Intelligent water-saving regulation and control method and system for water conservancy project

Abstract

The invention discloses an intelligent water-saving regulation and control method and system for a water conservancy project, and relates to the technical field of water conservancy regulation and control. An event ontology model is constructed, state parameters are initialized, basic environment data is collected, whether an event is triggered or not is judged, event-driven data pruning and reasoning chain selection are executed, and the event ontology model is established. The method comprises the steps of constructing a water demand prediction model, generating a thermal distribution diagram, executing a regulation and control instruction, and adaptively updating an event response model and a data path. Through structured event description, data pruning and an inference chain selection mechanism, an intelligent water-saving regulation and control process based on event driving is realized; the acquisition frequency can be dynamically adjusted, the water demand area can be accurately identified, and the thermal distribution diagram can be generated, so that the response accuracy, the self-adaptive capability and the water-saving regulation and control efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of water conservancy regulation and control, and in particular to an intelligent water-saving regulation and control method and system for a water conservancy project. Background Art

[0002] With the modernization of agriculture and the intensification of water resource shortages, the demand for intelligent and sophisticated water-saving control systems in water conservancy projects is increasing. Traditional irrigation control methods rely primarily on fixed-cycle data collection and preset rules for irrigation control. These methods lack the ability to dynamically perceive and respond to sudden meteorological changes, abnormal crop conditions, or hydrological fluctuations. This results in inefficient water resource allocation, lagging and generalized irrigation behavior, and can easily lead to localized over-irrigation or irrigation delays, which in turn affects crop yields and water resource utilization.

[0003] At present, the Chinese invention patent with application number CN202210270523.2 discloses a water-saving and salt-inhibiting irrigation and drainage coordinated regulation method and system. The method includes: obtaining a first irrigation and drainage pattern set; forming a first water footprint set, wherein the first water footprint set corresponds one-to-one to the first irrigation and drainage pattern set; determining a first crop irrigation and drainage pattern based on the first water footprint set; conducting a salt inhibition test on the first soil to obtain a first salt inhibition test result; determining a first soil covering scheme; obtaining a first yield set of the first crop, wherein the first yield set includes multiple test yields; using the first crop irrigation and drainage pattern as a preset mode, using the first soil covering scheme as a preset scheme, and combining the multiple test yields to perform coordinated regulation of irrigation and drainage.

[0004] Existing technologies make it difficult to proactively identify and respond to key irrigation events on demand, resulting in inaccurate regulation and waste of resources. Summary of the Invention

[0005] The technical problem solved by the present invention is that the above-mentioned technology is difficult to realize active identification and on-demand response to key irrigation events, resulting in inaccurate regulation and waste of resources.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: An intelligent water-saving control method for a water conservancy project comprises the following steps: Step S1, constructing an event ontology model and initializing state parameters; Step S2, collecting basic environmental data and determining whether an event is triggered; Step S3, performing event-driven data pruning and inference chain selection; Step S4, reconstructing the water demand prediction model and generating a thermal distribution map; Step S5, executing the control instruction; Step S6: adaptively update the event response model and data path.

[0007] Preferably, step S1 includes the following sub-steps: Step S101, defining typical water demand response events, wherein the typical water demand response events include meteorological events, crop status events, and hydrological fluctuation events; Meteorological events include the number of consecutive high-temperature days exceeding the set value, hourly precipitation exceeding the irrigation area infiltration limit, and wind speed continuously exceeding the critical threshold; The crop status events include abnormal crop spectral reflectance, increased leaf curling tendency and signs of growth stagnation; The hydrological fluctuation events include sudden changes in channel water inflow and fluctuations exceeding the threshold of flow velocity in the terminal channel system; Establish a structured description template for a typical water demand response event, which includes event number, event classification, event name, trigger logic, affected object type, control suggestion action, priority level, effective time range and response path number, and output the event ontology model; Step S102: Parse the structured description template of a typical water demand response event, identify the data fields of the structured description template, classify the sensing devices and sensing data sources, and output a standard collection channel directory. The sensing data sources include: Soil moisture sensing node: used to collect the moisture status of the surface and middle layers of soil; Crop status image acquisition node: used to collect crop canopy images, color index and morphological changes; Channel flow counting node: used to collect flow velocity and flow fluctuation information at all levels of the channel; Meteorological micro-station node: used to collect wind speed data, temperature data, humidity data, rainfall data and radiation data; Establish a mapping relationship between data fields and typical water demand response events.

[0008] Step S103: Initializing state parameters, which include the initial soil moisture value, crop type, growth stage, and historical irrigation record status corresponding to each plot.

[0009] Preferably, step S2 includes the following sub-steps: Step S201: Data is collected based on the perception data source, and current state data and current environmental data are output. The current state data includes: surface soil moisture ratio of the plot, mid-soil moisture ratio of the plot, crop canopy image, crop color index, crop morphology, real-time flow velocity and flow fluctuation trend of each level of channels. The current environmental data includes recorded wind speed data, temperature data, humidity data, rainfall data, and radiation data. Based on a preset low-risk data threshold, it is determined whether the current state data and current environmental data are within the low-risk data threshold. If so, the current state is determined to be a low-risk state, and a collection frequency reduction signal is output; Step S202: parse the trigger logic field in the data field corresponding to the typical water demand response event, extract the indicator value corresponding to the data field from the current state data and the current environment data, and determine whether the indicator value meets the trigger condition based on the priority level and the trigger logic field.

[0010] Preferably, the step S2 further includes the following sub-steps: Step S203: If there is a typical water demand response event with one or more indicator values ​​that meet the trigger conditions, the status of the current state data and the current environment data is set to an event occurrence, and the corresponding event number, event category, event name, current state data and current environment data are recorded. The inference chain number corresponding to the typical water demand response event is started, and a collection frequency increase signal is output; Step S204: If there is no typical water demand response event with one or more indicator values ​​meeting the trigger conditions, the status of the current state data and the current environmental data is set to no critical event, and enters a low-priority hold state. All computing nodes suspend high-level processing tasks and only retain status records.

[0011] Preferably, step S3 includes the following sub-steps: Step S301: Based on the currently triggered typical water demand response event, the corresponding data fields and control recommended actions are extracted. The minimum necessary data corresponding to the control recommended actions is filtered from the preset data processing path library and input into the control terminal. The perception data sources unrelated to the data fields are closed. This is defined as a data pruning operation. Step S302: selecting an inference chain that matches the event classification from a control module library, wherein the inference chain includes a data completion module, a state fusion module, a control target reorganization module, and a path simulation module connected in sequence, and the inference chain is executed in combination according to the event call rule; Step S303: activating the inference chain, and inputting the minimum necessary data into each module in the inference chain according to the priority level.

[0012] Preferably, step S4 includes the following sub-steps: Step S401: Obtain the current perception data in the inference chain and compare it with the corresponding data in the historical status database. If the current soil moisture is lower than the historical average or the decline rate of the crop image index is greater than the preset standard decline rate, the land parcel is marked as a potential water-demanding area and an abnormality attention list is output.

[0013] Preferably, the step S4 further includes the following sub-steps: Step S402: obtaining the current crop type, current crop growth stage, real-time soil moisture value, soil type, and meteorological forecast trend per unit time of the potential water-demanding area, outputting the data as potential water storage area state data, and inputting the potential water-demanding area state data into a preset state modeling model for analysis, wherein the analysis includes: if the current crop type is a shallow-rooted crop and the soil type is light soil, increasing the weight of the corresponding potential water-demanding area by a first percentage; if the meteorological forecast trend per unit time is no rainfall and continuous high temperature, increasing the water demand urgency; if the meteorological forecast trend per unit time is rain, the current crop growth stage is dormant, or the soil moisture value is higher than a preset soil moisture threshold, decreasing the water demand urgency; Output the dynamic water demand priority score of each plot as the basis for sorting and output a structured list; Step S403: Map the structured list to the irrigation area geographical layer and output a thermal distribution map.

[0014] Preferably, step S5 includes the following sub-steps: Step S501: Input the thermal distribution map into the control execution engine and generate zone control instructions according to the dynamic water demand priority score; In step S502, the edge controller opens and closes the irrigation valve and adjusts the flow rate of the pump station according to the instruction.

[0015] Preferably, step S6 includes the following sub-steps: Step S601: Evaluate the results of the control execution and output the evaluation results including the amount of soil moisture recovery and the recovery of crop status; Step S602: If the evaluation result deviates from the preset status standard, the result is fed back to the event response terminal to adjust the event determination threshold; Step S603: Synchronously update the data pruning strategy and the inference chain calling logic.

[0016] An intelligent water-saving control system for water conservancy projects, comprising an ontology model building module, a data acquisition and judgment module, a data pruning operation module, a model reconstruction and analysis module, a control instruction execution module, and a model response update module; The ontology model building module is used to build an event ontology model and initialize state parameters; The data collection and judgment module is used to collect basic environmental data and determine whether an event is triggered; The data pruning operation module is used to perform event-driven data pruning and inference chain selection; The model reconstruction analysis module is used to reconstruct the water demand prediction model and generate a thermal distribution map; The control instruction execution module is used to execute the control instruction; The model response update module is used to adaptively update the event response model and data path.

[0017] Beneficial effects of the present invention: The present invention realizes an event-driven intelligent water-saving control process through structured event description, data pruning and reasoning chain selection mechanism, which can dynamically adjust the collection frequency, accurately identify water-demand areas and generate thermal distribution maps, thereby improving response accuracy, adaptability and water-saving control efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A flowchart of the steps of an intelligent water-saving control method for a water conservancy project provided by one embodiment of the present invention; Figure 2 A basic flow chart of an intelligent water-saving control system for a water conservancy project provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0020] Example 1, with reference to Figure 1 , provides an intelligent water-saving control method for water conservancy projects, comprising the following steps: Step S1: construct an event ontology model and initialize state parameters.

[0021] Step S2: Collect basic environmental data and determine whether an event is triggered.

[0022] Step S3: Execute event-driven data pruning and inference chain selection.

[0023] Step S4: reconstruct the water demand prediction model and generate a thermal distribution map.

[0024] Step S5: executing the control instruction.

[0025] Step S6: adaptively update the event response model and data path.

[0026] Step S1 includes the following sub-steps: Step S101 : defining typical water demand response events, which include meteorological events, crop status events, and hydrological fluctuation events.

[0027] Meteorological events include the number of consecutive high temperature days exceeding the set value, hourly precipitation exceeding the infiltration limit of the irrigation area, and wind speed continuously exceeding the critical threshold.

[0028] Crop status events include abnormal crop spectral reflectance, increased leaf curling trends, and signs of growth stagnation.

[0029] Hydrological fluctuation events include sudden changes in channel water inflow and exceeding threshold fluctuations in flow velocity in the terminal channel system.

[0030] A structured description template is established for typical water demand response events. The structured description template includes event number, event classification, event name, trigger logic, affected object type, control recommendation action, priority level, effective time range and response path number, and outputs the event ontology model.

[0031] Step S101 systematically defines typical water demand response events and their classifications, covering common meteorological anomalies, crop stress, and hydrological fluctuations in irrigation district management, and converts them into standardized structured description templates, thereby improving the uniformity of event expression and the accuracy of system identification, and laying a logical foundation for implementing event-driven data processing processes.

[0032] Step S102: Parse the structured description template of a typical water demand response event, identify the data fields of the structured description template, classify the sensing devices and sensing data sources, and output a standard collection channel directory. The sensing data sources include: Soil moisture sensing node: used to collect the moisture status of the surface and middle layers of soil.

[0033] Crop status image acquisition node: used to collect crop canopy images, color index and morphological changes.

[0034] Channel flow counting node: used to collect flow velocity and flow fluctuation information at all levels of the channel.

[0035] Meteorological micro-station node: used to collect wind speed data, temperature data, humidity data, rainfall data and radiation data.

[0036] Establish a mapping relationship between data fields and typical water demand response events.

[0037] Step S102 parses the data fields in the structured description template to clarify the perception information that each type of event depends on, establishes a one-to-one correspondence between event fields and perception nodes, and outputs a standard collection channel directory, effectively reducing redundant data access and improving the controllability and response speed of the system perception structure.

[0038] Step S103: Initialize state parameters, which include the initial soil moisture value, crop type, growth stage, and historical irrigation record status corresponding to each plot.

[0039] Step S103 establishes the system's initial cognitive state by initializing the soil moisture, crop type, growth stage, and historical irrigation records of each regulated plot, providing data basis for event triggering judgment and subsequent generation of differentiated regulation strategies, thereby enhancing the personalization, pertinence, and continuity of system operation.

[0040] Step S1 constructs a structured description template for typical water demand response events, establishes a correspondence between events and perception data sources, and initializes the state parameters of each block, providing basic support for subsequent event identification, data collection scheduling, and reasoning path selection, significantly improving the system's adaptability to complex water demand situations and its pre-regulation judgment capabilities.

[0041] Step S2 includes the following sub-steps: Step S201: Data is collected based on the perception data source, and current state data and current environmental data are output. The current state data includes: surface soil moisture ratio of the plot, middle soil moisture ratio of the plot, crop canopy image, crop color index, crop morphology, real-time flow velocity and flow fluctuation trend of channels at all levels. The current environmental data includes recorded wind speed data, temperature data, humidity data, rainfall data and radiation data. According to the preset low-risk data threshold, it is determined whether the current state data and current environmental data are within the low-risk data threshold. If they are within the low-risk data threshold, it is determined that the current state is a low-risk state, and an acquisition frequency reduction signal is output.

[0042] In step S201, by collecting the perception data source in real time and outputting the current status data and current environmental data in a structured manner, the system can determine whether the irrigation area is in a stable state based on the preset low-risk data threshold, and automatically send a signal to reduce the collection frequency based on the judgment result, thereby actively reducing the system operation energy consumption during non-critical periods and improving the overall resource scheduling efficiency.

[0043] Step S202: parse the trigger logic field in the data field corresponding to the typical water demand response event, extract the indicator value corresponding to the data field from the current state data and the current environment data, and determine whether the indicator value meets the trigger condition based on the priority level and the trigger logic field.

[0044] In step S202, by parsing the trigger logic fields in the structured description template corresponding to the typical water demand response event and extracting key indicator values ​​from the collected data for judgment, the system can accurately identify whether there is a state that meets the trigger conditions of a specific event, providing a basis for subsequent reasoning chain calls and control preparations.

[0045] Step S203: If there is a typical water demand response event whose one or more indicator values ​​meet the trigger conditions, the status of the current state data and the current environmental data is set to the event occurrence, and the corresponding event number, event classification, event name, current state data and current environmental data are recorded, the inference chain number corresponding to the typical water demand response event is started, and an acquisition frequency increase signal is output.

[0046] When the system detects that one or more indicator values ​​meet the trigger conditions, step S203 immediately sets the current state to "event occurrence" and records the corresponding event key information. At the same time, it starts the corresponding reasoning chain processing process and increases the perception collection frequency to ensure the immediate activation of high-priority data processing tasks and decision-making accuracy.

[0047] Step S204: If there is no typical water demand response event with one or more indicator values ​​meeting the trigger conditions, the status of the current state data and the current environmental data is set to no critical event, and enters a low-priority hold state. All computing nodes suspend high-level processing tasks and only retain status records.

[0048] When it is determined that no typical water demand response event has been triggered, step S204 marks the status as "no critical event", and the system automatically enters a low-priority hold state, retaining only the basic data recording function and suspending high-level processing tasks to reduce unnecessary resource consumption and achieve energy-saving operation and efficient status monitoring.

[0049] Step S2 dynamically collects and determines the current status data and current environmental data, and combines it with the preset risk threshold and event trigger logic to achieve rapid identification and response level judgment of typical water demand response events. The system can autonomously switch the perception collection frequency and processing priority according to the actual status, thereby improving the system's resource management efficiency and intelligent identification capabilities.

[0050] Step S3 includes the following sub-steps: Step S301: Based on the typical water demand response event that has been triggered, the corresponding data fields and control suggestion actions are extracted, and the minimum necessary data corresponding to the control suggestion actions are screened out from the preset data processing path library and input into the control end. The perception data sources unrelated to the data fields are closed, which is defined as a data pruning operation. The control suggestion actions are associated with the required data fields in the structured description template. After parsing the control suggestion actions, their dependent fields can be directly indexed to further screen the relevant perception data sources.

[0051] In step S301, by extracting the recommended control actions corresponding to the current event and combining them with the established data field mapping relationship in the structured description template, the system can automatically screen out the minimum necessary data inputs closely related to the control actions, while closing irrelevant perception nodes and processing paths, performing data pruning operations, reducing redundant computing tasks and perception costs, and improving system execution efficiency.

[0052] Step S302: Select an inference chain that matches the event classification from the control module library. The inference chain includes a data completion module, a state fusion module, a control target reorganization module, and a path simulation module connected in sequence. The inference chain is executed in combination according to the event call rule.

[0053] Step S302 selects the most appropriate reasoning chain from the preset control module library based on the event classification. The reasoning chain is composed of multiple functional modules in a logical order and is dynamically adjusted according to the event calling rules to ensure that each processing step in the event response process has target adaptability and functional composability, thereby enhancing the versatility and scalability of the control decision-making model.

[0054] Step S303: activating the inference chain, and inputting the minimum necessary data into each module in the inference chain according to the priority level.

[0055] After the inference chain is activated, step S303 inputs the minimum necessary data filtered by data pruning into each functional module in the inference chain according to the priority level, ensuring that high-priority data is processed first, effectively supporting the execution process of data completion, state fusion, target reconstruction and path simulation, and realizing event cascade response and refined control strategy generation.

[0056] Step S3 filters the necessary data fields and processing modules based on the type of triggered event, builds an inference chain that adapts to the event, and executes the processing flow according to the priority input data, which significantly improves the system's data processing efficiency, resource utilization and the targetedness of regulatory decisions, and realizes an intelligent and dynamic differentiated response mechanism.

[0057] Step S4 includes the following sub-steps: Step S401: Obtain the current perception data in the inference chain and compare it with the corresponding data in the historical status database. If the current soil moisture is lower than the historical average or the decline rate of the crop image index is greater than the preset standard decline rate, the land parcel is marked as a potential water-demanding area and an abnormality attention list is output.

[0058] Step S401 compares and analyzes the current perception data with the historical status database, identifies areas where the soil moisture content is significantly low or the crop status index drops sharply, marks them as potential water-demanding areas, and outputs an abnormality watch list, thereby improving the ability to identify sudden water demand risks early.

[0059] Step S402, obtaining the current crop type, current crop growth stage, real-time soil moisture value, soil type and meteorological forecast trend per unit time of the potential water-demanding area. In one embodiment, the unit time is 24 hours, 48 ​​hours or 72 hours, and the output is potential water storage area status data. The potential water-demanding area status data is input into a preset state modeling model for analysis. The analysis includes: if the current crop type is a shallow-rooted crop and the soil type is light soil, the weight of the corresponding potential water-demanding area is increased by a first percentage; if the meteorological forecast trend per unit time is no rainfall and continuous high temperature, the water demand urgency is increased; if the meteorological forecast trend per unit time is rain, the current crop growth stage is dormant or the soil moisture value is higher than a preset soil moisture threshold, the water demand urgency is decreased.

[0060] Output the dynamic water demand priority score of each plot as the basis for sorting and output a structured list.

[0061] Step S402 obtains multi-source state parameters such as crop type, growth stage, real-time soil moisture, soil type, and meteorological forecast trends in potential water-demanding areas, and inputs them into the state modeling model for comprehensive analysis. The water demand urgency of each plot is determined based on the set rules, and a dynamic water demand priority score list is output, achieving accurate sorting and risk assessment driven by multiple factors.

[0062] Step S403: Map the structured list to the irrigation area geographical layer and output a thermal distribution map.

[0063] Step S403 maps the structured list of dynamic water demand priority scores to the irrigation district geographic layer, and outputs a thermal distribution map in a graphical manner to intuitively express the water demand levels of different areas, realize the spatial visualization of the control basis, and provide intuitive and accurate data support for the generation of intelligent control instructions.

[0064] Step S4 identifies potential water-demanding areas by comparing current perception data with historical status, and constructs a dynamic water demand priority model based on crop properties, soil type, and meteorological trends. It ultimately outputs a thermal distribution map that expresses the urgency of water demand for each plot, providing a decision-making basis for the precise issuance of subsequent control instructions, significantly improving the spatial resolution and scientific nature of water-saving control.

[0065] Step S5 includes the following sub-steps: Step S501: input the thermal distribution map into the control execution engine, and generate a zone control instruction according to the dynamic water demand priority score.

[0066] Step S501 inputs the thermal distribution map generated based on the water demand urgency of the plot into the control execution engine, and automatically generates a multi-region zoning control strategy based on the thermal level and priority score, including the irrigation sequence, water ratio and time control parameters of the specific plot, to achieve intelligent conversion from "layer output" to "control instructions".

[0067] In step S502, the edge controller opens and closes the irrigation valve and adjusts the flow rate of the pump station according to the instruction.

[0068] The edge controller in step S502 receives the control instructions issued by the control execution engine, accurately opens and closes the irrigation valves of the plots according to the specified priority and parameters, and adjusts the water flow of the pump station, completing precise irrigation actions at the plot level, reducing the load of the central system, improving the response speed, and achieving efficient closed-loop control.

[0069] Step S5 inputs the thermal distribution map into the control execution engine, and automatically generates differentiated control instructions based on the dynamic water demand priority score, driving the edge controller to accurately control the irrigation valves and pump station flows in each area, realizing intelligent water resource allocation with zone response and quantitative control, effectively improving water-saving efficiency and the timeliness and accuracy of irrigation execution.

[0070] Step S6 includes the following sub-steps: Step S601: Evaluate the results after the control is executed, and output the evaluation results including the amount of soil moisture recovery and the recovery status of the crop state.

[0071] Step S601 evaluates the executed irrigation control behavior and forms a complete control execution feedback result based on the key indicators such as the actual soil moisture recovery and crop status recovery, providing a basis for system performance evaluation and subsequent optimization.

[0072] In step S602, if there is a deviation between the evaluation result and the preset status standard, the deviation is fed back to the event response end to adjust the event determination threshold.

[0073] If there is a significant deviation between the control feedback result and the preset state standard, step S602 will feed back the deviation information to the event response end, automatically adjust the trigger logic threshold of the corresponding event, correct the model's ability to judge similar situations in the future, and improve the accuracy and foresight of event recognition.

[0074] Step S603: Synchronously update the data pruning strategy and the inference chain calling logic.

[0075] Step S603 synchronously updates the data pruning strategy and inference chain calling logic associated with the current event, optimizes the data collection path and computing resource allocation strategy, realizes the personalized evolution of data processing behavior under different event categories, and promotes the gradual improvement of the system intelligence level.

[0076] Step S6 evaluates the control execution results and feeds the results back to the event model and data processing strategy, supports the dynamic update of event judgment logic, data pruning strategy and reasoning path, realizes the system's adaptive learning and optimization capabilities for external environmental changes, and enhances the long-term stability, robustness and evolution capability of the water-saving control system.

[0077] Example 2, reference Figure 2 , provides an intelligent water-saving control system for water conservancy projects, including an ontology model construction module, a data acquisition and judgment module, a data pruning operation module, a model reconstruction and analysis module, a control instruction execution module and a model response update module.

[0078] The ontology model building module is used to build the event ontology model and initialize state parameters.

[0079] The data collection and judgment module is used to collect basic environmental data and determine whether to trigger an event.

[0080] The data pruning operation module is used to perform event-driven data pruning and inference chain selection.

[0081] The model reconstruction analysis module is used to reconstruct the water demand prediction model and generate thermal distribution maps.

[0082] The control instruction execution module is used to execute the control instructions.

[0083] The model response update module is used to adaptively update the event response model and data path.

[0084] The present invention uses predefined typical water demand response events combined with sensory data for real-time identification, can achieve rapid response to abnormal hydrological and meteorological conditions and crop conditions, accurately judge the timing and location of irrigation, and automatically reduce the frequency of operation through the acquisition frequency adjustment mechanism when no key events occur, suspend high-order computing tasks, effectively reduce unnecessary computing resources and perception loads, and through the data pruning mechanism, only retain the necessary data fields and processing paths related to the current event, realize on-demand activation of the reasoning chain, improve processing efficiency, and is suitable for data control tasks in large-scale irrigation areas. It supports multi-parameter state modeling and irrigation area heat map output, can dynamically adjust the control priority, and implement differentiated control strategies; and drives the adaptive update of event models and data paths through control feedback results, has self-evolution capabilities, adopts structured description templates and standard acquisition channel mapping, so that each link of event recognition, data acquisition, reasoning execution and control distribution has unified data standards and calling interfaces, enhancing stability and deployment flexibility.

[0085] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium may be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0086] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent water-saving control method for water conservancy projects, characterized in that: The steps include: Step S1, constructing an event ontology model and initializing state parameters; Step S2, collecting basic environmental data and determining whether an event is triggered; Step S3, performing event-driven data pruning and inference chain selection; Step S4, reconstructing the water demand prediction model and generating a thermal distribution map; Step S5, executing the control instruction; Step S6: adaptively update the event response model and data path.

2. The intelligent water-saving control method for water conservancy projects according to claim 1, characterized in that: The step S1 includes the following sub-steps: Step S101, defining typical water demand response events, wherein the typical water demand response events include meteorological events, crop status events, and hydrological fluctuation events; Meteorological events include the number of consecutive high-temperature days exceeding the set value, hourly precipitation exceeding the irrigation area infiltration limit, and wind speed continuously exceeding the critical threshold; The crop status events include abnormal crop spectral reflectance, increased leaf curling tendency and signs of growth stagnation; The hydrological fluctuation events include sudden changes in channel water inflow and fluctuations exceeding the threshold of flow velocity in the terminal channel system; Establish a structured description template for a typical water demand response event, which includes event number, event classification, event name, trigger logic, affected object type, control suggestion action, priority level, effective time range and response path number, and output the event ontology model; Step S102: Parse the structured description template of a typical water demand response event, identify the data fields of the structured description template, classify the sensing devices and sensing data sources, and output a standard collection channel directory. The sensing data sources include: Soil moisture sensing node: used to collect the moisture status of the surface and middle layers of soil; Crop status image acquisition node: used to collect crop canopy images, color index and morphological changes; Channel flow counting node: used to collect flow velocity and flow fluctuation information at all levels of the channel; Meteorological micro-station node: used to collect wind speed data, temperature data, humidity data, rainfall data and radiation data; Establish mapping relationships between data fields and typical water demand response events; Step S103: Initializing state parameters, which include the initial soil moisture value, crop type, growth stage, and historical irrigation record status corresponding to each plot.

3. The intelligent water-saving control method for water conservancy projects according to claim 2, characterized in that: The step S2 includes the following sub-steps: Step S201: Data is collected based on the perception data source, and current state data and current environmental data are output. The current state data includes: surface soil moisture ratio of the plot, mid-soil moisture ratio of the plot, crop canopy image, crop color index, crop morphology, real-time flow velocity and flow fluctuation trend of each level of channels. The current environmental data includes recorded wind speed data, temperature data, humidity data, rainfall data, and radiation data. Based on a preset low-risk data threshold, it is determined whether the current state data and current environmental data are within the low-risk data threshold. If so, the current state is determined to be a low-risk state, and a collection frequency reduction signal is output; Step S202: parse the trigger logic field in the data field corresponding to the typical water demand response event, extract the indicator value corresponding to the data field from the current state data and the current environment data, and determine whether the indicator value meets the trigger condition based on the priority level and the trigger logic field.

4. The intelligent water-saving control method for water conservancy projects according to claim 3, characterized in that: The step S2 further includes the following sub-steps: Step S203: If there is a typical water demand response event with one or more indicator values ​​that meet the trigger conditions, the status of the current state data and the current environment data is set to an event occurrence, and the corresponding event number, event category, event name, current state data and current environment data are recorded. The inference chain number corresponding to the typical water demand response event is started, and a collection frequency increase signal is output; Step S204: If there is no typical water demand response event with one or more indicator values ​​meeting the trigger conditions, the status of the current state data and the current environmental data is set to no critical event, and enters a low-priority hold state. All computing nodes suspend high-level processing tasks and only retain status records.

5. The intelligent water-saving control method for water conservancy projects according to claim 4, characterized in that: The step S3 includes the following sub-steps: Step S301: Based on the currently triggered typical water demand response event, the corresponding data fields and control recommended actions are extracted. The minimum necessary data corresponding to the control recommended actions is filtered from the preset data processing path library and input into the control terminal. The perception data sources unrelated to the data fields are closed. This is defined as a data pruning operation. Step S302: selecting an inference chain that matches the event classification from a control module library, wherein the inference chain includes a data completion module, a state fusion module, a control target reorganization module, and a path simulation module connected in sequence, and the inference chain is executed in combination according to the event call rule; Step S303: activating the inference chain, and inputting the minimum necessary data into each module in the inference chain according to the priority level.

6. The intelligent water-saving control method for water conservancy projects according to claim 5, characterized in that: The step S4 includes the following sub-steps: Step S401: Obtain the current perception data in the inference chain and compare it with the corresponding data in the historical status database. If the current soil moisture is lower than the historical average or the decline rate of the crop image index is greater than the preset standard decline rate, the land parcel is marked as a potential water-demanding area and an abnormality attention list is output.

7. The intelligent water-saving control method for water conservancy projects according to claim 6, characterized in that: The step S4 further includes the following sub-steps: Step S402: obtaining the current crop type, current crop growth stage, real-time soil moisture value, soil type, and meteorological forecast trend per unit time of the potential water-demanding area, outputting the data as potential water storage area state data, and inputting the potential water-demanding area state data into a preset state modeling model for analysis, wherein the analysis includes: if the current crop type is a shallow-rooted crop and the soil type is light soil, increasing the weight of the corresponding potential water-demanding area by a first percentage; if the meteorological forecast trend per unit time is no rainfall and continuous high temperature, increasing the water demand urgency; if the meteorological forecast trend per unit time is rain, the current crop growth stage is dormant, or the soil moisture value is higher than a preset soil moisture threshold, decreasing the water demand urgency; Output the dynamic water demand priority score of each plot as the basis for sorting and output a structured list; Step S403: Map the structured list to the irrigation area geographical layer and output a thermal distribution map.

8. The intelligent water-saving control method for water conservancy projects according to claim 7, characterized in that: The step S5 includes the following sub-steps: Step S501: Input the thermal distribution map into the control execution engine and generate zone control instructions according to the dynamic water demand priority score; In step S502, the edge controller opens and closes the irrigation valve and adjusts the flow rate of the pump station according to the instruction.

9. The intelligent water-saving control method for water conservancy projects according to claim 8, characterized in that: The step S6 includes the following sub-steps: Step S601: Evaluate the results of the control execution and output the evaluation results including the amount of soil moisture recovery and the recovery of crop status; Step S602: If the evaluation result deviates from the preset status standard, the result is fed back to the event response terminal to adjust the event determination threshold; Step S603: Synchronously update the data pruning strategy and the inference chain calling logic.

10. An intelligent water-saving control system for a water conservancy project, applied to an intelligent water-saving control method for a water conservancy project as claimed in any one of claims 1 to 9, characterized in that: It includes ontology model construction module, data collection and judgment module, data pruning operation module, model reconstruction analysis module, control instruction execution module and model response update module; The ontology model building module is used to build an event ontology model and initialize state parameters; The data collection and judgment module is used to collect basic environmental data and determine whether an event is triggered; The data pruning operation module is used to perform event-driven data pruning and inference chain selection; The model reconstruction analysis module is used to reconstruct the water demand prediction model and generate a thermal distribution map; The control instruction execution module is used to execute the control instruction; The model response update module is used to adaptively update the event response model and data path.

Images