Waterway salt tide evolution simulation system based on long-memory double-autoregression model

Abstract

The invention provides a water channel salt tide evolution simulation system based on a long-memory double-autoregression model, and belongs to the field of data processing. The simulation system comprises a data acquisition module, a data preprocessing module, a model training module, a simulation module and a visualization module. The data acquisition module is responsible for collecting hydrological data of a plurality of positions in a water channel, and the data preprocessing module carries out standardization and non-integer order differential transformation on the data so as to process time correlation of long-term numerical values. And the model training module carries out training by adopting a double-autoregression model based on the preprocessed data, and carries out parameter fitting by using an LSTM neural network to generate a salt tide evolution model. The model comprises a long memory part and an autoregression part which are respectively used for capturing long-term dependence and short-term fluctuation of data. The simulation module utilizes the trained model to predict future salt tide evolution, and the visualization module displays a prediction result in a visual graphic mode.

Description

Technical Field

[0001] The invention belongs to the technical field of data processing, and in particular relates to a waterway saltwater tide evolution simulation system based on a long memory double autoregressive model. Background Art

[0002] Saltwater tide refers to the intrusion of saltwater into inland rivers due to changes in tides and river flow, which leads to a significant increase in the concentration of dissolved salts in the water body. Traditional simulation methods for saltwater tide evolution usually use simple regression models or physical models, which have significant deficiencies in processing complex time series data and capturing long-term dependencies. In particular, traditional methods are difficult to effectively handle nonlinear relationships and time dependencies in data, resulting in low prediction accuracy and reliability. In recent years, with the development of big data technology and machine learning algorithms, the use of long memory models and deep learning technologies for time series analysis and prediction has become a research hotspot. The long memory double autoregressive model combines long memory characteristics and autoregressive characteristics, and can simultaneously capture the long-term dependency and short-term volatility characteristics of time series. The LSTM (long short-term memory) neural network has become an ideal choice for parameter fitting due to its superior performance in processing nonlinear time series data. By comprehensively applying these advanced technologies, it is possible to more accurately simulate and predict the evolution of saltwater tides, providing a scientific basis for water resources management and environmental protection.

[0003] According to the relevant public technologies, the technical scheme with the announcement number CN101510345B proposes a saltwater tide early warning system and method, which remotely monitors the chlorine content of water bodies at predetermined locations by setting up multiple chloride-containing sensors, and uses remote communication devices for real-time monitoring, thereby realizing rapid early warning of saltwater tide data; the technical scheme with the publication number CN105956360A proposes a fuzzy mathematical evaluation method for seawater intrusion, which can numerically reflect the water quality evolution process in the saltwater intrusion area by fitting and calculating multiple evaluation factors; the technical scheme with the publication number EP2851883A1 proposes a coastal groundwater monitoring system, which realizes real-time monitoring of the fresh / salty water interface by arranging sensors with specified intervals and sensitivities in multiple groundwater areas.

[0004] The above technical solutions all propose several technical solutions for monitoring or predicting the values ​​or trends of salt water or saltwater tides, but the simulation technology of saltwater tide evolution based on long observation time is rarely mentioned.

[0005] The foregoing discussion of the background art is intended only to facilitate an understanding of the present invention. This discussion does not acknowledge or admit that any of the material referred to is part of the common general knowledge. Summary of the invention

[0006] The object of the present invention is to provide a waterway saltwater tide evolution simulation system based on a long memory double autoregressive model, the simulation system comprising a data acquisition module, a data preprocessing module, a model training module, a simulation module and a visualization module. The data acquisition module is responsible for collecting hydrological data at multiple locations in the waterway, and the data preprocessing module performs standardization and non-integer order differential transformation on these data to process the time correlation of long-term numerical values. The model training module uses a double autoregressive model for training based on the preprocessed data, and uses an LSTM neural network for parameter fitting to generate a saltwater tide evolution model. The model includes a long memory part and an autoregressive part, which are respectively used to capture the long-term dependence and short-term fluctuations of the data. The simulation module uses the trained model to predict the future evolution of the saltwater tide, and the visualization module displays the prediction results in an intuitive graphical manner.

[0007] The present invention adopts the following technical solution: a waterway saltwater tide evolution simulation system based on a long memory double autoregressive model, the simulation system comprising: A data collection module is configured to collect time series based hydrological data at multiple locations in the waterway; A data preprocessing module is configured to preprocess the collected data; The model training module is configured to generate a saltwater tide evolution model by training with a long memory dual autoregressive model based on the preprocessed data; A simulation module is configured to use the trained saltwater tide evolution model to predict the future saltwater tide evolution; A visualization module, configured to display the prediction results in a visual form; The saltwater tide evolution model is trained and established based on the detection values ​​of dissolved salt concentration in water bodies at multiple designated waterway locations in a time series; the saltwater tide evolution model includes a long memory part and an autoregressive part, which are used to capture the long-term dependence and short-term fluctuation of the data respectively; the saltwater tide evolution model includes the following steps: S100: analyzing the seasonal characteristics of the original time series of dissolved salt concentration, and standardizing the original time series to obtain a standardized time series; S200: performing a non-integer order difference transformation on the standardized time series to process the time correlation of long-term values, thereby obtaining a difference transformation time series; S300: establishing a double autoregressive model based on the differential transformation time series to obtain a saltwater tide evolution model; Preferably, when the data preprocessing module performs the data preprocessing step, it includes performing one or more of the following processing steps on the original data sequence to be processed to ensure that the original data sequence has stability: Unit root test to ensure that the data is stable as required; Long memory test to ensure that the data has long memory; Time-varying volatility test to ensure that the data has time-varying volatility; Preferably, in step S300, one or more of the following sub-steps are performed: After setting the model order R, fit the parameters of the model based on the order R; Preferably, in step S300, it includes using an LSTM neural network to perform parameter fitting, and outputting models corresponding to different orders R; Preferably, in step S300, the model order and parameters corresponding to the minimum value in the RMSE check are selected as the final form of the saltwater tide evolution model; Preferably, the model training module includes: using a root mean square error calculation method to evaluate the error between the predicted value and the actual value of the saltwater tide evolution model; Preferably, for different waterway locations, a saltwater tide evolution model corresponding to the waterway location is trained and established; Preferably, the visualization module includes a display; the display displays the saltwater tide data including the original time series and the saltwater tide data predicted by the saltwater tide evolution model through a data interface.

[0008] The beneficial effects achieved by the present invention are: The simulation system of this technical solution uses a long memory dual autoregressive model and LSTM neural network to simultaneously capture the long-term dependence and short-term fluctuation characteristics in time series data; by standardizing and transforming the dissolved salt concentration data with non-integer order differences, it can effectively handle the nonlinear relationship and time correlation in the data; compared with traditional regression models and physical models, this solution has significant advantages in prediction accuracy and reliability, and can provide more accurate salt tide evolution predictions for water resources management; The simulation system of this technical solution can process hydrological data of multiple locations and multiple variables, and adapt to the prediction needs of saltwater tide evolution under different waterways and environmental conditions through the learning ability of the evolution model; the system can automatically adjust the model parameters according to different waterway locations and historical data to generate a highly adaptable saltwater tide evolution model, which is suitable for a variety of complex hydrological conditions and dynamically changing environments, and has good versatility and scalability; The simulation system of this technical solution can display the prediction results in an intuitive graphical way through the visualization module, including the time series changes of dissolved salt concentration and the future evolution trend of salt tide; through the visualization display, users can understand and analyze the prediction results more clearly, which helps to make scientific and reasonable decisions; The software and hardware parts of the simulation system of the present technical solution adopt a modular design. The various working modules and components of the hardware part of the system, as well as the instructions, parameters, and algorithms of the software part can be conveniently replaced and / or upgraded later, thereby reducing the construction cost and maintenance cost of the present system. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The present invention can be further understood from the following description in conjunction with the accompanying drawings. The components in the figures are not necessarily drawn to scale, but the emphasis is placed on illustrating the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.

[0010] Description of the accompanying drawings: 10-data acquisition module; 20-data preprocessing module; 30-model training module; 40-simulation module; 50-visualization module; 500-computer system; 502-bus; 504-processor; 506-main memory; 508-read-only memory; 510-storage device; 512-display; 514-input device; 516-cursor control device; 518-network device; Figure 1 A schematic diagram of a simulation system according to an embodiment of the present invention; Figure 2 A diagram showing the modeling steps of the saltwater tide evolution model described in an embodiment of the present invention; Figure 3 Schematic diagram of a memory unit of a long short-term neural network according to an embodiment of the present invention; Figure 4 A schematic diagram showing the evolution of a visualized saltwater tide in an embodiment of the present invention; Figure 5 It is a schematic diagram of the framework of the computer system used in the simulation system in the embodiment of the present invention. DETAILED DESCRIPTION

[0011] In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with its embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. For those skilled in the art, after reviewing the following detailed description, other systems, methods and / or features of the present embodiment will become apparent. It is intended that all such additional systems, methods, features and advantages are included in this specification. Included within the scope of the present invention and protected by the appended claims. Additional features of the disclosed embodiments are described in the following detailed description, and these features will be apparent from the following detailed description.

[0012] The same or similar reference numerals in the drawings of the embodiments of the present invention correspond to the same or similar components; in the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "left", "right" and the like indicating the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, it is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or component referred to must have a specific orientation. The invention is constructed and operated in a specific orientation, so the terms describing the positional relationship in the drawings are only used for exemplary description and cannot be understood as a limitation of this patent. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to the specific circumstances.

[0013] Embodiment 1: For example, a waterway saltwater tide evolution simulation system based on a long memory double autoregressive model is proposed. The simulation system is as shown in the attached Figure 1 As shown, including: The data collection module 10 is configured to collect hydrological data based on time series at multiple locations in the waterway; The data preprocessing module 20 is configured to preprocess the collected data; The model training module 30 is configured to generate a saltwater tide evolution model by training with a long memory dual autoregressive model based on the preprocessed data; The simulation module 40 is configured to use the trained saltwater tide evolution model to predict the future saltwater tide evolution; A visualization module 50, configured to display the prediction results in a visual form; The saltwater tide evolution model is trained and established based on the detection values ​​of dissolved salt concentration in water bodies at multiple designated waterway locations in a time series; the saltwater tide evolution model includes a long memory part and an autoregressive part, which are used to capture the long-term dependence and short-term fluctuation of the data respectively; Preferably, the saltwater tide evolution model comprises the following modeling steps: S100: analyzing the seasonal characteristics of the original time series of dissolved salt concentration, and standardizing the original time series to obtain a standardized time series; S200: performing a non-integer order difference transformation on the standardized time series to process the time correlation of long-term values, thereby obtaining a difference transformation time series; S300: establishing a double autoregressive model based on the differential transformation time series to obtain a saltwater tide evolution model; Preferably, when the data preprocessing module performs the data preprocessing step, it includes performing one or more of the following processing steps on the original data sequence to be processed to ensure that the original data sequence has stability: Unit root test to ensure that the data is stable as required; Long memory test to ensure that the data has long memory; Time-varying volatility test to ensure that the data has time-varying volatility; Preferably, in step S300, one or more of the following sub-steps are performed: After setting the model order R, fit the parameters of the model based on the order R; Preferably, in step S300, it includes using an LSTM neural network to perform parameter fitting, and outputting models corresponding to different orders R; Preferably, in step S300, the model order and parameters corresponding to the minimum value in the RMSE check are selected as the final form of the saltwater tide evolution model; Preferably, the model training module includes: using a root mean square error calculation method to evaluate the error between the predicted value and the actual value of the saltwater tide evolution model; Preferably, for different waterway locations, a saltwater tide evolution model corresponding to the waterway location is trained and established; Preferably, the visualization module includes a display; the display displays the saltwater tide data including the original time series and the saltwater tide data predicted by the saltwater tide evolution model through a data interface; In an exemplary embodiment, the data preprocessing module needs to perform data verification on the data to be processed; preferably, before training the saltwater tide evolution model, or when using the trained saltwater tide evolution model for evolution simulation, a data verification step is performed to help ensure the applicability of the data and improve the simulation effect; Preferably, data verification includes the following three sub-steps: (1) Unit root test, the purpose of which is to determine the stationarity of time series data; for time series models (such as AR, MA, ARMA, etc.), including the saltwater tide evolution model in this technical solution, the data is usually required to be stationary; the statistical characteristics of a stationary time series (such as mean and variance) do not change over time, which makes the estimation of model parameters more stable and reliable; if the data is non-stationary, this can be discovered through a unit root test, and measures (such as differential processing) can be taken to make the data stationary, thereby meeting the assumptions of the model; Specifically, the ADF test (Augmented Dickey-Fuller Test) can be used, and its calculation formula is: , formula 1; In formula 1, B0 is the bias constant term; B1*t is the trend term, Δy t =y t -y t-1 ,y t-1is the level variable lagged by one period, γ is the coefficient of the level variable lagged by one period, which is obtained by evaluating the least squares method; a i is the coefficient of the lagged difference term; δ t is the white noise error term, V is the lag order; (2) Long Memory Test: used to determine whether the time series data has long memory. Long memory (or long-term dependence) indicates that there is a long-term correlation between data points. Time series with long memory are more complex and challenging in long-term prediction. By testing long memory, we can select the corresponding sub-model suitable for processing this characteristic, and improve the model's ability to describe and predict long-term dependence. Among them, the Hurst index (H) is a common method to test whether a time series has long memory. Preferably, the R / S analysis method can be used to determine the H value. When H>0.5, it indicates that there is long memory in the time series. (3) Time-varying volatility test, to determine whether the time series data has time-varying volatility, such as ARCH effect); Time-varying volatility means that the volatility of data changes over time, which is very common in data in the fields of finance, meteorology, etc. If this feature is not taken into account, the model may not be able to accurately capture the true dynamics of the data, affecting the accuracy of the prediction; by detecting time-varying volatility, it can be determined whether it is necessary to use models such as GARCH to deal with this feature in order to better fit the data and make predictions; preferably, the Lagrange multiplier is used to test the time-varying volatility effect, and its null hypothesis is that the residual sequence of the conditional mean model does not have time-varying volatility. The specific calculation method is already well known to relevant technical personnel and will not be described here; Through the three tests, we can fully understand the characteristics of the data and ensure that the data is suitable for training a specific sub-model. At the same time, after passing the above tests, we can understand the characteristics of different data and select the appropriate sub-model for simulation, which can significantly improve the model's fit and prediction accuracy. In an exemplary embodiment, as shown in the attached Figure 2 As shown, the modeling process includes three main steps; illustratively, in step S100, the following calculation formula is used to obtain the standardized time series: , formula 1; γ(n) and σ(n) are the mean and standard deviation of dissolved salt concentration on the nth day (n = 1, ..., N) of a year, where n = 366 in leap years and n = 365 in ordinary years; O(t) = {o １ , o2…, o T} is the original time series of dissolved salt concentration values, Q(t)=q １ ,q2…,q TTo eliminate the seasonal effects of the standardized time series, the sequence lengths of O(t) and Q(t) are both T. It should be noted that multiple monitoring values ​​can be obtained in each day's monitoring; After the above processing, the data is converted to a standard normal distribution with a mean of 0 and a standard deviation of 1 by subtracting the mean and dividing by the standard deviation to eliminate the influence of seasonal characteristics and other non-stationary factors on the model; Furthermore, in step S200 of the modeling process, the standardized time series is transformed into a non-integer order difference using the following calculation formula: , formula 2; in, , formula 3; In equations 2 and 3, ω p is the corresponding weight of the value at different time nodes t in the time series, and the initial value ω0=1; f is the difference order (non-integer); p is the time lag value, p=1,…,t; q t represents the value of the time series at time point t after difference; Q t-p is the value of the original time series at time point tp; Furthermore, in step S300 of the modeling process, the following calculation formula is used to establish the final saltwater tide evolution model: , formula 4; In formula 4, d is a constant term and d>0; a i is the linear part coefficient, and a i are all real numbers; β i is the nonlinear part coefficient, and β i >0; set {ε t} is a random variable with a mean of 0 and a variance of 1; R is the model order; Among them, the linear part is:

[0014] By using the values ​​of the previous R time points, the current value q is predicted through linear combination t ; This part mainly describes q t The conditional mean information, that is, the q value before the given value (that is, q t-1 ,q t-2 ,……q0), q t Expected value; The nonlinear part, i.e. , Used to represent the change in volatility, the random variable ε acts as a noise or error term, which affects the nonlinear part; d acts as a constant term to ensure the positive value of volatility.

[0015] Embodiment 2: This embodiment should be understood to include at least all the features of any of the above embodiments, and further improve upon them; Further, in a preferred exemplary embodiment, an LSTM network is used for parameter fitting, and the parameters to be fitted include the linear part coefficient a i , nonlinear part coefficient β i and the constant term d; LSTM (Long Short-Term Memory) network is a deep learning model specially designed for processing and predicting time series data. It can effectively capture and maintain long-term dependencies in time series by introducing memory units and gating mechanisms. For the saltwater tide evolution model, the change of dissolved salt concentration may be affected by a variety of complex factors, including seasonal changes, meteorological conditions and tides. The saltwater tide evolution model needs to consider the change and dependency of dissolved salt concentration over a long period of time. The LSTM network can retain important information for a long time through its memory units, so that the model can use past historical data to make accurate predictions when predicting future saltwater tide evolution. At the same time, in the dual autoregressive model containing linear and nonlinear parts, the LSTM network can automatically learn the complex nonlinear relationships in the data, thereby optimizing the parameters of the model and improving the prediction accuracy. The LSTM network shows good robustness when processing noisy and irregularly spaced data. Through training with a large amount of data, the generalization ability of the model is improved to avoid overfitting problems. As attached Figure 3 As shown, a schematic diagram of the memory unit of the LSTM neural network used by the model training module is shown; the LSTM neural network may include multiple LSTM layers, wherein each LSTM layer includes multiple memory units; the number of LSTM layers is one of the hyperparameters of the saltwater tide evolution model, which can be determined by experimental methods to find the most accurate results; the number of memory units in each LSTM layer is user-configurable; preferably, the number of neural units is determined by the number of time steps that the user wishes to receive predictions in a given time period; Preferably, a time period can be one year, two years or other time lengths with periodic characteristics; how many time steps Δt can be included in a time period, and the time length of each time step Δt can be 12 hours, or 24 hours, or other values; the user can provide the number of time steps by configuring the input; for example, if the user configures a time period length of the saltwater tide evolution model to be one year, each time step length is 12 hours to generate a predicted dissolved salt concentration value; LSTM neural networks use gating mechanisms to add or remove information to the cell state; the gating mechanism allows information to be selectively passed through different cells / cell states; therefore, LSTM neural networks have the ability to selectively remember patterns and can retain information for a long time; the gating mechanism includes input gates, output gates, and forget gates; therefore, memory cells can retain values ​​at any time interval; the three gating mechanisms regulate the process of information flowing into and out of memory cells; the cell state can be understood as a selective memory of the past, while the hidden state of the cell carries the information received over a period of time and passes it to the current time step; Each memory cell has an internal cell state, denoted as "c", and outputs a hidden state, denoted as "h"; the hidden state h is the output of the LSTM cell or cell layer, and the internal state of the LSTM cell (not output) is denoted as the cell state "c"; each LSTM cell can output a hidden state h for each received input, denoted as "i" here; for each input i1, i2...ia, the memory cell outputs the corresponding hidden state value h1, h2...ha at multiple times t1, t2...ta of the user-defined time step Δt; in addition, the last input time step also generates a cell state ca; When the saltwater model is trained, the LSTM neural network receives a set of time series data "i1" of a waterway location within a user-defined time period T, which includes the data of the differential transformation time series, i.e., q t , t=1, 2, ...; The memory unit responds to the received data input "i1" at the user-defined time step Δt and outputs a series of hidden state values ​​h1, h2...ha; and finally outputs a set of unit state values ​​"ca" to predict the predicted value of the monitoring value at each time step in the next time period T+1; Specifically, the forget gate receives the current input x t and the hidden state h of the previous time step t-1 , a Sigmoid activation function is used to generate a value vector f between 0 and 1 t , each element represents the proportion that should be forgotten; where: f t =σ(W f ×[h t-1 ,x t ]+b f ), formula 5-1; Among them, W f is the weight matrix, b f is the bias vector, σ is the Sigmoid function; Furthermore, the input gate determines the amount of new information to be added to the current time step; respectively: Input conditioning: The input gate receives the current input x t and the hidden state h of the previous time step t-1 , a value vector i is generated through a Sigmoid activation function t , each element of which represents a scale that should be updated; and New information generation: Generate a new candidate information vector C(~) t , a Tanh activation function is used to determine the new information content to be added; that is: i t =σ(W i ×[h t-1 ,x t ]+b i ), formula 5-2; C(~) t =tanh(W c ×[h t-1 ,x t ]+b c ), formula 5-3; Among them, W i and W c is the weight matrix, b i and b c is the bias vector, σ is the Sigmoid function, and tanh is the Tanh function; Furthermore, the cell state is used to update and save long-term information through the adjustment of forgetting and input gates; the new cell state C t The cell state C of the previous time step t-1 and the newly generated candidate information C(~) t Joint decision; that is: C t =f t ×C t-1 +i t ×C(~) t , formula 5-4; Furthermore, the output gate determines the output content of the current time step; the output gate receives the current input x t and the hidden state h of the previous time step t-1 , a value vector O is generated through a Sigmoid activation function t , each element represents the proportion that should be output; and, generate a new hidden state h t ; New hidden state h t From the current cell state C t Adjusted by a Tanh activation function and multiplied by the output gate result Ot ;Right now: O t =σ(W o ×[h t-1 ,x t ]+b o ), formula 5-5; h t =O t ×tanh(C t ), formula 5-6; Among them, W o is the weight matrix, b o is the bias vector; In the above equations 5-1 to 5-6, the "×" symbol indicates that the elements in the matrix are multiplied one by one; Furthermore, the model training module can be used for the weight matrix W f , W i , W c , W o and the bias vector b f , b i , b c , b o The specific value of is optimized so that the LSTM neural network can obtain more accurate prediction values; Through the above steps, the LSTM layer can effectively manage and process the input information, capture long-term dependencies, and thus fit the required linear part coefficient a i , nonlinear part coefficient β i and a constant term d.

[0016] Embodiment 3: This embodiment should be understood to include at least all the features of any of the above embodiments, and further improve upon them; For example, as shown in the attached Figure 5 As shown, an implementation of a computer system 500 used in the simulation system is described; the computer system 500 can be used to identify and determine the data storage, calculation and result output process of each working module in the system; Illustratively, computer system 500 includes a bus 502 or other communication mechanism for communicating information, one or more processors 504 coupled to bus 502 for processing information; processor 504 may be, for example, one or more general-purpose microprocessors; The computer system 500 also includes a main memory 506, such as a random access memory (RAM), a cache, and / or other dynamic storage device, coupled to the bus 502 for storing information and instructions to be executed by the processor 504; the main memory 506 may also be used to store temporary variables or other intermediate information during the execution of instructions to be executed by the processor 504; these instructions, when stored in a storage medium accessible to the processor 504, present the computer system 500 as a special-purpose machine customized to perform the operations specified in the instructions; The computer system 500 may also include a read-only memory (ROM) 508 or other static storage device coupled to the bus 502 for storing static information and instructions for the processor 504; a storage device 510 such as a disk, an optical disk, or a USB drive (flash drive) will be coupled to the bus 502 for storing information and instructions; And further, coupled to the bus 502 may also include a display 512 for displaying various information, data, media, etc., an input device 514 for allowing a user of the computer system 500 to control, manipulate, and / or interact with the computer system 500; A preferred way to interact with the management system may be through a cursor control device 516, such as a computer mouse or similar control / navigation mechanism; Furthermore, the computer system 500 may also include a network device 518 coupled to the bus 502; wherein the network device 518 may include, for example, a wired network card, a wireless network card, a switching chip, a router, a switch, and other components; In general, the terms "engine", "component", "system", "database", etc., as used herein, may refer to logic embodied in hardware or firmware, or to a collection of software instructions, possibly with entry and exit points, written in a programming language such as Java, C, or C++; software components may be compiled and linked into executable programs, installed in a dynamic link library, or may be written in an interpreted programming language (e.g., BASIC, Perl, or Python); it should be understood that software components may be callable from other components or from themselves, and / or may be called in response to detected events or interrupts; Software components configured to execute on a computing device may be provided on a computer-readable medium, such as a compact disc, digital video disc, flash drive, diskette, or any other tangible medium, or as a digital download (and may be initially stored) in a compressed or installable format that requires installation, decompression, or decryption prior to execution); such software code may be stored in part or in whole on a memory device of the executing computing device for execution by the computing device; software instructions may be embedded in firmware, such as an EPROM; it is also understood that hardware components may be composed of connected logic units (such as gates and flip-flops), and / or may be composed of programmable units (such as programmable gate arrays or processors); Computer system 500 includes the techniques described herein that may be implemented using custom hardwired logic, one or more ASICs or FPGAs, firmware, and / or program logic that, in combination with the computer system, renders computer system 500 a special-purpose computing device; According to one or more embodiments, the techniques herein may be performed by computer system 500 in response to processor 504 executing one or more sequences of one or more instructions contained in main memory 506; such instructions may be read into main memory 506 from another storage medium, such as storage device 510; execution of the sequences of instructions contained in main memory 506 causes processor 504 to perform the process steps described herein; in alternative embodiments, hard-wired circuitry may be used in place of or in combination with software instructions; As used herein, the term "non-transitory media" and similar terms refer to any media that store data and / or instructions that cause a machine to operate in a specific manner; such non-transitory media may include non-volatile media and / or volatile media; non-volatile media include, for example, optical or magnetic disks, such as storage device 510; volatile media include dynamic memory, such as main memory 506; Among them, common forms of non-transitory media include, for example, floppy disks, diskettes, hard disks, solid-state drives, magnetic tapes or any other magnetic data storage medium, CD-ROMs, any other optical data storage medium, any physical medium having a pattern of holes, RAM, PROM and EPROM, FLASH-EPROM, NVRAM, any other memory chip or cartridge, and network versions thereof; Non-transient media are distinct from transmission media but may be used in conjunction with transmission media; transmission media participate in the transmission of information between non-transient media; for example, transmission media include coaxial cables, copper wires, and optical fibers, including the wires that make up bus 502; transmission media may also take the form of sound waves or light waves, such as radio waves and infrared data communications.

[0017] Although the present invention has been described above with reference to various embodiments, it should be understood that many changes and modifications may be made without departing from the scope of the present invention. That is, the methods, systems and devices discussed above are examples. Various configurations may appropriately omit, replace or add various processes or components. For example, in alternative configurations, the method may be performed in an order different from the order described, and / or various components may be added, omitted and / or combined. Moreover, the features described with respect to certain configurations may be combined in various other configurations, such as different aspects and elements of the configurations may be combined in a similar manner. In addition, the elements therein may be updated as the technology develops, i.e., many elements are examples and do not limit the scope of the present disclosure or claims.

[0018] Specific details are given in the specification to provide a thorough understanding of the exemplary configurations including implementations. However, the configurations may be practiced without these specific details. For example, well-known circuits, processes, algorithms, structures, and techniques have been shown without unnecessary details to avoid obscuring the configurations. This description provides only example configurations and does not limit the scope, applicability, or configurations of the claims. On the contrary, the foregoing description of the configurations will provide those skilled in the art with an enabling description for implementing the described techniques. Various changes may be made to the functions and arrangements of the elements without departing from the spirit or scope of the present disclosure.

[0019] In summary, it is intended that the above detailed description is considered to be illustrative rather than restrictive, and it should be understood that the above embodiments should be understood to be only used to illustrate the present invention and not to limit the scope of protection of the present invention. After reading the contents of the present invention, the technician can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. A waterway saltwater tide evolution simulation system based on a long memory double autoregressive model, characterized in that: The simulation system comprises: A data collection module is configured to collect time series based hydrological data at multiple locations in the waterway; A data preprocessing module is configured to preprocess the collected data; The model training module is configured to generate a saltwater tide evolution model by training with a long memory dual autoregressive model based on the preprocessed data; A simulation module is configured to use the trained saltwater tide evolution model to predict the future saltwater tide evolution; A visualization module, configured to display the prediction results in a visual form; The saltwater tide evolution model is trained and established based on the detection values ​​of dissolved salt concentration in water bodies at multiple designated waterway locations in a time series; the saltwater tide evolution model includes a long memory part and an autoregressive part, which are used to capture the long-term dependence and short-term fluctuation of the data respectively; the saltwater tide evolution model includes the following steps: S100: analyzing the seasonal characteristics of the original time series of dissolved salt concentration, and standardizing the original time series to obtain a standardized time series; S200: performing a non-integer order difference transformation on the standardized time series to process the time correlation of long-term values, thereby obtaining a difference transformation time series; S300: Establishing a double autoregressive model based on the differential transformation time series to obtain a saltwater tide evolution model.

2. The simulation system according to claim 1, characterized in that: When the data preprocessing module performs the data preprocessing step, it includes performing one or more of the following processing steps on the original data sequence to be processed to ensure that the original data sequence has stability: Unit root test to ensure that the data is stable as required; Long memory test to ensure that the data has long memory; Time-varying volatility test to ensure that the data has time-varying volatility.

3. The simulation system according to claim 2, characterized in that: In step S300, the following sub-steps are performed: After setting the model order R, fit the parameters of the model based on the order R.

4. The simulation system according to claim 3, characterized in that: In step S300, the LSTM neural network is used to perform parameter fitting, and models corresponding to different orders R are output.

5. The simulation system according to claim 4, characterized in that: In step S300, the model order and parameters corresponding to the minimum value in the RMSE check are selected as the final form of the saltwater tide evolution model.

6. The simulation system according to claim 5, characterized in that: The model training module includes: The root mean square error calculation method is used to evaluate the error between the predicted value and the actual value of the saltwater tide evolution model.

7. The simulation system according to claim 6, characterized in that: For different waterway locations, a saltwater tide evolution model corresponding to the waterway location is trained and established.

8. The simulation system according to claim 7, characterized in that: The visualization module includes a display; the display displays the saltwater tide data including the original time series and the saltwater tide data predicted by the saltwater tide evolution model through a data interface.

9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the functions of the simulation system as claimed in any one of claims 1 to 8 are performed.

Images