CN-121480888-B - Method, equipment and storage medium for predicting and monitoring time window of algal bloom in lakes and reservoirs

CN121480888BCN 121480888 BCN121480888 BCN 121480888BCN-121480888-B

Abstract

The application discloses a method, equipment and a storage medium for scheduling a time window for predicting and monitoring algal bloom in a lake and a reservoir, which comprise the steps of carrying out online assimilation update on the state and parameters of a digital twin body in the lake and the reservoir by adopting a data assimilation algorithm based on observation data; the method comprises the steps of carrying out algal bloom prediction in a future period by adopting an assimilation updated digital twin body, carrying out physical constraint correction and uncertain quantization on a prediction result to output prediction information, generating a detection time window candidate set according to the prediction information, carrying out optimization solution based on the detection time window candidate set and a preset operation constraint to obtain a detection time window scheduling strategy, transmitting the detection time window scheduling strategy to an unmanned detection platform, obtaining actual measurement data acquired after the detection task is executed, recharging the actual measurement data to observation data, and carrying out data registration and model assimilation on the digital twin body. According to the application, by constructing a closed loop system of sensing, assimilation, prediction, scheduling and feedback, the dynamic optimal configuration of monitoring resources along with algal bloom risks is realized, and the early warning timeliness and the resource utilization efficiency are improved.

Inventors

WANG AIJIE
GUO ERWEI
LI WEILONG
TAO YU
ZHOU YAN
XIAO PENG
Xie Luyang
ZHANG LINA
TANG DINGDING
Zhan de
LIU XUEJIN

Assignees

哈尔滨工业大学(深圳)(哈尔滨工业大学深圳科技创新研究院)
中建三局绿色产业投资有限公司

Dates

Publication Date: 20260508
Application Date: 20260108

Claims (8)

1. The lake and reservoir algal bloom prediction and monitoring time window scheduling method is characterized by comprising the following steps of: The method comprises the steps of obtaining multi-source detection data detected by a plurality of sensors in a lake and reservoir, and registering data errors of the multi-source detection data to obtain registered observation data; Performing model data assimilation treatment on the observed data, and performing online assimilation update on the model state and model parameters of a digital twin body based on a data assimilation result, wherein the digital twin body is a virtual dynamic mapping model constructed based on a physical entity of a lake and a reservoir; Predicting the algal bloom information of an future period based on the digital twin organisms after assimilation and update, and generating an algal bloom prediction result based on the algal bloom information, wherein the algal bloom prediction result comprises an algal bloom outbreak probability and a risk distribution; generating detection time window candidate sets for different lake and reservoir detection areas according to the algal bloom outbreak probability and the risk distribution; Carrying out optimization solution based on the detection time window candidate set and a preset operation constraint to obtain a detection time window scheduling strategy; Issuing the detection time window scheduling strategy to an unmanned detection platform to execute a detection task, acquiring actual measurement data corresponding to the executed detection task, and feeding the actual measurement data back to the observation data to perform homologous update on the digital twin; The step of predicting the algal bloom information of the future period based on the digital twin organisms after assimilation and update and generating an algal bloom prediction result based on the algal bloom information comprises the following steps: The method comprises the steps of adopting the model state and the model parameters as initial conditions, synchronously solving a hydrodynamic sub-model, a biochemical sub-model and a photo-thermal sub-model which are coupled in the digital twin body based on weather hydrologic forced data of a preset future period, circularly executing the coupling calculation of the hydrodynamic sub-model, the biochemical sub-model and the photo-thermal sub-model according to a preset time step, and outputting a space-time variation sequence in the future period; Taking the space-time variation sequence as the algal bloom preliminary prediction sequence; introducing physical constraints to correct the algal bloom preliminary prediction sequence, wherein the physical constraints comprise at least one of a mass conservation law and an energy conservation law; Uncertainty quantization is carried out on the corrected algal bloom preliminary prediction sequence to generate confidence information, wherein the confidence information comprises confidence regions or probability distribution; and comprehensively processing the corrected algal bloom preliminary prediction sequence and the confidence information to generate an algal bloom prediction result.
2. The method for predicting and monitoring time window scheduling of algal bloom in lakes and reservoirs according to claim 1, wherein the step of registering the data errors of the multi-source detection data to obtain registered observation data comprises: performing time synchronization and space alignment operation on the multi-source detection data to generate a fusion data set with consistent space-time; calculating a residual mean value based on the residual of the fusion dataset and the predicted value of the digital twin; When the residual average value exceeds a preset residual threshold value, determining that drift data exists in the fusion data set, wherein the drift data is an abnormal value caused by sensor errors or environmental interference; Performing numerical compensation on the drift data, and updating the fusion data set based on a numerical compensation result; And carrying out outlier identification and screening treatment on the updated fusion data set to obtain the observation data.
3. The method for predicting and monitoring time window scheduling of algal bloom in lakes and reservoirs according to claim 1, wherein the step of cyclically executing the coupling calculation of the hydrodynamic sub-model, the biochemical sub-model and the photothermal sub-model according to a preset time step and outputting a time-space variation sequence in a future period comprises: respectively driving the hydrodynamic sub-model and the photothermal sub-model to calculate flow field distribution and illumination water temperature vertical distribution in each time step; Taking the flow field distribution and the illumination water temperature vertical distribution as the input of the biochemical submodel, and calculating the space-time variation of algae biomass and nutrient salt concentration; And taking the calculation result of the current time step of the biochemical sub-model as an initial condition of the next time step, performing loop iteration until the simulation of a preset future period is completed, and outputting the continuous time-space change sequence.
4. The method for predicting and monitoring time window scheduling of algal bloom in lakes and reservoirs according to claim 1, wherein the step of generating detection time window candidate sets for different detection areas of the lakes and reservoirs according to the algal bloom outbreak probability and the risk distribution comprises the steps of: Dividing the lake and reservoir detection area into a plurality of space units based on the algal bloom outbreak probability and the risk distribution; Determining the number and the duration of time windows of the space units in a future scheduling period according to the monitoring priority of each space unit and the preset monitoring frequency requirement, wherein the monitoring priority is determined based on the outbreak probability of algal bloom; Calculating a feasible starting time range of each time window based on the geographic position of the space unit, the navigation speed of the unmanned monitoring platform and the preset operation constraint; And combining the space unit, the time window duration and the feasible starting time range to generate the detection time window candidate set containing a plurality of candidate monitoring task items.
5. The method for predicting and monitoring the time window scheduling of the algal bloom in the lake and reservoir according to claim 1, wherein the step of performing optimization solution based on the detection time window candidate set and a preset operation constraint to obtain a detection time window scheduling strategy comprises the following steps: acquiring platform parameters of an unmanned detection platform, wherein the platform parameters comprise at least one of endurance time, navigation speed and sensor type; Converting the platform parameters into resource constraint and time conflict constraint of a preset optimization model; screening out candidate time windows meeting the resource constraint and the time conflict constraint from the time window candidate set; Comprehensively evaluating the candidate time windows according to the comprehensive evaluation function, and obtaining candidate time window combinations according to the evaluation result; the detection time window scheduling policy is generated based on the candidate time window combination.
6. The method for predicting and monitoring time window scheduling of algal bloom in lakes and reservoirs according to claim 1, wherein the step of feeding back the measured data to the observed data for performing the assimilation update of the digital twin comprises: Marking the measured data as special data with a feedback identifier; Inputting the special data into a processing flow of multi-source registration and quality control, and assimilating with the observed data to generate a feedback observed vector; and updating the model state and model parameters of the digital twin body on line through the feedback observation vector.
7. A lake and reservoir algal bloom prediction and monitoring time window scheduling device, characterized in that the lake and reservoir algal bloom prediction and monitoring time window scheduling device stores a computer program, and the computer program when executed by a processor realizes the lake and reservoir algal bloom prediction and monitoring time window scheduling method of any one of claims 1-6.
8. A storage medium storing a computer program which when executed by a processor implements the method of predicting and monitoring time window of algal bloom in a lake according to any one of claims 1 to 6.

Description

Method, equipment and storage medium for predicting and monitoring time window of algal bloom in lakes and reservoirs Technical Field The application relates to the technical field of lake and reservoir water ring management and cyanobacteria bloom prevention and control, in particular to a method, equipment and storage medium for predicting and monitoring time windows of lake and reservoir algal bloom. Background In the fields of lake and reservoir water environment management and cyanobacteria bloom prevention and control, related technologies depend on schemes such as fixed station/buoy timing monitoring, unmanned ship inspection, statistics or numerical model prediction, path planning and scheduling. However, the technical schemes have obvious limitations and cutting property, and are difficult to meet urgent requirements of efficient early warning and accurate monitoring of algal bloom. Specifically, the technical system has the defects that the fixed period monitoring cannot capture the rapid time-space evolution of the algal bloom, so that the discovery delay is caused, the single-mode statistical model or the off-line running numerical model is identical with the real-time data due to the lack of physical mechanism constraint, the prediction error is rapidly amplified when the external condition is suddenly changed, and the error accumulation cannot be avoided. Meanwhile, the execution of the model prediction result and the monitoring task are mutually disjointed, so that information island of prediction return prediction and scheduling return scheduling is formed, the real-time compensation of sensor data drift, the uncertainty quantification of the prediction result and a degradation fault-tolerant strategy under abnormal conditions are generally lost in the detection process, and the data validity, decision scientificity and service continuity of the whole system are difficult to guarantee. The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present application and is not intended to represent an admission that the foregoing is prior art. Disclosure of Invention The application mainly aims to provide a method, equipment and a storage medium for predicting and monitoring time windows of algal bloom in lakes and reservoirs, and aims to solve the technical problems of resource mismatch and response delay caused by the dislocation of static prediction and rigid monitoring scheduling of algal bloom in lakes and reservoirs in the prior art. In order to achieve the above purpose, the application provides a method for predicting and monitoring time window of algal bloom in lakes and reservoirs, which comprises the following steps: The method comprises the steps of obtaining multi-source detection data detected by a plurality of sensors in a lake and reservoir, and registering data errors of the multi-source detection data to obtain registered observation data; Performing model data assimilation treatment on the observed data, and performing online assimilation update on the model state and model parameters of a digital twin body based on a data assimilation result, wherein the digital twin body is a virtual dynamic mapping model constructed based on a physical entity of a lake and a reservoir; Predicting the algal bloom information of an future period based on the digital twin organisms after assimilation and update, and generating an algal bloom prediction result based on the algal bloom information, wherein the algal bloom prediction result comprises an algal bloom outbreak probability and a risk distribution; generating detection time window candidate sets for different lake and reservoir detection areas according to the algal bloom outbreak probability and the risk distribution; Carrying out optimization solution based on the detection time window candidate set and a preset operation constraint to obtain a detection time window scheduling strategy; And issuing the detection time window scheduling strategy to an unmanned detection platform to execute a detection task, acquiring actual measurement data corresponding to the executed detection task, and feeding the actual measurement data back to the observation data to perform homologous update on the digital twin. In an embodiment, the step of registering the data errors of the multi-source detection data to obtain registered observation data includes: performing time synchronization and space alignment operation on the multi-source detection data to generate a fusion data set with consistent space-time; calculating a residual mean value based on the residual of the fusion dataset and the predicted value of the digital twin; When the residual average value exceeds a preset residual threshold value, determining that drift data exists in the fusion data set, wherein the drift data is an abnormal value caused by sensor errors or environmental interference; Performing numerical compensation on th