A comprehensive intelligent management and control method for port water resources

Abstract

The invention relates to the technical field of port water resource management, and discloses a comprehensive intelligent control method for port water resources, comprising dividing historical dispatch logs according to dispatch requirements, and constructing multiple demand dispatch log sets; extracting and analyzing historical demand data, historical water resource data, historical dispatch data, and historical port geographic attribute data, and determining a number of characteristic port geographic attribute data and the degree of influence according to the analysis results; setting weights according to the degree of influence, training historical water resource data, characteristic port geographic attribute data, and corresponding weights for different dispatch requirements, and obtaining a port water resource control model; judging whether water resources need to be dispatched, and if so, generating a water resource dispatch strategy and performing simulation; generating a dispatch evaluation value according to the simulation result and judging whether to generate an adjustment instruction, so as to improve the dispatch efficiency and dispatch rationality of port water resources under different dispatch requirements and the influencing factors of geographic attributes.

Description

Technical Field

[0001] The present application relates to the technical field of port water resource management, and in particular to a comprehensive intelligent management and control method for port water resources. Background Art

[0002] In port operations, water resource management is a vital task. Traditional port water resource management methods mostly rely on manual experience and fixed scheduling rules. This method often seems to be unable to cope with complex and changeable water resource demands and geographical environmental factors.

[0003] Specifically, the port's water resources management corresponds to a variety of scheduling needs. At the same time, the port's geographical attributes also put forward special requirements for the management and scheduling of water resources. If these factors are not fully considered, it will lead to unreasonable and inefficient water resources scheduling, which is directly related to the port's daily operating efficiency, environmental protection and sustainable development capabilities. Summary of the invention

[0004] To solve the above technical problems, the present application provides a comprehensive intelligent management and control method for port water resources. Multiple demand scheduling log sets are constructed through scheduling needs, each demand scheduling log set is analyzed, and the characteristic port geographical attributes and corresponding weights under different scheduling needs are determined. A port water resource scheduling model is constructed according to historical water resource data, characteristic port geographical attribute data and corresponding weights under different scheduling needs, and it is judged whether the actual demand data requires water resource scheduling. If necessary, the water resource scheduling strategy is determined and a scheduling simulation is performed. A scheduling evaluation value is generated according to the simulation results, and the water resource scheduling strategy is adjusted to improve the scheduling efficiency and scheduling rationality of port water resources under different scheduling needs and the influencing factors of geographical attributes.

[0005] In some embodiments of the present application, a comprehensive intelligent management and control method for port water resources is provided, including:

[0006] Obtain several historical scheduling logs and the scheduling requirements corresponding to the historical scheduling logs, divide the historical scheduling logs according to the scheduling requirements, and construct multiple demand scheduling log sets according to the division results;

[0007] Extract and analyze the historical demand data, historical water resource data, historical dispatch data and historical port geographical attribute data of the historical dispatch log in each demand dispatch log set, and determine several characteristic port geographical attribute data of the historical dispatch data under different dispatch demands and the corresponding impact degree according to the analysis results;

[0008] The weights of the corresponding characteristic port geographical attribute data are set according to the degree of influence, and the historical water resource data in the historical dispatch logs with different dispatch requirements, the characteristic port geographical attribute data and the corresponding weights are used for neural network training to obtain the port water resource management and control model;

[0009] Acquire real-time water resource data and generate actual demand data, determine whether water resources need to be dispatched based on the actual demand data, and if so, generate the corresponding water resource dispatch strategy based on the port water resource management and control model, and perform dispatch simulation on the water resource dispatch strategy;

[0010] A scheduling evaluation value is generated according to the scheduling simulation results, and it is determined whether to generate a water resources scheduling strategy and adjustment instructions for the port water resources management and control model according to the scheduling evaluation value.

[0011] In some embodiments of the present application, multiple demand scheduling log sets are constructed according to the division results, including:

[0012] Obtain several historical scheduling logs and the scheduling requirements corresponding to the historical scheduling logs;

[0013] Analyze the dispatch demand and determine relevant information of the dispatch demand, the relevant information including the dispatch type, dispatch purpose, dispatch demand amount and dispatch destination of the dispatch demand;

[0014] Analyze the correlation degree of the relevant information of the scheduling requirements of different historical scheduling logs to obtain the correlation degree of the scheduling requirements of different historical scheduling logs;

[0015] Dividing a plurality of historical scheduling logs according to the correlation degree of scheduling requirements, and obtaining a plurality of historical scheduling logs whose correlation degree of scheduling requirements is greater than a preset correlation degree threshold;

[0016] A demand scheduling log set is constructed according to a plurality of historical scheduling logs whose correlation degree is greater than a preset correlation degree threshold.

[0017] In some embodiments of the present application, several characteristic port geographical attribute data and corresponding impact levels under different scheduling requirements are determined according to the analysis results, including:

[0018] Obtain the historical demand data of each historical scheduling log in the same demand scheduling log set, and construct an initial demand data interval of the corresponding demand scheduling log set according to the historical demand data of all historical scheduling logs;

[0019] Compare each demand scheduling log set with the initial demand scheduling data interval of other demand scheduling log sets, and determine whether there is a duplicate data interval;

[0020] If there are repeated data intervals, calculate the occurrence frequency of the historical demand data in the corresponding repeated data intervals in the corresponding demand scheduling log set, allocate the repeated data intervals according to the occurrence frequency, and update the corresponding initial demand data intervals to obtain the final demand data interval of each demand scheduling log set;

[0021] Obtain historical scheduling data of each historical scheduling log in the same demand scheduling log set, randomly select a historical scheduling log as a target scheduling log, compare the target scheduling log with the historical scheduling data of the remaining historical scheduling logs, obtain a first difference feature of the historical scheduling data of the target scheduling log and the remaining historical scheduling logs, quantify the first difference feature, and obtain a first difference amount;

[0022] Generate a difference degree between the target scheduling log and the remaining historical scheduling logs according to a first difference amount of the historical scheduling data and a weight coefficient corresponding to the historical scheduling data;

[0023] Constructing an analysis reference sequence from historical scheduling logs whose difference degree is greater than a preset difference degree threshold, wherein the analysis reference sequence includes a target scheduling log and at least one historical scheduling log, and determining an arrangement order of the historical scheduling logs according to the difference degree;

[0024] Constructing several analysis reference sequences for each demand scheduling log set in turn;

[0025] Acquire the historical port geographic attribute data of the target scheduling log and the historical scheduling log in the same analysis reference sequence, and the second difference feature of the historical port geographic attribute data of the target scheduling log and the historical scheduling log, quantify the second difference feature, and obtain a second difference amount;

[0026] According to the arrangement order of the historical scheduling logs in the same analysis reference sequence, the first difference amount of the historical scheduling data and the second difference amount of the historical port geographical attribute data are sorted respectively to obtain the first difference amount matrix W1 and the second difference amount matrix W2 of the analysis reference sequence;

[0027] ;

[0028] ;

[0029] in, is the first difference of the i1th historical scheduling data in the rth historical scheduling log, r=1,…m, i1=1,2,…n1, is the second difference amount of the i2th historical port geographical attribute data in the gth historical dispatch log, where i2=1, 2, ... n2, m is the total number of historical dispatch logs in the analysis reference sequence, n1 is the total number of historical dispatch data, and n2 is the total number of historical port geographical attribute data in the historical dispatch log;

[0030] The characteristic port geographic attribute data of the corresponding demand scheduling log set and the influence degree of each characteristic port geographic attribute data are determined according to the first difference matrix W1 and the second difference matrix W2 of several analysis reference sequences of the same demand scheduling log set.

[0031] In some embodiments of the present application, determining a number of characteristic port geographical attribute data and corresponding impact levels under different scheduling requirements according to the analysis results also includes:

[0032] Determine a first change amount of each historical dispatch data of adjacent historical dispatch logs according to the first difference amount matrix W1, and determine a second change amount of each historical port geographical attribute data of adjacent historical dispatch logs according to the second difference amount matrix W2;

[0033] Constructing a reference curve graph of a corresponding analysis reference sequence according to the first change amount of each historical scheduling data, the second change amount of each historical port geographical attribute data and the difference degree difference value of adjacent historical scheduling logs of the same analysis reference sequence, wherein the reference curve graph includes a plurality of first change amount curves of historical scheduling data, a plurality of second change amount curves of historical port geographical attribute data and the difference degree difference value curve, and also includes a plurality of reference nodes;

[0034] Calculate a first similarity evaluation value between the same second change amount curve and each first change amount curve and a second similarity evaluation value between the same second change amount curve and the difference degree difference curve;

[0035] generating a comprehensive similarity evaluation value according to the first similarity evaluation value and the second similarity evaluation value;

[0036] If the comprehensive similarity evaluation value is greater than the preset similarity evaluation value threshold, it is determined that the corresponding historical port geographic attribute data has an impact on the historical dispatching data, and the impact degree of the corresponding historical port geographic attribute data is calculated according to the comprehensive similarity evaluation value and the weight coefficient of the corresponding historical dispatching data;

[0037] Pre-set impact thresholds;

[0038] If the impact degree is greater than the preset impact degree threshold, the corresponding historical port geographical attribute data is set as the undetermined characteristic port geographical attribute data of the corresponding analysis reference sequence;

[0039] Compare the geographic attribute data of the undetermined characteristic ports of several analysis reference sequences of the same demand scheduling log set to obtain the credibility of each geographic attribute data of the undetermined characteristic port;

[0040] The pending characteristic port geographic attribute data with a credibility greater than a preset credibility threshold is set as the characteristic port geographic attribute data of the corresponding demand scheduling log set.

[0041] In some embodiments of the present application, the calculation formula of the comprehensive similarity evaluation value is:

[0042] ;

[0043] Among them, P is the comprehensive similarity evaluation value, p1 is the first similarity evaluation conversion coefficient, a1 is the weight coefficient of the first similarity evaluation value, is the slope of the second variation curve at the rth reference node, is the slope of the first change curve of the first historical scheduling data at the rth reference node, tr is the weight coefficient of the rth reference node, is the slope of the first change curve of the n1th historical scheduling data at the rth reference node, p2 is the second similarity evaluation conversion coefficient, a2 is the weight coefficient of the second similarity evaluation value, is the slope of the difference degree difference curve at the rth reference node;

[0044] The calculation formula for the impact degree is:

[0045] ;

[0046] Where D is the degree of influence, h is the total number of historical dispatching data affected by historical port geographic attribute data, n1 is the total number of historical dispatching data, is the comprehensive similarity evaluation value between the historical port geographic attribute data and the i1th historical dispatch data, P0 is the preset similarity evaluation value threshold, is the weight coefficient of the i1th historical scheduling data, and d0 is the impact degree conversion coefficient.

[0047] In some embodiments of the present application, a port water resources management and control model is obtained, including:

[0048] The weight of each characteristic port geographical attribute data is set according to the number of characteristic port geographical attribute data of the same demand scheduling log set and the influence degree of the corresponding characteristic port geographical attribute data;

[0049] The final demand data interval of each demand scheduling log set, the historical water resources data of each historical scheduling log, the characteristic port geographical attribute data and the corresponding final weights are input into the preset CNN model, and the neural network training is performed to obtain the control sub-model of each demand scheduling log set;

[0050] The control sub-models of all demand scheduling log sets are integrated to obtain the port water resources control model.

[0051] In some embodiments of the present application, if yes, generating a corresponding water resource scheduling strategy according to the port water resource management and control model includes:

[0052] Pre-set multiple port management evaluation indicators and obtain real-time relevant data of each port management evaluation indicator;

[0053] Compare the real-time relevant data with the corresponding standard data interval, and generate the current actual demand data of the port based on the comparison results and combined with the real-time water resources data;

[0054] Compare the actual demand data with each final demand data interval respectively. If the actual demand data is not within the final demand data interval, there is no need to dispatch water resources.

[0055] If the actual demand data is within the final demand data interval, the actual demand data is input into the port water resources management and control model to obtain the water resources scheduling strategy for the actual demand data.

[0056] In some embodiments of the present application, a scheduling simulation is performed on a water resource scheduling strategy, including:

[0057] Obtain the actual demand data and scenario data involved in the water resources scheduling strategy, build a simulation application scenario of the water resources scheduling strategy based on the scenario data, and determine the focus factors in the simulation application scenario that are associated with the port management evaluation indicators;

[0058] Extract the simulated scheduling data in the water resource scheduling strategy, and obtain the dynamic relationship between the simulated scheduling data and the factors of interest in the simulation application scenario;

[0059] According to the dynamic relationship and the simulation application environment, the location connection is established to obtain the scheduling simulation model of the water resources scheduling strategy.

[0060] In some embodiments of the present application, generating a scheduling evaluation value according to a scheduling simulation result includes:

[0061] Obtaining a scheduling simulation result according to a scheduling simulation model of a water resources scheduling strategy, wherein the scheduling simulation result includes simulation-related data of multiple port management evaluation indicators and simulated demand data after simulation;

[0062] Subtracting the simulated relevant data of multiple port management evaluation indicators from the corresponding standard data interval to obtain multiple first simulated relevant data difference values, and subtracting the simulated relevant data of multiple port management evaluation indicators from the corresponding real-time relevant data to obtain multiple second simulated relevant data difference values ​​and relevant data change trends;

[0063] Generate a first dispatch sub-evaluation value corresponding to the port management evaluation index according to a plurality of first simulation related data differences, a plurality of second simulation related data differences and a related data change trend;

[0064] Generate a first scheduling evaluation value of the water resources scheduling strategy according to the first scheduling sub-evaluation value and a weight coefficient of the corresponding port management evaluation index;

[0065] Preset standard demand data of actual demand data, compare the standard demand data with the simulated demand data after simulation, and generate the simulated demand satisfaction of the current water resource scheduling strategy based on the comparison results;

[0066] Generate a second dispatch evaluation value of the water resources dispatch strategy according to the simulated demand satisfaction;

[0067] A scheduling evaluation value is generated according to the first scheduling evaluation value and the second scheduling evaluation value.

[0068] In some embodiments of the present application, judging whether to generate a water resource scheduling strategy and an adjustment instruction of a port water resource management and control model according to the scheduling evaluation value includes:

[0069] Presetting a first preset scheduling evaluation value threshold and a second preset scheduling evaluation value threshold;

[0070] If the dispatch evaluation value is less than the first preset dispatch evaluation value threshold, a water resource dispatch strategy and a secondary adjustment instruction of the port water resource control model are generated;

[0071] If the dispatch evaluation value is between the first preset dispatch evaluation value threshold and the second preset dispatch evaluation value threshold, a first-level adjustment instruction of a water resource dispatch strategy and a port water resource control model is generated;

[0072] If the scheduling evaluation value is greater than the second preset scheduling evaluation value, no adjustment instructions for the water resources scheduling strategy and the port water resources management and control model are generated.

[0073] Compared with the prior art, the comprehensive intelligent management and control method of port water resources in the embodiment of the present application has the following beneficial effects:

[0074] Multiple demand scheduling log sets are constructed based on scheduling needs. Each demand scheduling log set is analyzed to determine the characteristic port geographical attributes and corresponding weights under different scheduling needs. A port water resources scheduling model is constructed based on historical water resources data, characteristic port geographical attribute data and corresponding weights under different scheduling needs. It is judged whether the actual demand data requires water resources scheduling. If necessary, the water resources scheduling strategy is determined and a scheduling simulation is performed. A scheduling evaluation value is generated based on the simulation results, and the water resources scheduling strategy is adjusted to improve the scheduling efficiency and scheduling rationality of port water resources under different scheduling needs and the influencing factors of geographical attributes. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 It is a flow chart of a comprehensive intelligent management and control method of port water resources in an embodiment of the present application. DETAILED DESCRIPTION

[0076] The specific implementation methods of the present application are further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application but are not intended to limit the scope of the present application.

[0077] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0078] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0079] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0080] like Figure 1As shown, a comprehensive intelligent management and control method for port water resources in an embodiment of the present application includes:

[0081] Step S101: obtaining a number of historical scheduling logs and scheduling requirements corresponding to the historical scheduling logs, dividing the historical scheduling logs according to the scheduling requirements, and constructing a plurality of demand scheduling log sets according to the division results;

[0082] Step S102: extracting and analyzing historical demand data, historical water resource data, historical dispatch data, and historical port geographic attribute data of historical dispatch logs in each demand dispatch log set, and determining several characteristic port geographic attribute data of historical dispatch data under different dispatch demands and corresponding influence degrees according to the analysis results;

[0083] Step S103: setting the weight of the corresponding characteristic port geographical attribute data according to the degree of influence, and training the historical water resources data, characteristic port geographical attribute data and corresponding weights in the historical scheduling logs of different scheduling requirements through a neural network to obtain a port water resources management and control model;

[0084] Step S104: Acquire real-time water resource data and generate actual demand data, determine whether water resources need to be dispatched according to the actual demand data, and if so, generate a corresponding water resource dispatching strategy according to the port water resource management and control model, and perform dispatch simulation on the water resource dispatching strategy;

[0085] Step S105: Generate a scheduling evaluation value according to the scheduling simulation result, and determine whether to generate a water resource scheduling strategy and an adjustment instruction for the port water resource management and control model according to the scheduling evaluation value.

[0086] In this embodiment, the dispatching demand includes but is not limited to the fresh water demand of ships, cooling water for port facilities, irrigation of port greening, and emergency water use in unexpected situations. The dispatching demand is directly related to the daily operation efficiency, environmental protection and sustainable development capabilities of the port. Water resource data include rainfall, evaporation, water quality of water sources, etc. The dispatching data includes dispatching time, dispatching type (such as fresh water, seawater, etc.), dispatching quantity, etc. The port geographical attribute data includes but is not limited to geographical location, climate conditions (such as temperature, humidity), topography (such as slope, soil type), water source distribution (such as rivers, lakes, groundwater), etc.

[0087] In this embodiment, neural network training includes designing a convolutional neural network (CNN) model structure, including an input layer, a convolution layer, a pooling layer, a fully connected layer, and an output layer. The input layer receives historical demand data, historical water resource data, characteristic port geographic attribute data, and their weights, the convolution layer and the pooling layer are used for feature extraction, and the fully connected layer is used for decision making and output. The pre-processed historical demand data, historical water resource data, characteristic port geographic attribute data, and their weights under different scheduling requirements are input into the CNN model for training to obtain a port water resource management and control model, and during the training process, appropriate loss functions and optimization algorithms are used to adjust the model parameters to improve the prediction accuracy of the model.

[0088] In some embodiments of the present application, multiple demand scheduling log sets are constructed according to the division results, including:

[0089] Obtain several historical scheduling logs and the scheduling requirements corresponding to the historical scheduling logs;

[0090] Analyze the dispatch demand and determine relevant information of the dispatch demand, the relevant information including the dispatch type, dispatch purpose, dispatch demand amount and dispatch destination of the dispatch demand;

[0091] Analyze the correlation degree of the relevant information of the scheduling requirements of different historical scheduling logs to obtain the correlation degree of the scheduling requirements of different historical scheduling logs;

[0092] Dividing a plurality of historical scheduling logs according to the correlation degree of scheduling requirements, and obtaining a plurality of historical scheduling logs whose correlation degree of scheduling requirements is greater than a preset correlation degree threshold;

[0093] A demand scheduling log set is constructed according to a plurality of historical scheduling logs whose correlation degree is greater than a preset correlation degree threshold.

[0094] In this embodiment, a correlation degree analysis is performed based on each relevant information of the scheduling requirements of different historical scheduling logs, and the sub-correlation degrees of the scheduling type, scheduling purpose, scheduling demand amount and scheduling destination of the scheduling requirements of different historical scheduling logs are obtained respectively, and the correlation degree is generated based on multiple sub-correlation degrees and weight coefficients of the corresponding relevant information.

[0095] In this embodiment, the preset correlation degree threshold refers to the minimum correlation degree between historical scheduling logs with a large correlation relationship.

[0096] In this embodiment, by calculating the correlation degree of the scheduling requirements of the historical scheduling logs, several historical scheduling logs are divided into multiple demand scheduling log sets, thereby reducing the amount of subsequent data analysis. The historical scheduling logs are divided according to the scheduling requirements in advance, and the degree of influence of the historical port geographic attribute data on the historical scheduling data under different scheduling requirements is obtained, which lays the foundation for the subsequent construction of the port water resources management and control model and improves the efficiency of water resources scheduling.

[0097] In some embodiments of the present application, several characteristic port geographical attribute data and corresponding impact levels under different scheduling requirements are determined according to the analysis results, including:

[0098] Obtain the historical demand data of each historical scheduling log in the same demand scheduling log set, and construct an initial demand data interval of the corresponding demand scheduling log set according to the historical demand data of all historical scheduling logs;

[0099] Compare each demand scheduling log set with the initial demand scheduling data interval of other demand scheduling log sets, and determine whether there is a duplicate data interval;

[0100] If there are repeated data intervals, calculate the occurrence frequency of the historical demand data in the corresponding repeated data intervals in the corresponding demand scheduling log set, allocate the repeated data intervals according to the occurrence frequency, and update the corresponding initial demand data intervals to obtain the final demand data interval of each demand scheduling log set;

[0101] Obtain historical scheduling data of each historical scheduling log in the same demand scheduling log set, randomly select a historical scheduling log as a target scheduling log, compare the target scheduling log with the historical scheduling data of the remaining historical scheduling logs, obtain a first difference feature of the historical scheduling data of the target scheduling log and the remaining historical scheduling logs, quantify the first difference feature, and obtain a first difference amount;

[0102] Generate a difference degree between the target scheduling log and the remaining historical scheduling logs according to a first difference amount of the historical scheduling data and a weight coefficient corresponding to the historical scheduling data;

[0103] Constructing an analysis reference sequence from historical scheduling logs whose difference degree is greater than a preset difference degree threshold, wherein the analysis reference sequence includes a target scheduling log and at least one historical scheduling log, and determining an arrangement order of the historical scheduling logs according to the difference degree;

[0104] Constructing several analysis reference sequences for each demand scheduling log set in turn;

[0105] Acquire the historical port geographic attribute data of the target scheduling log and the historical scheduling log in the same analysis reference sequence, and the second difference feature of the historical port geographic attribute data of the target scheduling log and the historical scheduling log, quantify the second difference feature, and obtain a second difference amount;

[0106] According to the arrangement order of the historical scheduling logs in the same analysis reference sequence, the first difference amount of the historical scheduling data and the second difference amount of the historical port geographical attribute data are sorted respectively to obtain the first difference amount matrix W1 and the second difference amount matrix W2 of the analysis reference sequence;

[0107] ;

[0108] ;

[0109] in, is the first difference of the i1th historical scheduling data in the rth historical scheduling log, r=1,…m, i1=1,2,…n1, is the second difference amount of the i2th historical port geographical attribute data in the gth historical dispatch log, where i2=1, 2, ... n2, m is the total number of historical dispatch logs in the analysis reference sequence, n1 is the total number of historical dispatch data, and n2 is the total number of historical port geographical attribute data in the historical dispatch log;

[0110] The characteristic port geographic attribute data of the corresponding demand scheduling log set and the influence degree of each characteristic port geographic attribute data are determined according to the first difference matrix W1 and the second difference matrix W2 of several analysis reference sequences of the same demand scheduling log set.

[0111] In this embodiment, the frequency of occurrence refers to the ratio of the number of occurrences of historical demand data in the historical scheduling logs of the same demand scheduling log set to the total number of all historical scheduling logs of the same demand scheduling log. The frequency of occurrence indicates the degree of association between the historical demand data in the repeated data interval and the corresponding demand scheduling log set. The greater the frequency of occurrence, the stronger the degree of association. The corresponding historical demand data is divided into the initial demand data interval of the demand scheduling log set with the strongest degree of association according to the frequency of occurrence, thereby improving the accuracy of the demand data interval of each demand scheduling log set and laying the foundation for the subsequent formulation of a reasonable water resources scheduling strategy.

[0112] In this embodiment, the first difference feature refers to the differences in scheduling type, scheduling time, scheduling amount, etc. in the historical scheduling data. The first difference feature is quantified to obtain the first difference amount. The larger the first difference feature, the larger the corresponding first difference amount. The second difference feature refers to the differences in temperature, humidity, topography, water source distribution, etc. in the historical port geographic attribute data. The larger the second difference feature, the larger the corresponding second difference amount.

[0113] In this embodiment, the first difference matrix refers to the first difference of each historical scheduling data of all historical scheduling logs in the reference sequence, which is arranged according to the sorting results of the difference between the historical scheduling log and the target historical scheduling log. The second difference matrix refers to the second difference of each historical port geographical attribute data of all historical scheduling logs in the reference sequence, which is arranged according to the sorting results of the difference between the historical scheduling log and the target historical scheduling log.

[0114] In this embodiment, according to the first difference matrix and the second difference matrix, the characteristic port geographic attribute data can be clearly and accurately determined and the influence of each characteristic port geographic attribute data on different historical scheduling data can be evaluated, thereby reducing the amount of data processing and analysis, laying the foundation for the subsequent construction of a port water resources management and control model, and improving the efficiency of water resources scheduling.

[0115] In some embodiments of the present application, determining a number of characteristic port geographical attribute data and corresponding impact levels under different scheduling requirements according to the analysis results also includes:

[0116] Determine a first change amount of each historical dispatch data of adjacent historical dispatch logs according to the first difference amount matrix W1, and determine a second change amount of each historical port geographical attribute data of adjacent historical dispatch logs according to the second difference amount matrix W2;

[0117] Constructing a reference curve graph of a corresponding analysis reference sequence according to the first change amount of each historical scheduling data, the second change amount of each historical port geographical attribute data and the difference degree difference value of adjacent historical scheduling logs of the same analysis reference sequence, wherein the reference curve graph includes a plurality of first change amount curves of historical scheduling data, a plurality of second change amount curves of historical port geographical attribute data and the difference degree difference value curve, and also includes a plurality of reference nodes;

[0118] Calculate a first similarity evaluation value between the same second change amount curve and each first change amount curve and a second similarity evaluation value between the same second change amount curve and the difference degree difference curve;

[0119] generating a comprehensive similarity evaluation value according to the first similarity evaluation value and the second similarity evaluation value;

[0120] If the comprehensive similarity evaluation value is greater than the preset similarity evaluation value threshold, it is determined that the corresponding historical port geographic attribute data has an impact on the historical dispatching data, and the impact degree of the corresponding historical port geographic attribute data is calculated according to the comprehensive similarity evaluation value and the weight coefficient of the corresponding historical dispatching data;

[0121] Pre-set impact thresholds;

[0122] If the impact degree is greater than the preset impact degree threshold, the corresponding historical port geographical attribute data is set as the undetermined characteristic port geographical attribute data of the corresponding analysis reference sequence;

[0123] Compare the geographic attribute data of the undetermined characteristic ports of several analysis reference sequences of the same demand scheduling log set to obtain the credibility of each geographic attribute data of the undetermined characteristic port;

[0124] The pending characteristic port geographic attribute data with a credibility greater than a preset credibility threshold is set as the characteristic port geographic attribute data of the corresponding demand scheduling log set.

[0125] In this embodiment, the horizontal axis in the reference curve graph is all the historical scheduling logs in the same analysis reference sequence, each historical scheduling log is a reference node, and the vertical axis is the first change amount of each historical scheduling data of adjacent historical scheduling logs, the second change amount of each historical port geographic attribute data, and the difference in degree of difference.

[0126] In some embodiments of the present application, the calculation formula of the comprehensive similarity evaluation value is:

[0127] ;

[0128] Among them, P is the comprehensive similarity evaluation value, p1 is the first similarity evaluation conversion coefficient, a1 is the weight coefficient of the first similarity evaluation value, is the slope of the second variation curve at the rth reference node, is the slope of the first change curve of the first historical scheduling data at the rth reference node, tr is the weight coefficient of the rth reference node, is the slope of the first change curve of the n1th historical scheduling data at the rth reference node, p2 is the second similarity evaluation conversion coefficient, a2 is the weight coefficient of the second similarity evaluation value, is the slope of the difference degree difference curve at the rth reference node;

[0129] The calculation formula for the impact degree is:

[0130] ;

[0131] Where D is the degree of influence, h is the total number of historical dispatching data affected by historical port geographic attribute data, n1 is the total number of historical dispatching data, is the comprehensive similarity evaluation value between the historical port geographic attribute data and the i1th historical dispatch data, P0 is the preset similarity evaluation value threshold, is the weight coefficient of the i1th historical scheduling data, and d0 is the impact degree conversion coefficient.

[0132] In some embodiments of the present application, a port water resources management and control model is obtained, including:

[0133] The weight of each characteristic port geographical attribute data is set according to the number of characteristic port geographical attribute data of the same demand scheduling log set and the influence degree of the corresponding characteristic port geographical attribute data;

[0134] The final demand data interval of each demand scheduling log set, the historical water resources data of each historical scheduling log, the characteristic port geographical attribute data and the corresponding final weights are input into the preset CNN model, and the neural network training is performed to obtain the control sub-model of each demand scheduling log set;

[0135] The control sub-models of all demand scheduling log sets are integrated to obtain the port water resources control model.

[0136] In this embodiment, by constructing a control sub-model for each demand scheduling log set and fusing the control sub-models of different demand scheduling log sets for training, the port water resources control model can provide more reasonable and effective water resources scheduling strategies for different scheduling needs and characteristic port geographic attribute data with different weights.

[0137] In some embodiments of the present application, if yes, generating a corresponding water resource scheduling strategy according to the port water resource management and control model includes:

[0138] Pre-set multiple port management evaluation indicators and obtain real-time relevant data of each port management evaluation indicator;

[0139] Compare the real-time relevant data with the corresponding standard data interval, and generate the current actual demand data of the port based on the comparison results and combined with the real-time water resources data;

[0140] Compare the actual demand data with each final demand data interval respectively. If the actual demand data is not within the final demand data interval, there is no need to dispatch water resources.

[0141] If the actual demand data is within the final demand data interval, the actual demand data is input into the port water resources management and control model to obtain the water resources scheduling strategy for the actual demand data.

[0142] In this embodiment, the port management evaluation indicators include port operation efficiency evaluation indicators, port environmental protection evaluation indicators, port safety protection evaluation indicators, etc. The real-time related data refers to the data associated with each port management evaluation indicator, which is set in advance based on historical data.

[0143] In this embodiment, the standard data interval refers to the data interval corresponding to each data when the corresponding port management evaluation index is qualified. According to the comparison result between the actual demand data and the final demand data interval, it is judged whether water resources need to be dispatched to improve the accuracy and timeliness of water resources dispatching.

[0144] In some embodiments of the present application, a scheduling simulation is performed on a water resource scheduling strategy, including:

[0145] Obtain the actual demand data and scenario data involved in the water resources scheduling strategy, build a simulation application scenario of the water resources scheduling strategy based on the scenario data, and determine the focus factors in the simulation application scenario that are associated with the port management evaluation indicators;

[0146] Extract the simulated scheduling data in the water resource scheduling strategy, and obtain the dynamic relationship between the simulated scheduling data and the factors of interest in the simulation application scenario;

[0147] According to the dynamic relationship and the simulation application environment, the location connection is established to obtain the scheduling simulation model of the water resources scheduling strategy.

[0148] In this embodiment, the scene data includes the equipment involved, weather, geographical attributes, etc.

[0149] In this embodiment, a scheduling simulation model of the current water resources scheduling strategy is built by simulating application scenarios and dynamic relationships, so as to simulate the application process of the water resources scheduling strategy and calculate the scheduling evaluation value. The management effect of the water resources scheduling strategy is evaluated according to the scheduling evaluation value, which is conducive to discovering and improving the shortcomings of the scheduling strategy, thereby improving the port water resources scheduling effect.

[0150] In some embodiments of the present application, generating a scheduling evaluation value according to a scheduling simulation result includes:

[0151] Obtaining a scheduling simulation result according to a scheduling simulation model of a water resources scheduling strategy, wherein the scheduling simulation result includes simulation-related data of multiple port management evaluation indicators and simulated demand data after simulation;

[0152] Subtracting the simulated relevant data of multiple port management evaluation indicators from the corresponding standard data interval to obtain multiple first simulated relevant data difference values, and subtracting the simulated relevant data of multiple port management evaluation indicators from the corresponding real-time relevant data to obtain multiple second simulated relevant data difference values ​​and relevant data change trends;

[0153] Generate a first dispatch sub-evaluation value corresponding to the port management evaluation index according to a plurality of first simulation related data differences, a plurality of second simulation related data differences and a related data change trend;

[0154] Generate a first scheduling evaluation value of the water resources scheduling strategy according to the first scheduling sub-evaluation value and a weight coefficient of the corresponding port management evaluation index;

[0155] Preset standard demand data of actual demand data, compare the standard demand data with the simulated demand data after simulation, and generate the simulated demand satisfaction of the current water resource scheduling strategy based on the comparison results;

[0156] Generate a second dispatch evaluation value of the water resources dispatch strategy according to the simulated demand satisfaction;

[0157] A scheduling evaluation value is generated according to the first scheduling evaluation value and the second scheduling evaluation value.

[0158] In this embodiment, the calculation formula of the first scheduling sub-evaluation value is:

[0159] ;

[0160] in, is the v1th first simulation related data difference of the first port management evaluation index, f1 is the total number of the first simulation related data difference and the second simulation related data difference of the first port management evaluation index, is the vu-th first simulation related data difference of the u-th port management evaluation index, fu is the total number of the first simulation related data difference and the second simulation related data difference of the u-th port management evaluation index, u is the total number of port management evaluation indicators, is the difference of the o1th second simulation related data of the first port management evaluation index, The selection coefficient of the change trend of the o1th relevant data of the first port management evaluation index is 1 when the change trend of the relevant data is a positive trend (that is, the second simulated relevant data changes in the direction of the standard data corresponding to the real-time relevant data), otherwise it is -1. is the difference of the outh second simulation related data of the uth port management evaluation index, The selection coefficient of the change trend of the outh relevant data of the uth port management evaluation index, x1 is the first scheduling conversion coefficient of multiple first simulated relevant data differences, and x2 is the second scheduling conversion coefficient of multiple second simulated relevant data differences.

[0161] In this embodiment, when the difference between multiple first simulation related data is smaller and the difference between the second simulation related data is larger, the corresponding first scheduling sub-evaluation value of the conversion is larger, and when the first scheduling sub-evaluation value is larger, the corresponding first scheduling evaluation value is larger.

[0162] In this embodiment, the closer the simulated demand data after simulation is to the standard demand data, the greater the corresponding simulated demand satisfaction is. The simulated demand satisfaction is converted into a numerical value of the same dimension as the first scheduling sub-evaluation value and combined with the first scheduling sub-evaluation value to generate a scheduling evaluation value. When the simulated demand satisfaction and the first scheduling sub-evaluation value are larger, the corresponding scheduling evaluation value is larger, that is, the scheduling efficiency of the corresponding water resources scheduling strategy is higher.

[0163] In some embodiments of the present application, judging whether to generate a water resource scheduling strategy and an adjustment instruction of a port water resource management and control model according to the scheduling evaluation value includes:

[0164] Presetting a first preset scheduling evaluation value threshold and a second preset scheduling evaluation value threshold;

[0165] If the dispatch evaluation value is less than the first preset dispatch evaluation value threshold, a water resource dispatch strategy and a secondary adjustment instruction of the port water resource control model are generated;

[0166] If the dispatch evaluation value is between the first preset dispatch evaluation value threshold and the second preset dispatch evaluation value threshold, a first-level adjustment instruction of a water resource dispatch strategy and a port water resource control model is generated;

[0167] If the scheduling evaluation value is greater than the second preset scheduling evaluation value, no adjustment instructions for the water resources scheduling strategy and the port water resources management and control model are generated.

[0168] In this embodiment, the secondary adjustment instruction includes rebuilding the port water resources management model and redetermining the water resources scheduling strategy, and the primary adjustment instruction includes fine-tuning the water resources scheduling strategy until the scheduling evaluation value is greater than the second preset scheduling evaluation value threshold, and iterating the port water resources management model according to the fine-tuned water resources scheduling model.

[0169] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principles of the present application. These improvements and substitutions should also be regarded as the scope of protection of the present application.

Claims

1. A comprehensive intelligent management and control method for port water resources, characterized in that: include: Obtain several historical scheduling logs and the scheduling requirements corresponding to the historical scheduling logs, divide the historical scheduling logs according to the scheduling requirements, and construct multiple demand scheduling log sets according to the division results; Extract and analyze the historical demand data, historical water resource data, historical dispatch data, and historical port geographic attribute data of the historical dispatch log in each demand dispatch log set, and determine several characteristic port geographic attribute data and the corresponding impact degree under different dispatch demands according to the analysis results; The weights of the corresponding characteristic port geographical attribute data are set according to the degree of influence, and the historical water resource data in the historical dispatch logs with different dispatch requirements, the characteristic port geographical attribute data and the corresponding weights are used for neural network training to obtain the port water resource management and control model; Acquire real-time water resource data and generate actual demand data, determine whether water resources need to be dispatched based on the actual demand data, and if so, generate the corresponding water resource dispatch strategy based on the port water resource management and control model, and perform dispatch simulation on the water resource dispatch strategy; Generate a scheduling evaluation value based on the scheduling simulation results, and determine whether to generate a water resources scheduling strategy and an adjustment instruction for the port water resources management and control model based on the scheduling evaluation value; Based on the partitioning results, multiple demand scheduling log sets are constructed, including: Obtain several historical scheduling logs and the scheduling requirements corresponding to the historical scheduling logs; Analyze the dispatch demand and determine relevant information of the dispatch demand, the relevant information including the dispatch type, dispatch purpose, dispatch demand amount and dispatch destination of the dispatch demand; Analyze the correlation degree of the relevant information of the scheduling requirements of different historical scheduling logs to obtain the correlation degree of the scheduling requirements of different historical scheduling logs; Dividing a plurality of historical scheduling logs according to the correlation degree of scheduling requirements, and obtaining a plurality of historical scheduling logs whose correlation degree of scheduling requirements is greater than a preset correlation degree threshold; Building a demand scheduling log set according to multiple historical scheduling logs whose correlation degree is greater than a preset correlation degree threshold; According to the analysis results, several characteristic port geographical attribute data and corresponding impact levels under different scheduling requirements are determined, including: Obtain the historical demand data of each historical scheduling log in the same demand scheduling log set, and construct an initial demand data interval of the corresponding demand scheduling log set according to the historical demand data of all historical scheduling logs; Compare each demand scheduling log set with the initial demand scheduling data interval of other demand scheduling log sets, and determine whether there is a duplicate data interval; If there are repeated data intervals, calculate the occurrence frequency of the historical demand data in the corresponding repeated data intervals in the corresponding demand scheduling log set, allocate the repeated data intervals according to the occurrence frequency, and update the corresponding initial demand data intervals to obtain the final demand data interval of each demand scheduling log set; Obtain historical scheduling data of each historical scheduling log in the same demand scheduling log set, randomly select a historical scheduling log as a target scheduling log, compare the target scheduling log with the historical scheduling data of the remaining historical scheduling logs, obtain a first difference feature of the historical scheduling data of the target scheduling log and the remaining historical scheduling logs, quantify the first difference feature, and obtain a first difference amount; Generate a difference degree between the target scheduling log and the remaining historical scheduling logs according to a first difference amount of the historical scheduling data and a weight coefficient corresponding to the historical scheduling data; Constructing an analysis reference sequence from historical scheduling logs whose difference degree is greater than a preset difference degree threshold, wherein the analysis reference sequence includes a target scheduling log and at least one historical scheduling log, and determining an arrangement order of the historical scheduling logs according to the difference degree; Constructing several analysis reference sequences for each demand scheduling log set in turn; Acquire the historical port geographic attribute data of the target scheduling log and the historical scheduling log in the same analysis reference sequence, and the second difference feature of the historical port geographic attribute data of the target scheduling log and the historical scheduling log, quantify the second difference feature, and obtain a second difference amount; According to the arrangement order of the historical scheduling logs in the same analysis reference sequence, the first difference amount of the historical scheduling data and the second difference amount of the historical port geographical attribute data are sorted respectively to obtain the first difference amount matrix W1 and the second difference amount matrix W2 of the analysis reference sequence; ； ； in, is the first difference of the i1th historical scheduling data in the rth historical scheduling log, r=1,…m, i1=1,2,…n1, is the second difference amount of the i2th historical port geographical attribute data in the gth historical dispatch log, where i2=1, 2, ... n2, m is the total number of historical dispatch logs in the analysis reference sequence, n1 is the total number of historical dispatch data, and n2 is the total number of historical port geographical attribute data in the historical dispatch log; The characteristic port geographic attribute data of the corresponding demand scheduling log set and the influence degree of each characteristic port geographic attribute data are determined according to the first difference matrix W1 and the second difference matrix W2 of several analysis reference sequences of the same demand scheduling log set.

2. The method for comprehensive intelligent management and control of port water resources according to claim 1, characterized in that: According to the analysis results, several characteristic port geographical attribute data and corresponding impact levels under different scheduling requirements are determined, including: Determine a first change amount of each historical dispatch data of adjacent historical dispatch logs according to the first difference amount matrix W1, and determine a second change amount of each historical port geographical attribute data of adjacent historical dispatch logs according to the second difference amount matrix W2; Constructing a reference curve graph of a corresponding analysis reference sequence according to the first change amount of each historical scheduling data, the second change amount of each historical port geographical attribute data and the difference degree difference value of adjacent historical scheduling logs of the same analysis reference sequence, wherein the reference curve graph includes a plurality of first change amount curves of historical scheduling data, a plurality of second change amount curves of historical port geographical attribute data and the difference degree difference value curve, and also includes a plurality of reference nodes; Calculate a first similarity evaluation value between the same second change amount curve and each first change amount curve and a second similarity evaluation value between the same second change amount curve and the difference degree difference curve; generating a comprehensive similarity evaluation value according to the first similarity evaluation value and the second similarity evaluation value; If the comprehensive similarity evaluation value is greater than the preset similarity evaluation value threshold, it is determined that the corresponding historical port geographic attribute data has an impact on the historical dispatching data, and the impact degree of the corresponding historical port geographic attribute data is calculated according to the comprehensive similarity evaluation value and the weight coefficient of the corresponding historical dispatching data; Pre-set impact thresholds; If the impact degree is greater than the preset impact degree threshold, the corresponding historical port geographical attribute data is set as the undetermined characteristic port geographical attribute data of the corresponding analysis reference sequence; Compare the geographic attribute data of the undetermined characteristic ports of several analysis reference sequences of the same demand scheduling log set to obtain the credibility of each geographic attribute data of the undetermined characteristic port; The pending characteristic port geographic attribute data with a credibility greater than a preset credibility threshold is set as the characteristic port geographic attribute data of the corresponding demand scheduling log set.

3. The integrated intelligent management and control method for port water resources according to claim 2, characterized in that: The calculation formula of the comprehensive similarity evaluation value is: ； Where P is the comprehensive similarity evaluation value, p1 is the first similarity evaluation conversion coefficient, a1 is the weight coefficient of the first similarity evaluation value, is the slope of the second variation curve at the rth reference node, is the slope of the first change curve of the first historical scheduling data at the rth reference node, tr is the weight coefficient of the rth reference node, is the slope of the first change curve of the n1th historical scheduling data at the rth reference node, p2 is the second similarity evaluation conversion coefficient, a2 is the weight coefficient of the second similarity evaluation value, is the slope of the difference degree difference curve at the rth reference node; The calculation formula for the impact degree is: ； Where D is the degree of influence, h is the total number of historical dispatching data affected by historical port geographic attribute data, n1 is the total number of historical dispatching data, is the comprehensive similarity evaluation value between the historical port geographic attribute data and the i1th historical dispatch data, P0 is the preset similarity evaluation value threshold, is the weight coefficient of the i1th historical scheduling data, and d0 is the impact degree conversion coefficient.

4. The integrated intelligent management and control method for port water resources according to claim 3, characterized in that: The port water resources management and control model is obtained, including: The weight of each characteristic port geographical attribute data is set according to the number of characteristic port geographical attribute data of the same demand scheduling log set and the influence degree of the corresponding characteristic port geographical attribute data; The final demand data interval of each demand scheduling log set, the historical water resources data of each historical scheduling log, the characteristic port geographical attribute data and the corresponding final weights are input into the preset CNN model, and the neural network training is performed to obtain the control sub-model of each demand scheduling log set; The control sub-models of all demand scheduling log sets are integrated to obtain the port water resources control model.

5. The integrated intelligent management and control method for port water resources according to claim 4, characterized in that: If yes, generate corresponding water resource dispatching strategies according to the port water resource management and control model, including: Pre-set multiple port management evaluation indicators and obtain real-time relevant data of each port management evaluation indicator; Compare the real-time relevant data with the corresponding standard data interval, and generate the current actual demand data of the port based on the comparison results and combined with the real-time water resources data; Compare the actual demand data with each final demand data interval respectively. If the actual demand data is not within the final demand data interval, there is no need to dispatch water resources. If the actual demand data is within the final demand data interval, the actual demand data is input into the port water resources management and control model to obtain the water resources scheduling strategy for the actual demand data.

6. The integrated intelligent management and control method for port water resources according to claim 5, characterized in that: Simulate water resource scheduling strategies, including: Obtain the actual demand data and scenario data involved in the water resources scheduling strategy, build a simulation application scenario of the water resources scheduling strategy based on the scenario data, and determine the focus factors in the simulation application scenario that are associated with the port management evaluation indicators; Extract the simulated scheduling data in the water resource scheduling strategy, and obtain the dynamic relationship between the simulated scheduling data and the factors of interest in the simulation application scenario; According to the dynamic relationship and the simulation application environment, the location connection is established to obtain the scheduling simulation model of the water resources scheduling strategy.

7. The integrated intelligent management and control method for port water resources according to claim 6, characterized in that: Generate scheduling evaluation values ​​based on scheduling simulation results, including: Obtaining a scheduling simulation result according to a scheduling simulation model of a water resources scheduling strategy, wherein the scheduling simulation result includes simulation-related data of multiple port management evaluation indicators and simulated demand data after simulation; Subtracting the simulated relevant data of multiple port management evaluation indicators from the corresponding standard data interval to obtain multiple first simulated relevant data difference values, and subtracting the simulated relevant data of multiple port management evaluation indicators from the corresponding real-time relevant data to obtain multiple second simulated relevant data difference values ​​and relevant data change trends; Generate a first dispatch sub-evaluation value corresponding to the port management evaluation index according to a plurality of first simulation related data differences, a plurality of second simulation related data differences and a related data change trend; Generate a first scheduling evaluation value of the water resources scheduling strategy according to the first scheduling sub-evaluation value and a weight coefficient of the corresponding port management evaluation index; Preset standard demand data of actual demand data, compare the standard demand data with the simulated demand data after simulation, and generate the simulated demand satisfaction of the current water resource scheduling strategy based on the comparison results; Generate a second dispatch evaluation value of the water resources dispatch strategy according to the simulated demand satisfaction; A scheduling evaluation value is generated according to the first scheduling evaluation value and the second scheduling evaluation value.

8. The integrated intelligent management and control method for port water resources according to claim 7, characterized in that: According to the dispatch evaluation value, it is determined whether to generate a water resource dispatch strategy and an adjustment instruction of the port water resource management and control model, including: Presetting a first preset scheduling evaluation value threshold and a second preset scheduling evaluation value threshold; If the dispatch evaluation value is less than the first preset dispatch evaluation value threshold, a water resource dispatch strategy and a secondary adjustment instruction of the port water resource control model are generated; If the dispatch evaluation value is between the first preset dispatch evaluation value threshold and the second preset dispatch evaluation value threshold, a first-level adjustment instruction of a water resource dispatch strategy and a port water resource control model is generated; If the scheduling evaluation value is greater than the second preset scheduling evaluation value, no adjustment instructions for the water resources scheduling strategy and the port water resources management and control model are generated.

Images