state sub-indicators Indicates the first The dynamic correction coefficient of the state sub-indicator, Indicates the first Class-state sub-indicators, This indicates the topological importance of the node or control section where the state sub-index is located. This indicates the sensitivity of abnormal states to propagation along upstream and downstream. , and All represent preset adjustment coefficients, and each weighting coefficient satisfies ;

[0027]

[0028]

[0029]

[0030]

[0031]

[0032]

[0033] in, Indicates time Water level status sub-indicators Indicates time The rate of change of water level is a state sub-indicator. Indicates time Sensor status sub-indicators, Indicates time The actuator status sub-indices Indicates time Operating environment status sub-indicators Indicates time The controlled unit associated water level, Indicates the center water level of the normal operating area. Indicates the water level attenuation coefficient. Indicates time rate of water level change Indicates the central value of the target rate of change. This represents the rate of water level change attenuation coefficient. Indicates time The normalized statistic of the observation residuals This represents the sensor reliability adjustment coefficient. Indicates time Innovation covariance, Indicates time Sensor observation residuals Indicates time Actuator tracking error, This indicates the allowable tracking error reference value. Indicates the error amplification index. This indicates the target opening or target output given by the control algorithm. This represents the actuator feedback value. Indicates time Inflow at the upstream boundary, This represents the inflow center value corresponding to the normal operating range. Indicates the inflow state attenuation coefficient. Indicates time The downstream acceptance capacity coefficient This represents the central value of downstream receiving capacity corresponding to the normal operating range. This represents the state decay coefficient of the accepting capacity.

[0034] The beneficial effect of the above further scheme is that it improves the water level status sub-index. Water level change rate state sub-index Sensor status sub-indicators Actuator status sub-indicators and operating environment status sub-indicators By weighting and integrating the logical penalties corresponding to the logical state constraints, water level deviation, water level change trend, measurement reliability, actuator tracking capability, upstream inflow disturbance, and downstream receiving capacity changes can be uniformly mapped to the same operational design domain state indicators. Therefore, the current operating status of the water network no longer relies on a single threshold for judgment, but can reflect the degree of multi-source disturbance coupling and operating boundary contraction in a continuous and calculable manner. This provides a quantitative basis for the subsequent classification and judgment of stable operation, degraded operation and emergency operation, and improves the completeness of status identification and the rationality of control switching.

[0035] Furthermore, S5 includes: Define the red line trigger function; Under the condition that the red line is not triggered, the current water network operation status is classified and determined based on the comparison results between the water network operation design domain status index and the preset interval boundary. When the water network operation design domain status index is greater than or equal to the first threshold, the current water network operation status is determined to be stable operation. When the water network operation design domain status index is less than the first threshold but greater than or equal to the second threshold, the current water network operation status is determined to be degraded operation. When the water network operation design domain status index is less than the second threshold, the current water network operation status is determined to be emergency operation. Among them, the first threshold is greater than the second threshold. If the current water network operation status is determined to be stable, the target operation strategy shall be executed according to the predetermined target water level and target flow. When the current water network operation status is determined to be downgraded, the adjustment range is reduced and the rate of water level change is limited. Priority is given to ensuring the safety of key channels, critical equipment and end-point water receiving units, and a restricted operation strategy is implemented. When the current water network operation status is determined to be an emergency operation, a safety bottom-line strategy is implemented to prioritize preventing the water level of the controlled unit from entering the danger zone and to suppress the risk propagation caused by sudden changes in boundary flow.

[0036] The beneficial effect of the above-mentioned further scheme is that by combining the red line trigger determination with the interval-level determination based on the water network operation design domain state index, it is possible to determine the water network operation design domain state index under normal deviation operating conditions. The comparison results with the first and second thresholds are used to continuously classify and identify stable operation, degraded operation, and emergency operation. Simultaneously, when high-risk events occur, such as critical water levels entering danger zones, sudden changes in boundary flow, sensor failure with unavailable alternatives, or actuator malfunction with unavailable backups, the system can bypass the conventional interval comparison process and directly enter emergency operation mode. Therefore, this invention avoids the problem of insufficient identification of progressive risks by a single threshold determination, while ensuring rapid switching of control strategies under extreme risk events, improving the timeliness of risk response in water network operation and the reliability of safety-based control.

[0037] Furthermore, the expression for the red line trigger function is as follows:

[0038] in, Indicates time The red line trigger function value, Indicates the triggering amount for the critical water level danger zone. This indicates a sudden increase in inflow at the boundary. This indicates the trigger quantity caused by a sharp drop in downstream acceptance capacity. This indicates that the sensor has failed and an unavailable replacement trigger value has been found. This indicates that the actuator is faulty and the backup trigger is unavailable.

[0039] The beneficial effect of the above-mentioned further scheme is that it can trigger the critical water level danger zone. Boundary inflow surge trigger amount Downstream acceptance capacity suddenly drops, triggering volume Sensor failure and unavailable replacement trigger quantity And actuator failure and unavailable backup trigger quantity Unified inclusion in the red line trigger function The invention employs a maximum value function to determine whether the red line is triggered, ensuring that any high-risk event directly activates the emergency operation decision, preventing the weakening of the danger by averaging multiple risk events. Therefore, this invention can differentiate between deviations from normal conditions and high-risk red line events, rapidly initiating safety backup control in cases of critical water level breaches, sudden changes in boundary inflow, a sharp drop in downstream receiving capacity, or failure of critical equipment, thus improving the agility of risk identification and the certainty of safety response under extreme conditions.

[0040] This invention also provides a real-time calculation and hierarchical control system for water network operation design domain state indicators. The first processing module is used to acquire multidimensional observation data characterizing the current operating status of the water network; the multidimensional observation data includes continuous variable data and logical state data. The second processing module is used to establish a set of constraints for the water network operation design domain based on multi-dimensional observation data, the water network topology, the upstream and downstream transmission and distribution relationship, and the downstream water constraints. The third processing module is used to determine the upper and lower bounds of the operation design domain corresponding to each continuous variable by using the set of constraints of the water network operation design domain, and to perform dimensionless normalization processing on the continuous variable data to construct the state sub-indicators corresponding to each continuous variable. The fourth processing module is used to construct a weight function based on the constructed state sub-indicators and the coupling relationship between multiple dimension indicators, and to calculate the state indicators of the water network operation design domain in combination with logical state constraints. The fifth processing module is used to classify the current water network operation status into stable operation, degraded operation, or emergency operation based on the comparison results between the water network operation design domain status indicators and the preset interval boundaries, and to execute the corresponding control strategies.

[0041] The beneficial effects of this invention are as follows: Addressing the difficulties in safety assessment caused by multi-source disturbance coupling, time-varying equipment states, and dynamic contraction of operating boundaries during water network operation, this invention uses water level status, rate of change, perception reliability, execution capability, and operating environment as unified evaluation objects to construct a real-time operating condition-oriented operation design domain state index. Combined with a red-line triggering mechanism, it achieves collaborative discrimination between routine deviations and high-risk events. Based on this, the water network operating conditions are classified and identified according to the interval of the state index, and corresponding control strategies are matched to achieve an orderly switch from normal regulation to restricted regulation and safety-based control. This method can improve the completeness of state cognition, the timeliness of risk response, and the rationality of control switching in autonomous water network operation scenarios without relying on a single static threshold judgment, thereby enhancing the safety assurance capability and engineering applicability of water network operation under complex conditions. Attached Figure Description

[0042] Figure 1 A schematic diagram of the overall process of the present invention is shown.

[0043] Figure 2 A schematic diagram showing the structure and monitoring location of the dual-channel gate-controlled water conveyance unit is presented.

[0044] Figure 3 The diagram illustrates the hierarchical control and mode switching between stable operation, degraded operation, and emergency operation.

[0045] Figure 4 This diagram illustrates the coupling relationship between the operational design domain state indicators and disturbances, faults, and the operating environment.

[0046] Figure 5 A schematic diagram of the sensor state assessment and actuator tracking error handling based on Kalman filter observation residuals is shown.

[0047] Figure 6 The diagram shows the results of the dual-channel water level response and the switching of the three-stage control mode during simulation verification.

[0048] Figure 7 A schematic diagram illustrating the coupling relationship between the running design domain state indicators and inflow disturbances and fault events during simulation verification is shown.

[0049] Figure 8 This is a schematic diagram of the system structure of the present invention. Detailed Implementation

[0050] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0051] Example 1 like Figure 1 As shown, this invention provides a method for real-time calculation and hierarchical control of water network operation design domain state indicators. This method is applied to a water network autonomous operation monitoring system, where the water network consists of at least some of the following: canal sections, pipe sections, gates, pumping stations, valves, and storage units. The implementation method is as follows: S1. Obtain multidimensional observation data characterizing the current operating status of the water network; among which, multidimensional observation data includes continuous variable data and logical state data; Continuous variable data includes one or more of the following: channel flow, water level associated with the controlled unit, rate of water level change, gate opening, actuator control quantity corresponding to pump station output or valve opening, actuator change rate, water level or remaining storage capacity of downstream receiving pool, rainfall intensity, equipment health, and link latency. Logical status data includes one or more of the following: health status of key sensors, availability status of key actuators, downstream water supply conditions met, availability status of main communication link, and control interlock status. S2. Based on multi-dimensional observation data, and according to the water network topology, upstream and downstream transmission and distribution relationship, and downstream water constraints, establish a water network operation design domain constraint set; wherein, the water network operation design domain constraint set is used in S3 to determine the upper and lower bounds of the operation design domain corresponding to each continuous variable, and is used in S4 together with the state sub-indices as inputs for weight function construction and logical state constraint fusion, so as to calculate the state index of the water network operation design domain; S3. Using the constraint set of the water network operation design domain, determine the upper and lower bounds of the operation design domain corresponding to each continuous variable, and perform dimensionless normalization on the continuous variable data to construct the state sub-indices corresponding to each continuous variable. These state sub-indices include the controlled unit associated water level state sub-indice, water level change rate state sub-indice, sensor state sub-indice, actuator state sub-indice, and operating environment state sub-indice. The state sub-indices are used in S4 to construct the weighting function and serve as inputs for calculating the state indicators of the water network operation design domain. The implementation method is as follows: Using the set of constraints in the water network operation design domain, determine the upper and lower bounds of the operation design domain for each continuous variable; Based on the i The current observation of the nth continuous variable and the nth i The continuous variables correspond to the lower bound of the water network operation design domain and the first continuous variable. i The upper bound of the water network operation design domain is corresponding to each continuous variable, and the dimensionless normalized value is obtained. Normalization is performed based on dimensionless normalized values; Based on the normalization results, according to the first... i The current observation value of the nth continuous variable, the nth i The corresponding continuous variable is the central state value of the water network operation design domain and the first continuous variable is the central state value of the water network operation design domain. i The half-interval width of the water network operation design domain corresponding to the continuous variables is determined. i State sub-indicators corresponding to continuous variables; Status sub-indicators include: The water level status sub-index is used to characterize the degree of deviation of the water level in key channels from the center of the normal operating area; The rate of change of water level is a state sub-index used to characterize the stability of the evolution trend of channel water level; Sensor status sub-indices are used to characterize the reliability of measurement information; Actuator status sub-indices are used to characterize the gate actuator's ability to track control commands; Operating environment status sub-indicators are used to characterize the degree of deviation between the upstream boundary inflow and the downstream receiving conditions relative to normal operating conditions; During the sensor status assessment process, the actual sensor values ​​are compared with the Kalman filter predicted observations. Based on the comparison results, the sensor observation residuals are used to construct a normalized statistic to build a sensor status sub-index. When the observation residuals do not exceed the threshold, the sensor status is normal. If the observation residuals exceed the threshold, the sensor is abnormally warned, and the Kalman filter predicted observations are switched. During actuator status assessment, the actuator tracking error between the target opening and the actuator feedback opening is obtained to construct an execution status sub-indicator. When the actuator tracking error exceeds a preset threshold, an actuator fault alarm is triggered, and a backup actuator is switched to limit the scheduling amplitude. When the actuator tracking error does not exceed the preset threshold, the actuator status is normal. When the main actuator fails and the backup actuator is unavailable, the trigger value for actuator fault and backup unavailability is set to [value missing]. and make the red line trigger function value Retrieve the trigger state; at this time, do not enter the state index based on the water network operation design domain. Instead of using a tiered threshold determination process, it directly determines the system to be in an emergency operating state; when the actuator tracking error exceeds a preset threshold but the standby actuator is still available, the red line is not triggered, and the actuator status sub-index is lowered. and the actuator status sub-index As a domain status indicator for water network operation design The input participates in the subsequent classification determination; S4. Based on the constructed state sub-indicators, a weight function is constructed according to the coupling relationship between multiple dimension indicators, and the state indicators of the water network operation design domain are calculated in combination with logical state constraints. The water network operation design domain status indicators are obtained by weighted fusion of the water level status associated with the controlled unit, the water level change rate status, the sensor status, and the actuator status. Among them, the water level status associated with the controlled unit is determined based on the deviation between the current associated water level of the controlled unit and the center water level of the normal operation zone, as well as the water level status attenuation coefficient; the water level change rate status is determined based on the deviation between the current water level change rate and the center value of the target change rate, as well as the rate status attenuation coefficient; the sensor status is determined based on the normalized statistics of the observation residuals based on Kalman filtering and the sensor reliability adjustment coefficient; and the actuator status is determined based on the tracking error between the control command and the actuator feedback, the allowable tracking error reference value, and the error amplification index. The weight coefficients corresponding to each state are determined by a weight function, which is constructed based on the sub-indicators of each state, the topological hierarchy of the nodes where continuous variables are located, the scheduling effect of the control section, the sensitivity of abnormal states to propagate upstream and downstream, and the degree of impact on water supply continuity. The weight coefficients are determined by offline Monte Carlo simulation and satisfy normalization constraints. Logical state constraints are represented by logical penalties, which are determined based on the output values ​​of the logical constraint functions and their penalty coefficients corresponding to the health status of key sensors, the availability status of key actuators, the availability status of the main communication link, the downstream water supply conditions, and the control lockout status. The state indicators of the water network operation design domain are constructed based on the sub-indicators of each state, the weight coefficients output by the weighting function, and the logical penalties. S5. Based on the comparison results between the water network operation design domain state indicators and the preset interval boundaries, the current water network operation state is divided into stable operation, degraded operation, or emergency operation, and the corresponding control strategies are executed. The implementation method is as follows: Define the red line trigger function; Under the condition that the red line is not triggered, the current water network operation status is classified and determined based on the comparison results between the water network operation design domain status index and the preset interval boundary. When the water network operation design domain status index is greater than or equal to the first threshold, the current water network operation status is determined to be stable operation. When the water network operation design domain status index is less than the first threshold but greater than or equal to the second threshold, the current water network operation status is determined to be degraded operation. When the water network operation design domain status index is less than the second threshold, the current water network operation status is determined to be emergency operation. Among them, the first threshold is greater than the second threshold. If the current water network operation status is determined to be stable, the target operation strategy shall be executed according to the predetermined target water level and target flow. When the current water network operation status is determined to be downgraded, the adjustment range is reduced and the rate of water level change is limited. Priority is given to ensuring the safety of key channels, critical equipment and end-point water receiving units, and a restricted operation strategy is implemented. When the current water network operation status is determined to be an emergency operation, a safety bottom-line strategy is implemented to prioritize preventing the water level of the controlled unit from entering the danger zone and to suppress the risk propagation caused by sudden changes in boundary flow.

[0052] This invention is executed repeatedly according to a preset sampling period, and in each sampling period, the operational design domain status indicators and their corresponding operational status judgment results are updated in real time based on the latest observation data.

[0053] In this embodiment, the expression for the constraint set of the water network operation design domain is as follows:

[0054] in, Indicates time Water network operation design domain constraint set, Represents a continuous variable state vector. Indicates the first A continuous variable, and They represent the first Lower and upper bounds of the operational design domain for a continuous variable. n This represents the total number of continuous variables included in the set of constraints of the operational design domain. Represents the set of edges in the water network topology. Represents a node With nodes The transmission flow between them This indicates the transmission and distribution relationship determined by upstream and downstream water levels and actuator control parameters. Indicates the first A single actuator control quantity, r This represents the total number of actuator control variables included in the set of constraints of the operating design domain. , and These represent the lower limit, upper limit, and maximum allowable variation of the actuator control quantity, respectively. Indicates downstream acceptance capacity. This indicates downstream demand or constrained water supply. Indicates the first A number of logical state constraints are satisfied, where s represents the total number of logical state constraints included in the set of constraints of the running design domain.

[0055] In this embodiment, to comprehensively characterize the combined impact of the safety of the water level associated with the controlled unit, the water level evolution trend, the reliability of the sensors, and the tracking capability of the actuators on the applicability of the water network operation, the water network operation design domain state index is defined as follows:

[0056]

[0057] in, Indicates time Water network operation design domain status indicators , , , and These represent the weighting coefficients corresponding to the water level status, water level change rate status, sensor status, actuator status, and operating environment status, respectively. , , , and Representing time respectively The sub-indicators include water level status, water level change rate status, sensor status, actuator status, and operating environment status. This represents the logical penalty term corresponding to the logical state constraint. , , , and These represent the health status of key sensors, the availability status of key actuators, the availability status of the main communication link, the downstream water supply conditions, and the control interlock release status, respectively. The value is set to [value] when the corresponding logical state is satisfied. If the condition is not met, the value is taken as follows: ; , , , and These represent the penalty coefficients corresponding to each logical state constraint. m The index representing the logical state constraint. Indicates time Key sensor health status indicators Indicates time The key actuators have available status indicators. Indicates time The main communication link availability status indicator. Indicates time The downstream water receiving conditions meet the status indication quantity. Indicates time The control interlock release status indication quantity.

[0058]

[0059]

[0060]

[0061] in, Indicates the first State-like sub-indicators at time The weighting coefficients, This represents a set of status sub-indicators, including water level status, water level change rate status, sensor status, actuator status, and operating environment status. Indicates the first The base weights of the class state sub-indicators Indicates the first The dynamic correction coefficient of the state sub-indicator, Indicates the first Class-state sub-indicators, This indicates the topological importance of the node or control section where the state sub-index is located. This indicates the sensitivity of abnormal states to propagation along upstream and downstream. , and All represent preset adjustment coefficients, and each weighting coefficient satisfies .

[0062] Water network operation design domain status indicators The larger the value, the closer the current water network operation is to a safe and stable condition.

[0063] To reflect the deviation of the associated water level of the controlled unit from the normal operation center, the water level state function based on the center of the normal zone is defined as follows:

[0064] Water level status sub-indicators Water level associated with the controlled unit Deviating from the center water level of the normal operating area The degree of decrease is increased; when the critical water level enters the danger zone, the evaluation is no longer based solely on this state quantity, but rather the emergency operation strategy is directly triggered.

[0065] To reflect the impact of the rate of water level change on the safety of water network operation, the state function for the rate of water level change is defined as:

[0066] Water level change rate state sub-index It is used to characterize the degree to which the current rate of water level change deviates from the target center value.

[0067] To reflect the degree of agreement between sensor observations and state estimates, a sensor state based on Kalman filter observation residuals is defined:

[0068] in, Indicates time Sensor observation residuals Indicates time The actual measured value of the sensor, Indicates time The Kalman filter predicts the observations. This state function is used to evaluate the accuracy of the measurement information; when the main sensor malfunctions, the system issues an alarm to the staff and prioritizes the Kalman filter estimate or switches to the backup sensor to maintain the continuity of the state assessment.

[0069] To reflect the actuator's ability to track control commands, the actuator state function is defined as follows:

[0070] To reflect the degree of agreement between sensor observations and state estimates, the sensor state function based on the Kalman filter observation residuals is defined as follows:

[0071] To reflect the actuator's ability to track control commands, the actuator tracking error is defined as:

[0072] To reflect the actuator's ability to track control commands, the actuator state function is defined as follows:

[0073] To reflect the compression effect of the operating environment on the boundary of the water network operation design domain, the operating environment state function is defined as follows:

[0074] in, Indicates time Water level status sub-indicators Indicates time The rate of change of water level is a state sub-indicator. Indicates time Sensor status sub-indicators, Indicates time The actuator status sub-indices Indicates time Operating environment status sub-indicators Indicates time The controlled unit associated water level, Indicates the center water level of the normal operating area. Indicates the water level attenuation coefficient. Indicates time rate of water level change Indicates the central value of the target rate of change. This represents the rate of water level change attenuation coefficient. Indicates time The normalized statistic of the observation residuals This represents the sensor reliability adjustment coefficient. Indicates time Innovation covariance, Indicates time Sensor observation residuals Indicates time Actuator tracking error, This indicates the allowable tracking error reference value. Indicates the error amplification index. This indicates the target opening or target output given by the control algorithm. This represents the actuator feedback value. Indicates time Inflow at the upstream boundary, This represents the inflow center value corresponding to the normal operating range. Indicates the inflow state attenuation coefficient. Indicates time The downstream acceptance capacity coefficient This represents the central value of downstream receiving capacity corresponding to the normal operating range. This represents the state decay coefficient of the accepting capacity.

[0075] To isolate critical risk events such as dangerous water levels, sudden changes in boundary flow, and equipment failures from routine comprehensive evaluations, the redline trigger function is defined as follows:

[0076] in, Indicates time The red line trigger function value, Indicates the triggering amount of the critical water level danger zone. This indicates the trigger amount for a sudden increase or decrease in the boundary inflow. This indicates the trigger quantity caused by a sharp drop in downstream acceptance capacity. This indicates that the sensor has failed and an unavailable replacement trigger value has been found. This indicates that the actuator is faulty and the backup trigger is unavailable.

[0077] If the red line is not triggered, the current operating status is determined based on the relationship between the water network operation design domain status indicators and the classification thresholds, and is expressed as follows:

[0078] in, Indicates time The running status, Indicates a stable operating state. Indicates a downgraded operating status. Indicates an emergency operation status. This represents the threshold for classifying the state between stable operation and degraded operation. This represents the threshold for classifying the state between degraded and emergency operation, and satisfies... .

[0079] This judgment rule ensures that when there are no red-line events, graded control is implemented based on comprehensive status indicators, and when there are red-line events, emergency control is directly initiated.

[0080]

[0081] The aforementioned weighting coefficients satisfy the normalization constraint.

[0082] After determining the current operating state range, when the operating state S(k) corresponds to a stable operating state, the control system performs routine optimized scheduling according to the predetermined target water level and target flow rate; when the operating state S(k) corresponds to a degraded operating state, the control system reduces the adjustment amplitude and limits the rate of water level change, while prioritizing the safety of key channels, critical equipment, and end-point water receiving units; when the operating state S(k) corresponds to an emergency operating state, the control system immediately implements a safety bottom-line strategy, prioritizing the prevention of the water level associated with the controlled unit from entering the danger zone and suppressing the risk propagation caused by sudden changes in boundary flow.

[0083] The present invention will be further described below.

[0084] To address the shortcomings of existing technologies, this invention integrates continuous variable state quantization, logical constraint identification, and hierarchical control strategy switching. This solves the technical problems of existing technologies, such as the difficulty in uniformly quantifying and representing multidimensional heterogeneous variables, the difficulty in identifying boundary contraction trends in advance, and the difficulty in balancing control accuracy and system safety. Specifically: Acquire multidimensional observational data characterizing the current operational status of the water network, including continuous variable data and logical state data. Specifically, take a dual-channel gate-controlled water conveyance unit consisting of the upstream inflow boundary, upstream channel B1, primary control gate G1, downstream channel B2, and secondary control gate G2 as the example. Figure 2 As shown, the upstream inflow boundary provides water to channel B1. G1 regulates the flow capacity from channel B1 to channel B2, and G2 regulates the outflow capacity from channel B2 downstream. The water level in channel B1 is the primary control object of G1, and the water level in channel B2 is the primary control object of G2, thus forming a hydraulic regulation chain that is progressively transmitted along the water conveyance direction. Since the upstream inflow process, the gate flow capacity, and the downstream receiving conditions all affect the dynamic evolution of the channel water level, the dual-channel gate-controlled water conveyance unit is a controlled object with obvious upstream and downstream coupling characteristics. In this embodiment, the continuous variable data collected by the monitoring system includes upstream boundary inflow, B1 channel water level, B1 channel water level change rate, B2 channel water level, B2 channel water level change rate, G1 gate opening, G1 gate opening change rate, G2 gate opening, G2 gate opening change rate, and downstream receiving capacity coefficient; the collected logical status data includes one or more of the following: key sensor health status, key actuator availability status, downstream water receiving conditions met status, main communication link availability status, and control interlock status.

[0085] Based on the water network topology, upstream and downstream transmission and distribution relationships, water level constraints associated with key controlled units, gate and pump linkage constraints, and downstream water receiving capacity constraints, a set of constraints for the water network operation design domain is established. Specifically, based on design data, scheduling procedures, and historical safe operating conditions, the upper and lower limits of the associated water levels of each controlled unit, the allowable range of water level change rates, the gate opening or pump station output range, the maximum change rate of actuators, the upstream and downstream transmission and distribution flow relationships, and downstream receiving capacity constraints are determined, forming a set of constraints for the operation design domain composed of water level constraints, water level change rate constraints, actuator constraints, upstream and downstream transmission and distribution constraints, and operating environment constraints. Among them, the upstream and downstream transmission and distribution constraints are used to characterize the transmission relationship between upstream boundary inflow, gate flow capacity, and downstream channel water level; the operating environment constraints are used to characterize the compression effect of upstream inflow disturbances and downstream receiving capacity changes on the applicable boundary of the current operating conditions.

[0086] Based on the boundary conditions of the operational design domain corresponding to each continuous variable, the continuous variable data are subjected to dimensionless normalization, and state sub-indices corresponding to each continuous variable are constructed. Specifically, to uniformly quantify the applicable operational states of the dual-channel gate-controlled water conveyance unit, sub-indices for water level status, water level change rate status, sensor status, actuator status, and operating environment status are constructed. The water level status sub-indice is used to characterize the degree of deviation of the key channel water level from the center of the normal operating range; the water level change rate status sub-indice is used to characterize the stability of the channel water level evolution trend; the sensor status sub-indice is used to characterize the reliability of measurement information; the actuator status sub-indice is used to characterize the gate actuator's ability to track control commands; and the operating environment status sub-indice is used to characterize the degree of deviation between the upstream boundary inflow and the downstream receiving conditions from the normal operating conditions.

[0087] To comprehensively characterize the combined impact of the aforementioned state variables on the operational suitability of the dual-channel gate-controlled water conveyance unit, a time interval is defined. k The water network operation design domain state indicators are obtained by weighted fusion of water level status, water level change rate status, sensor status, actuator status, and operating environment status. Each weight coefficient is determined using offline Monte Carlo analysis and satisfies normalization constraints. A larger value for the water network operation design domain state indicator indicates that the current operating condition is closer to the safe and stable operating zone; a smaller value indicates that the current operating condition deviates further from the normal applicable boundary.

[0088] In one specific implementation, the water level status sub-indicators for channels B1 and B2 are calculated based on the center of their respective normal operating zones. When the channel water level is near the center of the normal operating zone, the water level status sub-indicator maintains a high value; when the channel water level continuously deviates from the center of the normal operating zone and approaches the warning boundary, the water level status sub-indicator continuously decreases. For example... Figure 5 As shown, the water level change rate sub-index is calculated based on the deviation between the current channel water level change rate and the target change rate center value. For steady-state water conveyance conditions, the target change rate center value can be taken as a small range near zero, thus both rapid rises and falls in water level are judged as unfavorable operating trends.

[0089] In the sensor condition assessment process, real-time measurements are compared with Kalman filter-predicted observations, and normalized statistics are constructed based on the observation residuals to obtain sensor condition sub-indices. For example... Figure 5As shown, when the key water level sensor is in normal working condition, the observation residual is small, and the corresponding sensor status sub-indicator is high; when the sensor is abnormal, drifts, or the degree of distortion increases, the observation residual increases, and the corresponding sensor status sub-indicator decreases. In the event of a main sensor malfunction but alternative observations are still available, the system issues an alarm to the staff and prioritizes using Kalman filter estimates or backup sensor data to maintain the continuity of short-term status assessment.

[0090] During actuator status evaluation, the tracking error between the target opening degree output by the control algorithm and the actuator feedback opening degree is used as the main criterion, and actuator status sub-indicators are constructed accordingly. When both G1 and G2 can track the control command well, the actuator status sub-indicators maintain a high value; when the gate actuator experiences jamming, hysteresis, saturation, or tracking mismatch, the tracking error increases, and the actuator status sub-indicators decrease. In the case of a main actuator failure but a standby actuator available, the system issues an actuator failure alarm and switches the standby actuator to perform the corresponding control task; in the case of a main actuator failure and the standby actuator unavailable, the relevant red-line triggering quantity will be activated.

[0091] To reflect the compressive effect of the operating environment on the boundary of the water network operation design domain, this embodiment uses the upstream boundary inflow and downstream receiving capacity as the main input quantities for the operating environment status sub-indicator. The greater the deviation of the upstream boundary inflow from the center value of the normal operation zone, the smaller the operating environment status sub-indicator; the more significantly the downstream receiving capacity is lower than the center value of the normal operation zone, the smaller the operating environment status sub-indicator. By incorporating the operating environment status sub-indicator into the ODD(k) main function, the system can identify the contraction trend of the applicable operating boundary earlier when the upstream inflow increases, the downstream receiving conditions weaken, or both occur simultaneously, and provide a basis for subsequent graded control switching.

[0092] Based on the topological hierarchy of the nodes containing continuous variables and the scheduling role of control sections, a weighting function is constructed. This function integrates state sub-indicators, the weighting function, and logical constraints to calculate the state indicators of the water network operation design domain, as detailed below: In addition to routine comprehensive assessments, to isolate high-risk events such as dangerous water levels, sudden changes in boundary inflow, sharp drops in downstream receiving capacity, and failures of critical equipment from routine condition assessments, a redline trigger function Γ(k) is defined. For example... Figure 3 and Figure 5As shown, the red-line trigger function includes at least the trigger amounts for critical water level danger zones, sudden increases or decreases in boundary inflow, sudden drops in downstream receiving capacity, sensor failure and unavailable replacements, and actuator failure and unavailable backups. When the red-line trigger function value Γ(k) is in the triggered state, the system no longer relies solely on the conventional interval position of the water network operation design domain state index ODD(k) for judgment, but directly enters the emergency operation state and executes a safety bottom-line strategy.

[0093] Under the condition that the red line is not triggered, the operational design domain state index ODD(k) is compared with the first threshold T1 and the second threshold T2 to classify the current operational state. The operational state judgment rule ensures that graded control is implemented based on the comprehensive state index when there is no red line event, and directly enters emergency control when a red line event occurs, thus taking into account both the need for continuous evaluation of normal state and rapid response to extreme risks.

[0094] Based on the comparison results between the water network operation design domain status indicators and the preset interval boundaries, the operation status is divided into stable operation, degraded operation, or emergency operation, as detailed below: When the system is in a stable operating state, G1 executes control strategy 1 based on the water level set during normal operation of channel B1, and G2 executes control strategy 1 based on the water level set during normal operation of channel B2. For example... Figure 3 As shown, since both G1 and G2 are water intake or discharge actuators, their opening changes have a negative gain effect on the upstream water level. Therefore, the control error is defined as the set water level minus the current channel water level. The control objective under stable operation is to maintain the water levels of channels B1 and B2 around the normal operating set value with small fluctuations, and to avoid unnecessary large oscillations in the gate opening.

[0095] When the system is in degraded operation mode, G1 and G2 switch to control strategy 2 respectively, and adjust the target water levels of channels B1 and B2 to more conservative warning setpoints. At this time, the control objective changes from conventional fine tracking to suppressing the expansion of water level deviation in both channels, reducing the rate of water level change, and limiting the propagation of risk upstream and downstream. Since the inflow disturbance in channel B1 will continue to be transmitted to channel B2 via G1, the regulation of G1 and G2 in degraded operation mode is not independent of each other, but rather reflects the coordinated control of upstream storage and discharge capacity and downstream outflow capacity.

[0096] When the system is in emergency operation mode, the adjustment method based primarily on conventional continuous parameter tuning is discontinued, and control strategy 3 is adopted instead. Control strategy 3 coordinates the output of safety-support actions G1 and G2 based on the water levels in channels B1 and B2, the rate of water level change, abnormal boundary inflow, a sudden drop in downstream receiving capacity, and the red line triggering results. This prioritizes preventing the water levels in critical channels from entering the danger zone and suppressing the continued spread of risk along the water conveyance direction. The control objective in emergency operation mode is not to continue pursuing precise tracking of conventional setpoints, but rather to prioritize ensuring the safety of critical water levels in both channels and the overall operational stability of the system.

[0097] Execution of corresponding control strategies: In a continuous operation example, initially, the upstream boundary inflow is within the normal range, and the water levels of both channels B1 and B2 remain near the center of the normal operation zone, indicating a stable operating state. As the upstream inflow gradually increases, channel B1 first shows an upward trend, and this disturbance is then transmitted to channel B2 via G1, causing the water level in channel B2 to rise synchronously. At this time, the operating environment status sub-indicators decrease, which in turn leads to a decrease in the water network operation design domain status indicators. When the water network operation design domain status indicators fall below the first threshold but have not yet triggered the red line condition, the system enters a degraded operation state, and G1 and G2 switch to control strategy 2 respectively. When communication link anomalies, actuator jamming, or a sudden drop in downstream receiving capacity coincides with the superposition of water level deviations in both channels, the red line trigger function is activated, and the system directly enters an emergency operation state, executing control strategy 3. After the anomaly is resolved and the water levels of channels B1 and B2 exit the high-risk zone, the system returns to the degraded operation state and, after further recovery of operating conditions, returns to a stable operation state.

[0098] In this embodiment, a one-dimensional hydrodynamic model is used to construct the hydrodynamic simulation model of the dual-channel gate-controlled water conveyance unit. For example... Figure 6 As shown, the model calculation time test shows that the hydrodynamic simulation model takes approximately 0.1 hours to complete a 12-hour simulation and approximately 0.001 hours to calculate the indicators, meeting the real-time calculation requirements. Therefore, the method described in this invention is selected to construct the operational design domain state indicator system and hierarchical control model for real-time operating condition identification and control switching.

[0099] When constructing the above-mentioned operational design domain state index system, sub-index calculation models were independently built for the applicable operational states of channels B1 and B2. Each sub-index model used the full state data of the dual-channel gate-controlled water conveyance unit as input to predict the state sub-index value for the next calculation step of each channel. For example... Figure 4 As shown, the constructed sub-index calculation model comprehensively utilizes the exponential decay function mapping relationship, and the weight coefficients are determined using the offline Monte Carlo simulation method. The model is considered converged when the index calculation error on 20 consecutive validation sets no longer decreases.

[0100] To ensure the accuracy of the calculated state indices of the operational design domain, an L1 regularization coefficient of 0.01 was set during model construction. Simulation parameters were set as follows: the initial upstream flow was 22.5 m. 3 / s, with an amplitude of 4 m applied at the 40th minute 3 A step disturbance of / s lasted until the 80th minute, of which 22.5 m 3 / s is the inflow center value corresponding to the normal operating range, 4 m 3 / s represents the inflow disturbance amplitude; the center water level of the normal operating zone of Channel B1 is 2.50 m, the warning water level is 2.80 m, and the red line water level is 3.00 m; the center water level of the normal operating zone of Channel B2 is 2.00 m, the warning water level is 2.20 m, and the red line water level is 2.50 m; the first threshold T1 = 0.60, the second threshold T2 = 0.40; the weighting coefficients wh = 0.25, wv = 0.15, ws = 0.20, wa = 0.20, we = 0.20; the water level state attenuation coefficient λh = 1.0, the water level change rate attenuation coefficient λv = 1.5, the sensor attenuation coefficient λs = 2.0, the actuator attenuation coefficient λa = 2.0, the inflow state attenuation coefficient λq = 1.0, and the receiving capacity state attenuation coefficient λc = 1.0. For example... Figure 7 As shown, the calculation time for ODD status indicators was less than 0.001 hours in all test scenarios, meeting the requirements for real-time calculation.

[0101] During the condition assessment process, real-time measurements are compared with Kalman filter-predicted observations as a double-insurance mechanism. Normalized statistics are constructed based on the observation residuals, and sensor condition sub-indices are obtained accordingly. In the objective function, the water level condition attenuation coefficient λh for both channels B1 and B2 is set to 1.0, and the water level change rate condition attenuation coefficient λv is set to 1.5; the sensor reliability attenuation coefficient λs is set to 2.0, and the actuator condition attenuation coefficient λa is set to 2.0.

[0102] After completing the simulation model construction, it was applied to verify the step disturbance condition. The verification condition was that the initial upstream flow was 22.5 m³ / s, and a step disturbance with an amplitude of +4 m³ / s was applied at the 40th minute, lasting until the 80th minute. Figure 6 and Figure 7As shown, simulation results indicate that compared to the fixed-threshold binary judgment method, the real-time calculation and hierarchical control method for the operational design domain state index in this embodiment of the invention achieves the following results: In the initial stable operation phase, the average water level of B1 is 2.52 ± 0.04 m, the average water level of B2 is 2.01 ± 0.04 m, and the average ODD index is 0.95. After the disturbance is applied, the ODD index drops below the first threshold T1 at the 47th minute, and the system automatically switches to a degraded operation state and switches to control strategy 2, providing an early warning time of approximately 7 minutes. At the 55th minute, the B1 water level deviation widens, and the system gradually converges after adjustment by control strategy 2, with a peak water level of 2.87 m for B1 and 2.24 m for B2. After the disturbance is removed, the system returns to a stable operation state at the 90th minute. No red line triggers or safety red line violations occur throughout the entire process. This is mainly due to the operational design domain state index's ability to identify the boundary contraction trend in advance and proactively switch control modes during the inflow disturbance enhancement phase. This demonstrates that the real-time calculation and hierarchical control method for the operational design domain state index of this invention has better overall performance than the fixed threshold binary judgment method, thus proving the effectiveness of this invention.

[0103] The method for real-time calculation and hierarchical control of water network operation design domain state indicators in this embodiment can actively switch the operation state about 7 minutes before the inflow disturbance is applied, so that the control system can switch to restricted regulation before the critical water level approaches the warning boundary. This solves the technical problem in the prior art that it is difficult to uniformly quantify and represent multidimensional heterogeneous variables and support hierarchical control in a timely manner.

[0104] The real-time calculation and hierarchical control method for the operational design domain status index in this embodiment decouples the comprehensive evaluation of the water network operational design domain status index (ODD) from the red line triggering conditions. Under normal operating conditions, hierarchical control is implemented based on the comprehensive status index, and under extreme risk conditions, emergency strategies are directly triggered, thus achieving a balance between control accuracy and system safety.

[0105] The real-time calculation and hierarchical control method for the operational design domain state index in this embodiment effectively suppresses the propagation and expansion of water level deviation along the water conveyance direction by identifying the boundary contraction trend in advance during the inflow disturbance enhancement stage and switching to the warning set value. After the disturbance recedes, the control mode is driven to gradually recover to a stable operating state, which enhances the timeliness of risk response and the rationality of control switching in the autonomous operation scenario of the water network.

[0106] Example 2 like Figure 8 As shown, this invention provides a real-time calculation and hierarchical control system for water network operation design domain state indicators, used to execute the real-time calculation and hierarchical control method for water network operation design domain state indicators described in Example 1, comprising: The first processing module is used to acquire multidimensional observation data characterizing the current operating status of the water network; the multidimensional observation data includes continuous variable data and logical state data. The second processing module is used to establish a set of constraints for the water network operation design domain based on multi-dimensional observation data, the water network topology, the upstream and downstream transmission and distribution relationship, and the downstream water constraints. The third processing module is used to determine the upper and lower bounds of the operation design domain corresponding to each continuous variable by using the set of constraints of the water network operation design domain, and to perform dimensionless normalization processing on the continuous variable data to construct the state sub-indicators corresponding to each continuous variable. The fourth processing module is used to construct a weight function based on the constructed state sub-indicators and the coupling relationship between multiple dimension indicators, and to calculate the state indicators of the water network operation design domain in combination with logical state constraints. The fifth processing module is used to classify the current water network operation status into stable operation, degraded operation, or emergency operation based on the comparison results between the water network operation design domain status indicators and the preset interval boundaries, and to execute the corresponding control strategies.

[0107] In this embodiment, the functional units can be divided according to the real-time calculation and hierarchical control method of water network operation design domain state indicators. For example, each function can be divided into a separate functional unit, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the division of units in this invention is illustrative and only represents a logical division; other division methods may be used in actual implementation.

[0108] In this embodiment, the real-time calculation and hierarchical control system for the water network operation design domain state indicators, in order to achieve the principles and beneficial effects of Embodiment 1, includes hardware structures and / or software modules corresponding to the execution of various functions. Those skilled in the art should readily recognize that, in conjunction with the illustrative units and algorithm steps described in the embodiments disclosed herein, this invention can be implemented in hardware and / or a combination of hardware and computer software. Whether a function is executed by hardware or computer software depends on the specific application and design constraints of the technical solution. Different methods can be used to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

Claims

1. A method for real-time calculation and hierarchical control of water network operation design domain state indicators, characterized in that, Includes the following steps: S1. Obtain multidimensional observation data characterizing the current operating status of the water network; among which, multidimensional observation data includes continuous variable data and logical state data; S2. Based on multi-dimensional observation data, and according to the water network topology, upstream and downstream transmission and distribution relationships, and downstream water constraints, establish a set of constraints for the water network operation design domain. S3. Using the set of constraints of the water network operation design domain, determine the upper and lower bounds of the operation design domain corresponding to each continuous variable, and perform dimensionless normalization on the continuous variable data to construct the state sub-indicators corresponding to each continuous variable. S4. Based on the constructed state sub-indicators, a weight function is constructed according to the coupling relationship between multiple dimension indicators, and the state indicators of the water network operation design domain are calculated in combination with logical state constraints. S5. Based on the comparison results between the water network operation design domain status indicators and the preset interval boundaries, the current water network operation status is divided into stable operation, degraded operation, or emergency operation, and the corresponding control strategies are executed.

2. The method for real-time calculation and hierarchical control of water network operation design domain state indicators according to claim 1, characterized in that, The continuous variable data includes one or more of the following: channel flow, water level associated with the controlled unit, rate of water level change, gate opening, actuator control quantity corresponding to pump station output or valve opening, actuator change rate, water level or remaining storage capacity of downstream receiving pool, rainfall intensity, equipment health, and link delay. The logical status data includes one or more of the following: health status of key sensors, availability status of key actuators, downstream water supply conditions met, availability status of the main communication link, and control lockout status.

3. The method for real-time calculation and hierarchical control of water network operation design domain state indicators according to claim 1, characterized in that, The expression for the set of constraints in the water network operation design domain is as follows: in, Indicates time Water network operation design domain constraint set, Represents a continuous variable state vector. Indicates the first A continuous variable, and They represent the first Lower and upper bounds of the operational design domain for a continuous variable. n This represents the total number of continuous variables included in the set of constraints of the operational design domain. Represents the set of edges in the water network topology. Represents a node With nodes The transmission flow between them This indicates the transmission and distribution relationship determined by upstream and downstream water levels and actuator control parameters. Indicates the first A single actuator control quantity, r This represents the total number of actuator control variables included in the set of constraints of the operating design domain. , and These represent the lower limit, upper limit, and maximum allowable variation of the actuator control quantity, respectively. Indicates downstream acceptance capacity. This indicates downstream demand or constrained water supply. Indicates the first A number of logical state constraints are satisfied, where s represents the total number of logical state constraints included in the set of constraints of the running design domain.

4. The method for real-time calculation and hierarchical control of water network operation design domain state indicators according to claim 1, characterized in that, S3 includes: Using the set of constraints in the water network operation design domain, determine the upper and lower bounds of the operation design domain for each continuous variable; Based on the i The current observation of the nth continuous variable and the nth i The continuous variables correspond to the lower bound of the water network operation design domain and the first continuous variable. i The upper bound of the water network operation design domain is corresponding to each continuous variable, and the dimensionless normalized value is obtained. Normalization is performed based on dimensionless normalized values; Based on the normalization results, according to the first... i The current observation value of the nth continuous variable, the nth i The corresponding continuous variable is the central state value of the water network operation design domain and the first continuous variable is the central state value of the water network operation design domain. i The half-interval width of the water network operation design domain corresponding to the continuous variables is determined. i The state sub-indicators corresponding to continuous variables.

5. The method for real-time calculation and hierarchical control of water network operation design domain state indicators according to claim 4, characterized in that, The status sub-indicators include: The water level status sub-index is used to characterize the degree of deviation of the water level in key channels from the center of the normal operating area; The rate of change of water level is a state sub-index used to characterize the stability of the evolution trend of channel water level; Sensor status sub-indices are used to characterize the reliability of measurement information; Actuator status sub-indices are used to characterize the gate actuator's ability to track control commands; The operating environment status sub-index is used to characterize the degree of deviation between the upstream boundary inflow and the downstream receiving conditions relative to normal operating conditions.

6. The method for real-time calculation and hierarchical control of water network operation design domain state indicators according to claim 5, characterized in that, In the sensor condition assessment process, the actual sensor values ​​are compared with the Kalman filter predicted observations. Based on the comparison results, the sensor observation residuals are used to construct normalized statistics to build sensor condition sub-indices. If the observation residual does not exceed the threshold, the sensor is in normal condition. If the observation residual exceeds the threshold, the sensor will issue an abnormal warning and switch to Kalman filtering to predict the observation value. During actuator status assessment, the actuator tracking error between the target opening and the actuator feedback opening is obtained to construct an execution status sub-indicator. When the actuator tracking error exceeds a preset threshold, an actuator fault alarm is triggered, and a backup actuator is switched to limit the scheduling amplitude. When the actuator tracking error does not exceed the preset threshold, the actuator status is normal. When the main actuator fails and the backup actuator is unavailable, the trigger value for actuator fault and backup unavailability is set to [value missing]. and make the red line trigger function value The trigger state is determined as an emergency operation state; when the actuator tracking error exceeds the preset threshold but the standby actuator can still be deployed, the red line is not triggered, and the actuator status sub-index is reduced. .

7. The method for real-time calculation and hierarchical control of water network operation design domain state indicators according to claim 1, characterized in that, The expression for the water network operation design domain state index is as follows: in, Indicates time Water network operation design domain status indicators , , , and These represent the weighting coefficients corresponding to the water level status, water level change rate status, sensor status, actuator status, and operating environment status, respectively. , , , and Representing time respectively The sub-indicators include water level status, water level change rate status, sensor status, actuator status, and operating environment status. This represents the logical penalty term corresponding to the logical state constraint. , , , and These represent the health status of key sensors, the availability status of key actuators, the availability status of the main communication link, the downstream water supply conditions, and the control interlock release status, respectively. The value is set to [value] when the corresponding logical state is satisfied. If the condition is not met, the value is taken as follows: ; , , , and These represent the penalty coefficients corresponding to each logical state constraint. m Indicates the sequence number of the logical state constraint. Indicates time Key sensor health status indicators Indicates time The key actuators have available status indicators. Indicates time The main communication link availability status indicator. Indicates time The downstream water receiving conditions meet the status indication quantity. Indicates time The control interlock release status indication quantity; in, Indicates the first State-like sub-indicators at time The weighting coefficients, This represents a set of status sub-indicators, including water level status, water level change rate status, sensor status, actuator status, and operating environment status. Indicates the first The base weights of the class state sub-indicators Indicates the first The dynamic correction coefficient of the state sub-indicator, Indicates the first Class-state sub-indicators, This indicates the topological importance of the node or control section where the state sub-index is located. This indicates the sensitivity of abnormal states to propagation along upstream and downstream. , and All represent preset adjustment coefficients, and each weighting coefficient satisfies ; in, Indicates time Water level status sub-indicators Indicates time The rate of change of water level is a state sub-indicator. Indicates time Sensor status sub-indicators, Indicates time The actuator status sub-indices Indicates time Operating environment status sub-indicators Indicates time The controlled unit associated water level, Indicates the center water level of the normal operating area. Indicates the water level attenuation coefficient. Indicates time rate of water level change Indicates the central value of the target rate of change. This represents the rate of water level change attenuation coefficient. Indicates time The normalized statistic of the observation residuals This represents the sensor reliability adjustment coefficient. Indicates time Innovation covariance, Indicates time Sensor observation residuals Indicates time Actuator tracking error, This indicates the allowable tracking error reference value. Indicates the error amplification index. This indicates the target opening or target output given by the control algorithm. This represents the actuator feedback value. Indicates time Inflow at the upstream boundary, This represents the inflow center value corresponding to the normal operating range. Indicates the inflow state attenuation coefficient. Indicates time The downstream acceptance capacity coefficient This represents the central value of downstream receiving capacity corresponding to the normal operating range. This represents the state decay coefficient of the accepting capacity.

8. The method for real-time calculation and hierarchical control of water network operation design domain state indicators according to claim 1, characterized in that, S5 includes: Define the red line trigger function; Under the condition that the red line is not triggered, the current water network operation status is classified and determined based on the comparison results between the water network operation design domain status index and the preset interval boundary. When the water network operation design domain status index is greater than or equal to the first threshold, the current water network operation status is determined to be stable operation. When the water network operation design domain status index is less than the first threshold but greater than or equal to the second threshold, the current water network operation status is determined to be degraded operation. When the water network operation design domain status index is less than the second threshold, the current water network operation status is determined to be emergency operation. Among them, the first threshold is greater than the second threshold. If the current water network operation status is determined to be stable, the target operation strategy shall be executed according to the predetermined target water level and target flow. When the current water network operation status is determined to be downgraded, the adjustment range is reduced and the rate of water level change is limited. Priority is given to ensuring the safety of key channels, critical equipment and end-point water receiving units, and a restricted operation strategy is implemented. When the current water network operation status is determined to be an emergency operation, a safety bottom-line strategy is implemented to prioritize preventing the water level of the controlled unit from entering the danger zone and to suppress the risk propagation caused by sudden changes in boundary flow.

9. The method for real-time calculation and hierarchical control of water network operation design domain state indicators according to claim 8, characterized in that, The expression for the red line trigger function is as follows: in, Indicates time The red line trigger function value, Indicates the triggering amount of the critical water level danger zone. This indicates a sudden increase in inflow at the boundary. This indicates the trigger quantity caused by a sharp drop in downstream acceptance capacity. This indicates that the sensor has failed and an unavailable replacement trigger value has been found. This indicates that the actuator is faulty and the backup trigger is unavailable.

10. A real-time calculation and hierarchical control system for the state indicators of a water network operation design domain, used to execute the real-time calculation and hierarchical control method for the state indicators of a water network operation design domain as described in any one of claims 1-9, characterized in that, include: The first processing module is used to acquire multidimensional observation data characterizing the current operating status of the water network; the multidimensional observation data includes continuous variable data and logical state data. The second processing module is used to establish a set of constraints for the water network operation design domain based on multi-dimensional observation data, the water network topology, the upstream and downstream transmission and distribution relationship, and the downstream water constraints. The third processing module is used to determine the upper and lower bounds of the operation design domain corresponding to each continuous variable by using the set of constraints of the water network operation design domain, and to perform dimensionless normalization processing on the continuous variable data to construct the state sub-indicators corresponding to each continuous variable. The fourth processing module is used to construct a weight function based on the constructed state sub-indicators and the coupling relationship between multiple dimension indicators, and to calculate the state indicators of the water network operation design domain in combination with logical state constraints. The fifth processing module is used to classify the current water network operation status into stable operation, degraded operation, or emergency operation based on the comparison results between the water network operation design domain status indicators and the preset interval boundaries, and to execute the corresponding control strategies.

Images