Port operation optimization method based on Internet of Things and big data intelligent analysis

Abstract

The invention relates to the technical field of port operation, big data processing and Internet of Things, and discloses a port operation optimization method based on Internet of Things and big data intelligent analysis, and the method comprises the steps: collecting various types of data based on an Internet of Things gateway, and constructing a standard data set according to the various types of preprocessed data; performing feature mining and analysis on the data in the standard data set, determining a predicted to-be-optimized link and predicted to-be-optimized data, and generating a first prediction optimization strategy; carrying out analogue simulation on the first prediction optimization strategy, generating a simulation optimization coefficient according to an analogue simulation result, judging whether the first prediction optimization strategy is corrected or not, and if yes, obtaining a second prediction optimization strategy; and judging whether a conflict exists in the second prediction optimization strategy or not, if so, calculating a priority coefficient, and generating a port operation optimization strategy according to the priority coefficient and the second prediction optimization strategy, thereby improving the port operation efficiency.

Description

Technical Field

[0001] The present application relates to the fields of port operation, big data processing and Internet of Things technology, and in particular to a port operation optimization method based on the Internet of Things and big data intelligent analysis. Background Art

[0002] With the continuous development of global trade, ports, as important hubs connecting sea and land transportation, have a significant impact on the overall logistics chain in terms of operational efficiency and management level. However, there are many problems in the operation of traditional ports, such as unreasonable resource scheduling, low accuracy of data collection, and poor optimization of business links. These problems limit the further improvement of port operation efficiency. Therefore, there is an urgent need for a port operation optimization method based on the Internet of Things and big data intelligent analysis to improve port operation efficiency and ensure smooth port operation. Summary of the invention

[0003] In order to solve the above technical problems, the present application provides a port operation optimization method based on the Internet of Things and big data intelligent analysis. By collecting various types of data in the port operation process and performing feature mining and analysis, the predicted links to be optimized and the predicted data to be optimized are determined, the corresponding first prediction optimization strategy is determined and simulated to obtain the simulation optimization coefficient, the first prediction optimization strategy is corrected to obtain a second prediction optimization strategy, and a port operation optimization strategy is generated according to the second prediction optimization strategy and the optimization priority order to improve the port operation efficiency.

[0004] In some embodiments of the present application, a port operation optimization method based on the Internet of Things and big data intelligent analysis is provided, including: Based on the IoT gateway, various data in the port operation process of the current monitoring node are collected, pre-processed, and a standard data set is constructed based on the pre-processed data. Perform feature mining and analysis on the data in the standard data set, determine the predicted link to be optimized and the predicted data to be optimized of the current monitoring node according to the feature mining and analysis results, and generate a first prediction optimization strategy corresponding to the predicted link to be optimized in combination with a preset optimization model; Simulating the first prediction optimization strategy, generating a simulation optimization coefficient according to the simulation result, and judging whether to modify the first prediction optimization strategy according to the simulation optimization coefficient, and if so, obtaining a second prediction optimization strategy; Determine whether there is a conflict between the second prediction optimization strategies of multiple prediction links to be optimized. If so, calculate the priority coefficient and generate the port operation optimization strategy based on the priority coefficient and the second prediction optimization strategy.

[0005] In some embodiments of the present application, a standard data set is constructed based on various types of preprocessed data, including: Obtain the port operation business in the current monitoring cycle and divide each port operation business into multiple operation links; Preprocessing various types of data, including unifying time units, removing outliers, filling missing values, and data cleaning; Obtain the data sources and data attributes of various types of preprocessed data, classify various types of data according to the data sources and data attributes, and obtain internal and external data of different operation links; Build standard data sets from internal and external data across multiple operations.

[0006] In some embodiments of the present application, the predicted link to be optimized and the predicted data to be optimized of the current monitoring node are determined according to the feature mining and analysis results, including: Pre-set multiple evaluation indicators for each operation link; Obtain the historical monitoring log of each internal data and external data in the standard data set, obtain the historical change characteristics of each internal data and external data in the historical monitoring log, and evaluate the correlation between the historical change characteristics and the historical evaluation values ​​of multiple evaluation indicators of the corresponding operation links; Setting the internal data whose correlation degree is greater than a preset first correlation degree threshold as the internal correlation data corresponding to the evaluation indicator; Setting the external data with a correlation degree greater than a preset second correlation degree threshold as the external correlation data corresponding to the evaluation index; Obtain demand information for each operation link, including standard time nodes and several standard data that meet the demand; Based on the actual change characteristics of the internal and external related data of all evaluation indicators of the same operation link in the current monitoring period and combined with the historical change characteristics, the predicted change characteristics of the corresponding internal and external related data before the standard time node are generated; Generate predicted internal correlation data and predicted external correlation data at a standard time node according to the predicted change characteristics and the internal correlation data and external correlation data of the current monitoring node; Comparing the predicted internal correlation data and the predicted external correlation data with the corresponding standard data respectively, to obtain a first data difference between the predicted internal correlation data and the corresponding standard data and a second data difference between the predicted external correlation data and the corresponding standard data; Generate a predicted operation status coefficient of a corresponding operation link according to a plurality of first data differences and second data differences; Pre-set operating condition coefficient thresholds; If the predicted operating condition coefficient is less than the operating condition coefficient threshold, the corresponding operating link is set as the predicted link to be optimized; The internal associated data whose first data difference is greater than the preset difference threshold and the external associated data whose second data difference is greater than the preset difference threshold in the predicted optimization link are set as the predicted data to be optimized.

[0007] In some embodiments of the present application, generating a predicted operating condition coefficient of a corresponding operating link according to a plurality of first data differences and second data differences includes: Comprehensively analyzing the first data differences corresponding to the predicted internal correlation data of the same evaluation indicator in each operation link, and screening out the predicted internal correlation data whose first data difference is greater than a preset difference threshold; Comprehensively analyzing the second data differences corresponding to the predicted external related data of the same evaluation indicator in the operation link, and screening out the predicted external related data whose second data difference is greater than a preset difference threshold; Generate the prediction difference coefficient of the corresponding evaluation index at the standard time node according to the screened prediction internal correlation data and prediction external correlation data; The calculation formula of the prediction difference coefficient is: ; Among them, Y is the prediction difference coefficient, y0 is the difference conversion coefficient, x1 is the first weight coefficient, n1 is the number of predicted internal correlation data screened out, and N1 is the total number of predicted internal correlation data of the current evaluation index. is the value of the ith first data difference greater than the preset difference threshold, d1i is the weight coefficient of the ith predicted internal correlation data, x2 is the second weight coefficient, n2 is the number of screened predicted external correlation data, N2 is the total number of predicted external correlation data of the current evaluation index, is the value of the s-th second data difference value greater than the preset difference threshold value, and d2s is the weight coefficient of the s-th predicted external correlation data; Generate the predicted operating status coefficient of the corresponding operating link according to the predicted difference coefficient of all evaluation indicators in each operating link at the standard time node; The calculation formula of the predicted operating condition coefficient is: ; Among them, Y2 is the predicted operating status coefficient, y2 is the operating status conversion coefficient, m is the total number of evaluation indicators of the current operating link, Y1v is the predicted difference coefficient of the vth evaluation indicator, and bv is the weight coefficient of the vth evaluation indicator.

[0008] In some embodiments of the present application, the first prediction optimization strategy is simulated, including: Based on the historical optimization logs of the predicted link to be optimized, a preset optimization model corresponding to the predicted link to be optimized is constructed; Inputting the prediction data to be optimized of each prediction link to be optimized into the corresponding preset optimization model to obtain the first prediction optimization strategy of the corresponding prediction link to be optimized; Determine the location information and structural information of the equipment involved in the predicted link to be optimized, lock the location of the preset panoramic map, and generate a static simulation scene corresponding to the predicted link to be optimized based on the locked location information; Acquire multiple dynamic nodes of the first prediction optimization strategy for the prediction link to be optimized; According to the predicted dependency relationship between the internal associated data and the external associated data of the link to be optimized, the dynamic dependency characteristics of the internal associated data and the external associated data under each dynamic node are established; Establish the dynamic operation characteristics of the corresponding equipment at each dynamic node according to the operation relationship between the involved equipment; According to the dynamic dependency relationship and the dynamic operation relationship under multiple identical dynamic nodes, a position connection is established with the static simulation scene, and a simulation model of the first prediction optimization strategy corresponding to the prediction link to be optimized is constructed.

[0009] In some embodiments of the present application, generating a simulation optimization coefficient according to a simulation result includes: Preset the test time of the first prediction optimization strategy of the link to be optimized, and set the data collection node according to the preset time interval; Collect simulation internal correlation data and simulation external correlation data in the corresponding simulation model according to the data collection node, and map them to the corresponding inspection duration to obtain a simulation data change curve diagram, wherein the simulation data change curve diagram includes a plurality of first simulation data change curves for predicting data to be optimized and a second simulation data change curve for predicting data; Generate a standard data line from multiple standard data in the predicted link to be optimized, and map it to the corresponding simulation data change curve diagram, compare the standard data line with the corresponding first simulation data change curve, and determine whether there is an intersection between the first simulation data change curve and the corresponding standard data line within the inspection time; If it exists, the simulation optimization period is determined according to the initial time node where the intersection exists; If it does not exist, obtain the simulation data change trend corresponding to the first simulation data change curve, perform trend extrapolation according to the simulation data change trend, until there is an intersection with the corresponding standard data line, obtain the predicted time node with the intersection, and determine the predicted simulation optimization period; Generate a first simulation optimization coefficient of the first prediction optimization strategy for predicting the data to be optimized according to the simulation optimization period and the prediction simulation optimization period; Acquire the simulation data change trend of the second simulation data change curve, and determine the distance relationship between the simulation data change trend and the corresponding standard data line, wherein the distance relationship includes a close relationship and a far relationship; When the relationship is far away, the mean value of the data difference between the predicted data corresponding to the second simulation data change curve and the corresponding standard data line is calculated; When the relationship is close, the predicted closeness between the simulation data change trend corresponding to the second simulation data change curve and the corresponding standard data line is calculated, and compared with the actual closeness before optimization to obtain the closeness difference; Generate a second simulation optimization coefficient of the second prediction optimization strategy for the remaining prediction data according to the data difference mean and the proximity difference; Generate a compensation coefficient according to the predicted optimization cost of the second prediction optimization strategy of the link to be optimized; Generate a simulation optimization coefficient according to the first simulation optimization coefficient, the second simulation optimization coefficient and the compensation coefficient; Pre-set the simulation optimization coefficient threshold; If the simulation optimization coefficient is less than the simulation optimization coefficient threshold, a correction instruction corresponding to the first prediction optimization strategy is generated.

[0010] In some embodiments of the present application, the calculation formula of the first simulation optimization coefficient is: ; Among them, H1 is the first simulation optimization coefficient, h1 is the simulation optimization conversion coefficient of the simulation optimization period, u1 is the number of prediction data to be optimized that have intersections within the test time in the current prediction to be optimized link, T0 is the test time, This is the simulation optimization period for the z1th predicted data to be optimized. is the weight coefficient of the z1th prediction data to be optimized, h2 is the simulation optimization conversion coefficient of the prediction simulation optimization period, u2 is the number of prediction data to be optimized that have no intersection within the test time in the current prediction to be optimized link, The prediction simulation optimization period for the z2th prediction data to be optimized, is the weight coefficient of the z2th predicted data to be optimized; The calculation formula of the second simulation optimization coefficient is: ; Among them, H2 is the second simulation optimization coefficient, h3 is the simulation optimization conversion coefficient of the difference in the degree of proximity of the predicted data, and u3 is the number of predicted data with a close relationship. is the difference in the degree of closeness of the z3th predicted data, is the weight coefficient of the z3th predicted data, h4 is the simulation optimization conversion coefficient of the mean value of the data difference of the predicted data, u3 is the number of predicted data far away from the relationship, is the mean value of the data difference of the z4th predicted data, is the weight coefficient of the z4th predicted data; The calculation formula of the simulation optimization coefficient is: ; Among them, H0 is the simulation optimization coefficient, f0 is the compensation coefficient, f1 is the weight coefficient of the first simulation optimization coefficient, and f2 is the weight coefficient of the second simulation optimization coefficient.

[0011] In some embodiments of the present application, determining whether there is a conflict between second prediction optimization strategies of multiple prediction links to be optimized includes: Obtain the second prediction optimization strategy of all prediction links to be optimized for the current monitoring node, and extract the optimization target, optimization period, and optimization constraint conditions of each second prediction optimization strategy; A conflict analysis is performed on the optimization objectives, optimization time periods, and optimization constraints of different second prediction optimization strategies to obtain conflicting second prediction optimization strategies and prediction links to be optimized of the second prediction optimization strategies.

[0012] In some embodiments of the present application, if present, calculating the priority coefficient includes: Obtain each predicted link to be optimized that has conflicts, and randomly select one predicted link to be optimized as the target link; Obtain the historical evaluation log of the target link, obtain multiple historical comprehensive evaluation values ​​of the target link according to the historical evaluation log, and calculate the difference of the historical comprehensive evaluation values; Screen out the historical comprehensive evaluation value differences with increasing differences, and construct the historical comprehensive evaluation value difference sequence of the target link; Determine the corresponding adjacent preset time nodes according to the multiple historical evaluation nodes in the historical comprehensive evaluation value difference sequence, and obtain the historical comprehensive evaluation values ​​of other operation links according to the adjacent preset time nodes; According to the time sequence of the historical evaluation nodes of the historical comprehensive evaluation value difference sequence of the target link, the historical comprehensive evaluation value difference sequence of the corresponding time nodes of other operation links is constructed; Calculate the sequence similarity between the historical comprehensive evaluation value difference sequence of each operating link and the historical comprehensive evaluation value difference sequence of the target link. If the sequence similarity is greater than the preset similarity threshold, determine the corresponding operating link as the associated link of the target link, and calculate the correlation coefficient between the associated link and the target link. Generate a priority coefficient of a target link according to the number of associated links, the associated coefficient and the weight coefficient of the corresponding associated link; Generate the priority coefficient of each conflicting prediction link to be optimized in turn.

[0013] In some embodiments of the present application, generating a port operation optimization strategy according to the priority coefficient and the second prediction optimization strategy includes: Setting the optimization order of the corresponding second prediction optimization strategy according to the priority coefficient of the conflicting prediction link to be optimized; The port operation optimization strategy of the current monitoring node is generated according to the optimization sequence and the second prediction optimization strategy for predicting the link to be optimized.

[0014] Compared with the prior art, the port operation optimization method based on the Internet of Things and big data intelligent analysis in the embodiment of the present application has the following beneficial effects: By collecting various types of data during the port operation process and conducting feature mining and analysis, the predicted links and data to be optimized are determined, the corresponding first prediction optimization strategy is determined and simulated to obtain the simulation optimization coefficient, the first prediction optimization strategy is corrected to obtain the second prediction optimization strategy, and the port operation optimization strategy is generated according to the second prediction optimization strategy and optimization priority to improve the port operation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flow chart of a port operation optimization method based on the Internet of Things and big data intelligent analysis in an embodiment of the present application. DETAILED DESCRIPTION

[0016] 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.

[0017] 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.

[0018] 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.

[0019] 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.

[0020] like Figure 1 As shown, a port operation optimization method based on the Internet of Things and big data intelligent analysis in an embodiment of the present application includes: Step S101: collecting various data in the port operation process of the current monitoring node based on the Internet of Things gateway, preprocessing the various data, and constructing a standard data set based on the preprocessed data; Step S102: performing feature mining and analysis on the data in the standard data set, determining the predicted link to be optimized and the predicted data to be optimized of the current monitoring node according to the feature mining and analysis results, and generating a first prediction optimization strategy corresponding to the predicted link to be optimized in combination with a preset optimization model; Step S103: simulate the first prediction optimization strategy, generate a simulation optimization coefficient according to the simulation result, and determine whether to modify the first prediction optimization strategy according to the simulation optimization coefficient. If so, obtain a second prediction optimization strategy; Step S104: determine whether there is a conflict between the second prediction optimization strategies of multiple prediction links to be optimized. If so, calculate the priority coefficient and generate a port operation optimization strategy based on the priority coefficient and the second prediction optimization strategy.

[0021] In this embodiment, the IoT gateway collects various data of each link of port logistics in real time, including RFID tag information, sensor data (such as temperature, humidity, pressure, etc.), vehicle GPS location information, etc.

[0022] In this embodiment, the preset optimization model is constructed based on the historical optimization data of each link predicted to be optimized and the corresponding historical business needs. For example, the link predicted to be optimized is loading and unloading operations. Neural network training is performed based on the historical ship arrival time set, the historical cargo type set and the corresponding historical loading and unloading demand set to obtain the preset optimization model of loading and unloading operations. The preset optimization model outputs the optimized loading and unloading operation sequence and schedule, which reduces waiting time and resource waste and realizes the optimization of loading and unloading operations.

[0023] In some embodiments of the present application, a standard data set is constructed based on various types of preprocessed data, including: Obtain the port operation business in the current monitoring cycle and divide each port operation business into multiple operation links; Preprocessing various types of data, including unifying time units, removing outliers, filling missing values, and data cleaning; Obtain the data sources and data attributes of various types of preprocessed data, classify various types of data according to the data sources and data attributes, and obtain internal and external data of different operation links; Build standard data sets from internal and external data across multiple operations.

[0024] In this embodiment, the port operation process refers to monitoring the operation links of multiple port operation businesses in the current monitoring period to ensure the smooth operation of the port. The port operation business includes transportation business, cargo loading and unloading and warehousing business, tallying business, equipment leasing business, etc. The operation link refers to multiple processes in the port operation business, for example, the loading link, transportation link, stacking and picking link in the cargo loading and unloading business, etc.

[0025] In this embodiment, data cleaning can be performed according to the data cleaning formula, which is: Clean Data = Remove Redundant And Error Data (Raw Data) where: Raw Data is a collection of various types of data; RemoveRedundant And Error Data is a function used to remove redundant and erroneous data; Clean Data is the cleaned data. ‌‌

[0026] In this embodiment, the operation links corresponding to various types of data are determined according to the data sources, and the data of each operation link are divided into internal data and external data according to the data attributes. The internal data refers to the data collected by the corresponding operation link in the port, and the external data refers to the data collected during the operation of the corresponding operation link. For example, the internal data includes the location of containers, the occupancy of the yard, etc., and the external data includes the arrival time of ships, the cargo throughput, etc.

[0027] In this embodiment, by determining the internal and external data of each operation link and constructing a standard data set, the credibility of the data is improved, and the foundation is laid for subsequent feature mining and determining the prediction optimization link.

[0028] In some embodiments of the present application, the predicted link to be optimized and the predicted data to be optimized of the current monitoring node are determined according to the feature mining and analysis results, including: Pre-set multiple evaluation indicators for each operation link; Obtain the historical monitoring log of each internal data and external data in the standard data set, obtain the historical change characteristics of each internal data and external data in the historical monitoring log, and evaluate the correlation between the historical change characteristics and the historical evaluation values ​​of multiple evaluation indicators of the corresponding operation links; Setting the internal data whose correlation degree is greater than a preset first correlation degree threshold as the internal correlation data corresponding to the evaluation indicator; Setting the external data with a correlation degree greater than a preset second correlation degree threshold as the external correlation data corresponding to the evaluation index; Obtain demand information for each operation link, including standard time nodes and several standard data that meet the demand; Based on the actual change characteristics of the internal and external related data of all evaluation indicators of the same operation link in the current monitoring period and combined with the historical change characteristics, the predicted change characteristics of the corresponding internal and external related data before the standard time node are generated; Generate predicted internal correlation data and predicted external correlation data at a standard time node according to the predicted change characteristics and the internal correlation data and external correlation data of the current monitoring node; Comparing the predicted internal correlation data and the predicted external correlation data with the corresponding standard data respectively, to obtain a first data difference between the predicted internal correlation data and the corresponding standard data and a second data difference between the predicted external correlation data and the corresponding standard data; Generate a predicted operation status coefficient of a corresponding operation link according to a plurality of first data differences and second data differences; Pre-set operating condition coefficient thresholds; If the predicted operating condition coefficient is less than the operating condition coefficient threshold, the corresponding operating link is set as the predicted link to be optimized; The internal associated data whose first data difference is greater than the preset difference threshold and the external associated data whose second data difference is greater than the preset difference threshold in the predicted optimization link are set as the predicted data to be optimized.

[0029] In this embodiment, the evaluation index of each operation link refers to the index for evaluating the service quality, cost and benefit, operation safety and stability of the corresponding operation link. For example, the vehicle working rate index, vehicle intact rate index, actual load rate index, etc. in the distribution link reflect the utilization rate and distribution efficiency of vehicles or ships in the distribution process.

[0030] In this embodiment, the historical change characteristics refer to the change trend, change rate and fluctuation degree of the historical data change curve of the corresponding internal data or external data in the historical monitoring log, and the correlation degree between the historical change characteristics and the historical evaluation values ​​is evaluated, that is, the influence of the historical change characteristics of the historical data change curve on the historical evaluation value change curve is evaluated. If one or more of the change trend, change rate and fluctuation degree of the historical data change curve of the internal or external data undergoes a significant change, it will have a significant impact on the historical evaluation value at the corresponding time node in the historical evaluation value change curve, and the significant changes and significant impacts will be quantified to obtain the corresponding correlation degree.

[0031] In this embodiment, the preset first correlation degree threshold is set according to the minimum correlation degree of historical internal correlation data in the port link, and the preset second correlation degree threshold is set according to the minimum correlation degree of historical external correlation data in the operation process of the port link.

[0032] In this embodiment, demand information refers to the operational goals that each operational link needs to achieve, standard time nodes refer to the best time nodes for completing the operational goals, and standard data refers to the standard internal related data and standard external related data corresponding to the completion of the operational goals.

[0033] In this embodiment, the predicted operating condition coefficient refers to the evaluation of whether the corresponding operating link can achieve the required operating target at the standard time node, and the operating condition coefficient threshold is the minimum operating condition coefficient for the corresponding operating link to achieve the required operating target at the standard time node.

[0034] In this embodiment, by determining the predicted internal correlation data and the predicted external correlation data at the standard time node, generating the predicted operating status coefficient, selecting the predicted links to be optimized and the predicted data to be optimized, the foundation is laid for the subsequent generation of optimization strategies, the operation links are optimized, and the port operation efficiency is improved.

[0035] In some embodiments of the present application, generating a predicted operating condition coefficient of a corresponding operating link according to a plurality of first data differences and second data differences includes: Comprehensively analyzing the first data differences corresponding to the predicted internal correlation data of the same evaluation indicator in each operation link, and screening out the predicted internal correlation data whose first data difference is greater than a preset difference threshold; Comprehensively analyzing the second data differences corresponding to the predicted external related data of the same evaluation indicator in the operation link, and screening out the predicted external related data whose second data difference is greater than a preset difference threshold; Generate the prediction difference coefficient of the corresponding evaluation index at the standard time node according to the screened prediction internal correlation data and prediction external correlation data; The calculation formula of the prediction difference coefficient is: ; Among them, Y is the prediction difference coefficient, y0 is the difference conversion coefficient, x1 is the first weight coefficient, n1 is the number of predicted internal correlation data screened out, and N1 is the total number of predicted internal correlation data of the current evaluation index. is the value of the ith first data difference greater than the preset difference threshold, d1i is the weight coefficient of the ith predicted internal correlation data, x2 is the second weight coefficient, n2 is the number of screened predicted external correlation data, N2 is the total number of predicted external correlation data of the current evaluation index, is the value of the s-th second data difference value greater than the preset difference threshold value, and d2s is the weight coefficient of the s-th predicted external correlation data; Generate the predicted operating status coefficient of the corresponding operating link according to the predicted difference coefficient of all evaluation indicators in each operating link at the standard time node; The calculation formula of the predicted operating condition coefficient is: ; Among them, Y2 is the predicted operating status coefficient, y2 is the operating status conversion coefficient, m is the total number of evaluation indicators of the current operating link, Y1v is the predicted difference coefficient of the vth evaluation indicator, and bv is the weight coefficient of the vth evaluation indicator.

[0036] In this embodiment, the difference conversion coefficient refers to converting the first data difference and the second data difference of the predicted internal correlation data and the predicted external correlation data of the same evaluation index that are greater than the preset difference threshold into a value of the same dimension as the predicted difference coefficient. The larger the value, the larger the corresponding predicted difference coefficient, and vice versa.

[0037] In this embodiment, the operating condition conversion coefficient refers to converting the predicted difference coefficients of all evaluation indicators into a coefficient with the same dimension as the predicted operating condition coefficient. When the predicted difference coefficient is larger and the corresponding evaluation indicator weight coefficient is larger, the corresponding predicted operating condition coefficient is smaller, and vice versa.

[0038] In this embodiment, by evaluating the degree of difference between the predicted internal correlation data and the predicted external correlation data of each evaluation indicator and the standard data, a predicted operating status coefficient is generated, the prediction accuracy of the status of each operating link is improved, and the predicted links and predicted data to be optimized are screened out, laying the foundation for the subsequent formulation of optimization strategies and improving port operating efficiency.

[0039] In some embodiments of the present application, the first prediction optimization strategy is simulated, including: Based on the historical optimization logs of the predicted link to be optimized, a preset optimization model corresponding to the predicted link to be optimized is constructed; Inputting the prediction data to be optimized of each prediction link to be optimized into the corresponding preset optimization model to obtain the first prediction optimization strategy of the corresponding prediction link to be optimized; Determine the location information and structural information of the equipment involved in the predicted link to be optimized, lock the location of the preset panoramic map, and generate a static simulation scene corresponding to the predicted link to be optimized based on the locked location information; Acquire multiple dynamic nodes of the first prediction optimization strategy for the prediction link to be optimized; According to the predicted dependency relationship between the internal associated data and the external associated data of the link to be optimized, the dynamic dependency characteristics of the internal associated data and the external associated data under each dynamic node are established; Establish the dynamic operation characteristics of the corresponding equipment at each dynamic node according to the operation relationship between the involved equipment; According to the dynamic dependency relationship and the dynamic operation relationship under multiple identical dynamic nodes, a position connection is established with the static simulation scene, and a simulation model of the first prediction optimization strategy corresponding to the prediction link to be optimized is constructed.

[0040] In this embodiment, the dependency relationship refers to the fact that internal related data and external related data change with the change of other internal related data and other external related data in the same operation link, and the degree of change is greater than the preset change degree threshold. The dynamic dependency feature refers to the degree of dependency corresponding to the dependency relationship. The greater the degree of change, the greater the corresponding degree of dependency.

[0041] In this embodiment, the operating relationship refers to the correlation between the operating conditions of the devices involved, that is, changes in the operating conditions of the devices cause significant changes in the operating conditions of other devices in the same operation link. The dynamic operating characteristics refer to the degree of correlation between the devices with operating relationships. The greater the change in the operating conditions, the greater the corresponding degree of correlation.

[0042] In this embodiment, by constructing a simulation of the first prediction optimization strategy for predicting the link to be optimized, the first prediction optimization strategy is accurately simulated, and the optimization effect of the corresponding first prediction optimization strategy is accurately evaluated according to the simulation results to improve the port operation efficiency.

[0043] In some embodiments of the present application, generating a simulation optimization coefficient according to a simulation result includes: Preset the test time of the first prediction optimization strategy of the link to be optimized, and set the data collection node according to the preset time interval; Collect simulation internal correlation data and simulation external correlation data in the corresponding simulation model according to the data collection node, and map them to the corresponding inspection duration to obtain a simulation data change curve diagram, wherein the simulation data change curve diagram includes a plurality of first simulation data change curves for predicting data to be optimized and a second simulation data change curve for predicting data; Generate a standard data line from multiple standard data in the predicted link to be optimized, and map it to the corresponding simulation data change curve diagram, compare the standard data line with the corresponding first simulation data change curve, and determine whether there is an intersection between the first simulation data change curve and the corresponding standard data line within the inspection time; If it exists, the simulation optimization period is determined according to the initial time node where the intersection exists; If it does not exist, obtain the simulation data change trend corresponding to the first simulation data change curve, perform trend extrapolation according to the simulation data change trend, until there is an intersection with the corresponding standard data line, obtain the predicted time node with the intersection, and determine the predicted simulation optimization period; Generate a first simulation optimization coefficient of the first prediction optimization strategy for predicting the data to be optimized according to the simulation optimization period and the prediction simulation optimization period; Acquire the simulation data change trend of the second simulation data change curve, and determine the distance relationship between the simulation data change trend and the corresponding standard data line, wherein the distance relationship includes a close relationship and a far relationship; When the relationship is far away, the mean value of the data difference between the predicted data corresponding to the second simulation data change curve and the corresponding standard data line is calculated; When the relationship is close, the predicted closeness between the simulation data change trend corresponding to the second simulation data change curve and the corresponding standard data line is calculated, and compared with the actual closeness before optimization to obtain the closeness difference; Generate a second simulation optimization coefficient of the second prediction optimization strategy for the remaining prediction data according to the data difference mean and the proximity difference; Generate a compensation coefficient according to the predicted optimization cost of the second prediction optimization strategy of the link to be optimized; Generate a simulation optimization coefficient according to the first simulation optimization coefficient, the second simulation optimization coefficient and the compensation coefficient; Pre-set the simulation optimization coefficient threshold; If the simulation optimization coefficient is less than the simulation optimization coefficient threshold, a correction instruction corresponding to the first prediction optimization strategy is generated.

[0044] In this embodiment, the inspection duration is obtained by averaging the required optimization durations of historical preferred optimization strategies. The historical preferred optimization strategy refers to a strategy with better optimization effect among the historical optimization strategies. The required optimization duration is the time required for the historical operating conditions of the links to be optimized to meet the historical operating goals.

[0045] In this embodiment, the predicted data refers to other data in the predicted to-be-optimized link except for the predicted to-be-optimized data.

[0046] In this embodiment, the simulation optimization period refers to the length of time required to predict the simulated changes of the data to be optimized to the corresponding standard data, and the simulation optimization period is less than the inspection time. The predicted simulation optimization period refers to the length of time required to predict the standard data corresponding to the simulated change value of the data to be optimized, and the predicted simulation optimization period is greater than the inspection time.

[0047] In this embodiment, the correction instruction refers to extracting the predicted data to be optimized that needs to be optimized again to fine-tune the first prediction optimization strategy. If the number of predicted data that needs to be optimized increases, the first prediction optimization strategy is newly formulated and simulated until the simulation results meet the simulation optimization coefficient threshold, and the corresponding second prediction optimization strategy is determined.

[0048] In this embodiment, a simulation optimization coefficient is obtained by comprehensively evaluating the optimization effects of the predicted data to be optimized and the remaining predicted data in the simulation model. The optimization effect of the first prediction optimization strategy is accurately evaluated based on the simulation optimization coefficient, and the corresponding optimization strategy is discovered and adjusted in time to improve the overall operational efficiency of the port.

[0049] In some embodiments of the present application, the calculation formula of the first simulation optimization coefficient is: ; Among them, H1 is the first simulation optimization coefficient, h1 is the simulation optimization conversion coefficient of the simulation optimization period, u1 is the number of prediction data to be optimized that have intersections within the test time in the current prediction to be optimized link, T0 is the test time, This is the simulation optimization period for the z1th predicted data to be optimized. is the weight coefficient of the z1th prediction data to be optimized, h2 is the simulation optimization conversion coefficient of the prediction simulation optimization period, u2 is the number of prediction data to be optimized that have no intersection within the test time in the current prediction to be optimized link, The prediction simulation optimization period for the z2th prediction data to be optimized, is the weight coefficient of the z2th predicted data to be optimized; The calculation formula of the second simulation optimization coefficient is: ; Among them, H2 is the second simulation optimization coefficient, h3 is the simulation optimization conversion coefficient of the difference in the degree of proximity of the predicted data, and u3 is the number of predicted data with a close relationship. is the difference in the degree of closeness of the z3th predicted data, is the weight coefficient of the z3th predicted data, h4 is the simulation optimization conversion coefficient of the mean value of the data difference of the predicted data, u3 is the number of predicted data far away from the relationship, is the mean value of the data difference of the z4th predicted data, is the weight coefficient of the z4th predicted data; The calculation formula of the simulation optimization coefficient is: ; Among them, H0 is the simulation optimization coefficient, f0 is the compensation coefficient, f1 is the weight coefficient of the first simulation optimization coefficient, and f2 is the weight coefficient of the second simulation optimization coefficient.

[0050] In this embodiment, the simulation optimization conversion coefficients refer to coefficients of the same dimension converted from the simulation optimization period, the prediction simulation optimization period, the difference in the degree of proximity of the prediction data, and the mean of the data difference of the prediction data to the simulation optimization coefficient.

[0051] In this embodiment, when the difference between the inspection time and the simulation optimization period is greater, the corresponding first simulation optimization coefficient is greater, and vice versa. When the difference between the predicted simulation optimization period and the inspection time is greater, the corresponding first simulation optimization coefficient is smaller, and vice versa. When the difference in the degree of proximity of the predicted data is greater, the corresponding second simulation optimization coefficient is greater, and vice versa. When the mean value of the data difference of the predicted data is greater, the corresponding second simulation optimization coefficient is smaller, and vice versa.

[0052] In this embodiment, when the prediction optimization cost is larger, the compensation coefficient is smaller, and vice versa. The value range of the compensation coefficient is (0.8, 1.2).

[0053] In this embodiment, when the first simulation optimization coefficient, the second simulation optimization coefficient and the compensation coefficient are larger, the corresponding simulation optimization coefficient is larger, that is, the corresponding optimization effect of the first prediction optimization strategy is better.

[0054] In some embodiments of the present application, determining whether there is a conflict between second prediction optimization strategies of multiple prediction links to be optimized includes: Obtain the second prediction optimization strategy of all prediction links to be optimized for the current monitoring node, and extract the optimization target, optimization period, and optimization constraint conditions of each second prediction optimization strategy; A conflict analysis is performed on the optimization objectives, optimization time periods, and optimization constraints of different second prediction optimization strategies to obtain conflicting second prediction optimization strategies and prediction links to be optimized of the second prediction optimization strategies.

[0055] In some embodiments of the present application, if present, calculating the priority coefficient includes: Obtain each predicted link to be optimized that has conflicts, and randomly select one predicted link to be optimized as the target link; Obtain the historical evaluation log of the target link, obtain multiple historical comprehensive evaluation values ​​of the target link according to the historical evaluation log, and calculate the difference of the historical comprehensive evaluation values; Screen out the historical comprehensive evaluation value differences with increasing differences, and construct the historical comprehensive evaluation value difference sequence of the target link; Determine the corresponding adjacent preset time nodes according to the multiple historical evaluation nodes in the historical comprehensive evaluation value difference sequence, and obtain the historical comprehensive evaluation values ​​of other operation links according to the adjacent preset time nodes; According to the time sequence of the historical evaluation nodes of the historical comprehensive evaluation value difference sequence of the target link, the historical comprehensive evaluation value difference sequence of the corresponding time nodes of other operation links is constructed; Calculate the sequence similarity between the historical comprehensive evaluation value difference sequence of each operating link and the historical comprehensive evaluation value difference sequence of the target link. If the sequence similarity is greater than the preset similarity threshold, determine the corresponding operating link as the associated link of the target link, and calculate the correlation coefficient between the associated link and the target link. Generate a priority coefficient of a target link according to the number of associated links, the associated coefficient and the weight coefficient of the corresponding associated link; Generate the priority coefficient of each conflicting prediction link to be optimized in turn.

[0056] In this embodiment, the historical comprehensive evaluation value refers to the historical comprehensive evaluation value obtained by weighting the historical evaluation values ​​of multiple evaluation indicators in the corresponding historical evaluation log, which is used to comprehensively evaluate the operation effect of the corresponding link.

[0057] In this embodiment, the historical comprehensive evaluation value differences with increasing differences are differences between multiple historical comprehensive evaluation values ​​and the same historical comprehensive evaluation value, and the differences show an increasing trend.

[0058] In this embodiment, the adjacent preset time node refers to the next adjacent time node of each historical evaluation node of the target link. The historical comprehensive evaluation values ​​of other operating links are collected according to the adjacent preset time nodes to evaluate the influence of the historical operating effect of the target link on the historical operating effects of other operating links.

[0059] In this embodiment, sequence similarity refers to the similarity analysis of the difference fluctuation degree of the historical comprehensive evaluation value difference in the historical comprehensive evaluation value difference sequence of the target link and the difference fluctuation degree of the historical comprehensive evaluation value difference at the corresponding time node in the historical comprehensive evaluation value difference sequence of other operating links. The greater the similarity of the difference fluctuation degree and the more numbers in the sequence, the greater the corresponding sequence similarity.

[0060] In this embodiment, as the change in the historical comprehensive evaluation value difference of the target link is smaller, the historical comprehensive evaluation value difference of the associated link corresponding time node is larger, and the corresponding correlation coefficient is larger, and vice versa.

[0061] In this embodiment, the importance of each conflicting predicted link to be optimized is evaluated by determining the associated links of each target link, thereby obtaining a priority coefficient and setting a priority, generating a port operation optimization strategy, and ensuring coordinated and efficient operation of the port.

[0062] In some embodiments of the present application, generating a port operation optimization strategy according to the priority coefficient and the second prediction optimization strategy includes: Setting the optimization order of the corresponding second prediction optimization strategy according to the priority coefficient of the conflicting prediction link to be optimized; The port operation optimization strategy of the current monitoring node is generated according to the optimization sequence and the second prediction optimization strategy for predicting the link to be optimized.

[0063] 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 port operation optimization method based on the Internet of Things and big data intelligent analysis, characterized in that: include: Based on the IoT gateway, various data in the port operation process of the current monitoring node are collected, pre-processed, and a standard data set is constructed based on the pre-processed data. Perform feature mining and analysis on the data in the standard data set, determine the predicted link to be optimized and the predicted data to be optimized of the current monitoring node according to the feature mining and analysis results, and generate a first prediction optimization strategy for the corresponding predicted link to be optimized in combination with a preset optimization model; Simulating the first prediction optimization strategy, generating a simulation optimization coefficient according to the simulation result, and judging whether to modify the first prediction optimization strategy according to the simulation optimization coefficient, and if so, obtaining a second prediction optimization strategy; Determine whether there is a conflict between the second prediction optimization strategies of multiple prediction links to be optimized. If so, calculate the priority coefficient and generate the port operation optimization strategy based on the priority coefficient and the second prediction optimization strategy.

2. The port operation optimization method based on the Internet of Things and big data intelligent analysis as claimed in claim 1 is characterized in that: Construct standard data sets based on various types of preprocessed data, including: Obtain the port operation business in the current monitoring cycle and divide each port operation business into multiple operation links; Preprocessing various types of data, including unifying time units, removing outliers, filling missing values, and data cleaning; Obtain the data sources and data attributes of various types of preprocessed data, classify various types of data according to the data sources and data attributes, and obtain internal and external data of different operation links; Build standard data sets from internal and external data across multiple operations.

3. The port operation optimization method based on the Internet of Things and big data intelligent analysis as claimed in claim 2 is characterized in that: According to the feature mining and analysis results, the predicted optimization links and data to be optimized for the current monitoring node are determined, including: Pre-set multiple evaluation indicators for each operation link; Obtain the historical monitoring log of each internal data and external data in the standard data set, obtain the historical change characteristics of each internal data and external data in the historical monitoring log, and evaluate the correlation between the historical change characteristics and the historical evaluation values ​​of multiple evaluation indicators of the corresponding operation links; Setting the internal data whose correlation degree is greater than a preset first correlation degree threshold as the internal correlation data corresponding to the evaluation indicator; Setting the external data with a correlation degree greater than a preset second correlation degree threshold as the external correlation data corresponding to the evaluation index; Obtain demand information for each operation link, including standard time nodes and several standard data that meet the demand; Based on the actual change characteristics of the internal and external related data of all evaluation indicators of the same operation link in the current monitoring period and combined with the historical change characteristics, the predicted change characteristics of the corresponding internal and external related data before the standard time node are generated; Generate predicted internal correlation data and predicted external correlation data at a standard time node according to the predicted change characteristics and the internal correlation data and external correlation data of the current monitoring node; Comparing the predicted internal correlation data and the predicted external correlation data with the corresponding standard data respectively, to obtain a first data difference between the predicted internal correlation data and the corresponding standard data and a second data difference between the predicted external correlation data and the corresponding standard data; Generate a predicted operation status coefficient of a corresponding operation link according to a plurality of first data differences and second data differences; Pre-set operating condition coefficient thresholds; If the predicted operating condition coefficient is less than the operating condition coefficient threshold, the corresponding operating link is set as the predicted link to be optimized; The internal associated data whose first data difference is greater than the preset difference threshold and the external associated data whose second data difference is greater than the preset difference threshold in the predicted optimization link are set as the predicted data to be optimized.

4. The port operation optimization method based on the Internet of Things and big data intelligent analysis as claimed in claim 3 is characterized in that: Generate a predicted operation status coefficient of a corresponding operation link according to a plurality of first data differences and second data differences, including: Comprehensively analyzing the first data differences corresponding to the predicted internal correlation data of the same evaluation indicator in each operation link, and screening out the predicted internal correlation data whose first data difference is greater than a preset difference threshold; Comprehensively analyzing the second data differences corresponding to the predicted external related data of the same evaluation indicator in the operation link, and screening out the predicted external related data whose second data difference is greater than a preset difference threshold; Generate the prediction difference coefficient of the corresponding evaluation index at the standard time node according to the screened prediction internal correlation data and prediction external correlation data; The calculation formula of the prediction difference coefficient is: ； Among them, Y is the prediction difference coefficient, y0 is the difference conversion coefficient, x1 is the first weight coefficient, n1 is the number of predicted internal correlation data screened out, and N1 is the total number of predicted internal correlation data of the current evaluation index. is the value of the ith first data difference greater than the preset difference threshold, d1i is the weight coefficient of the ith predicted internal correlation data, x2 is the second weight coefficient, n2 is the number of screened predicted external correlation data, N2 is the total number of predicted external correlation data of the current evaluation index, is the value of the s-th second data difference value greater than the preset difference threshold value, and d2s is the weight coefficient of the s-th predicted external correlation data; Generate the predicted operating status coefficient of the corresponding operating link according to the predicted difference coefficient of all evaluation indicators in each operating link at the standard time node; The calculation formula of the predicted operating condition coefficient is: ； Among them, Y2 is the predicted operating status coefficient, y2 is the operating status conversion coefficient, m is the total number of evaluation indicators of the current operating link, Y1v is the predicted difference coefficient of the vth evaluation indicator, and bv is the weight coefficient of the vth evaluation indicator.

5. The port operation optimization method based on the Internet of Things and big data intelligent analysis as claimed in claim 4 is characterized in that: The first prediction optimization strategy is simulated, including: Based on the historical optimization logs of the predicted links to be optimized, a preset optimization model corresponding to the predicted links to be optimized is constructed; Inputting the prediction data to be optimized of each prediction link to be optimized into the corresponding preset optimization model to obtain the first prediction optimization strategy of the corresponding prediction link to be optimized; Determine the location information and structural information of the equipment involved in the predicted link to be optimized, lock the location of the preset panoramic map, and generate a static simulation scene corresponding to the predicted link to be optimized based on the locked location information; Acquire multiple dynamic nodes of the first prediction optimization strategy for the prediction link to be optimized; According to the predicted dependency relationship between the internal associated data and the external associated data of the link to be optimized, the dynamic dependency characteristics of the internal associated data and the external associated data under each dynamic node are established; Establish the dynamic operation characteristics of the corresponding equipment at each dynamic node according to the operation relationship between the involved equipment; According to the dynamic dependency relationship and the dynamic operation relationship under the same multiple dynamic nodes, a position connection is established with the static simulation scene, and a simulation model of the first prediction optimization strategy corresponding to the prediction link to be optimized is constructed.

6. The port operation optimization method based on the Internet of Things and big data intelligent analysis as claimed in claim 5 is characterized in that: Generate simulation optimization coefficients based on simulation results, including: Preset the test time of the first prediction optimization strategy of the link to be optimized, and set the data collection node according to the preset time interval; Collect simulation internal correlation data and simulation external correlation data in the corresponding simulation model according to the data collection node, and map them to the corresponding inspection duration to obtain a simulation data change curve diagram, wherein the simulation data change curve diagram includes a plurality of first simulation data change curves for predicting data to be optimized and a second simulation data change curve for predicting data; Generate a standard data line from multiple standard data in the predicted link to be optimized, and map it to the corresponding simulation data change curve diagram, compare the standard data line with the corresponding first simulation data change curve, and determine whether there is an intersection between the first simulation data change curve and the corresponding standard data line within the inspection time; If it exists, the simulation optimization period is determined according to the initial time node where the intersection exists; If not, obtain the simulation data change trend corresponding to the first simulation data change curve, perform trend extrapolation according to the simulation data change trend until there is an intersection with the corresponding standard data line, obtain the predicted time node with the intersection, and determine the predicted simulation optimization period; Generate a first simulation optimization coefficient of the first prediction optimization strategy for predicting the data to be optimized according to the simulation optimization period and the prediction simulation optimization period; Acquire the simulation data change trend of the second simulation data change curve, and determine the distance relationship between the simulation data change trend and the corresponding standard data line, wherein the distance relationship includes a close relationship and a far relationship; When the relationship is far away, the mean value of the data difference between the predicted data corresponding to the second simulation data change curve and the corresponding standard data line is calculated; When the relationship is close, the predicted closeness between the simulation data change trend corresponding to the second simulation data change curve and the corresponding standard data line is calculated, and compared with the actual closeness before optimization to obtain the closeness difference; Generate a second simulation optimization coefficient of the second prediction optimization strategy for the remaining prediction data according to the data difference mean and the proximity difference; Generate a compensation coefficient according to the predicted optimization cost of the second prediction optimization strategy of the link to be optimized; Generate a simulation optimization coefficient according to the first simulation optimization coefficient, the second simulation optimization coefficient and the compensation coefficient; Pre-set the simulation optimization coefficient threshold; If the simulation optimization coefficient is less than the simulation optimization coefficient threshold, a correction instruction corresponding to the first prediction optimization strategy is generated.

7. The port operation optimization method based on the Internet of Things and big data intelligent analysis as claimed in claim 6 is characterized in that: The calculation formula of the first simulation optimization coefficient is: ； Among them, H1 is the first simulation optimization coefficient, h1 is the simulation optimization conversion coefficient of the simulation optimization period, u1 is the number of prediction data to be optimized that have intersections within the test time in the current prediction to be optimized link, T0 is the test time, This is the simulation optimization period for the z1th predicted data to be optimized. is the weight coefficient of the z1th prediction data to be optimized, h2 is the simulation optimization conversion coefficient of the prediction simulation optimization period, u2 is the number of prediction data to be optimized that have no intersection within the test time in the current prediction to be optimized link, The prediction simulation optimization period for the z2th prediction data to be optimized, is the weight coefficient of the z2th predicted data to be optimized; The calculation formula of the second simulation optimization coefficient is: ； Among them, H2 is the second simulation optimization coefficient, h3 is the simulation optimization conversion coefficient of the difference in the degree of proximity of the predicted data, and u3 is the number of predicted data with a close relationship. is the difference in the degree of closeness of the z3th predicted data, is the weight coefficient of the z3th predicted data, h4 is the simulation optimization conversion coefficient of the mean value of the data difference of the predicted data, u3 is the number of predicted data far away from the relationship, is the mean value of the data difference of the z4th predicted data, is the weight coefficient of the z4th predicted data; The calculation formula of the simulation optimization coefficient is: ； Among them, H0 is the simulation optimization coefficient, f0 is the compensation coefficient, f1 is the weight coefficient of the first simulation optimization coefficient, and f2 is the weight coefficient of the second simulation optimization coefficient.

8. The port operation optimization method based on the Internet of Things and big data intelligent analysis as claimed in claim 7 is characterized in that: Determine whether there is a conflict between the second prediction optimization strategies of multiple prediction links to be optimized, including: Obtain the second prediction optimization strategy of all prediction links to be optimized for the current monitoring node, and extract the optimization target, optimization period, and optimization constraint conditions of each second prediction optimization strategy; A conflict analysis is performed on the optimization objectives, optimization time periods, and optimization constraints of different second prediction optimization strategies to obtain conflicting second prediction optimization strategies and prediction links to be optimized of the second prediction optimization strategies.

9. The port operation optimization method based on the Internet of Things and big data intelligent analysis as claimed in claim 8, characterized in that: If present, calculate the priority coefficient, including: Obtain each predicted link to be optimized that has conflicts, and randomly select a predicted link to be optimized as the target link; Obtain the historical evaluation log of the target link, obtain multiple historical comprehensive evaluation values ​​of the target link according to the historical evaluation log, and calculate the difference of the historical comprehensive evaluation values; Screen out the historical comprehensive evaluation value differences with increasing differences, and construct the historical comprehensive evaluation value difference sequence of the target link; Determine the corresponding adjacent preset time nodes according to the multiple historical evaluation nodes in the historical comprehensive evaluation value difference sequence, and obtain the historical comprehensive evaluation values ​​of other operation links according to the adjacent preset time nodes; According to the time sequence of the historical evaluation nodes of the historical comprehensive evaluation value difference sequence of the target link, the historical comprehensive evaluation value difference sequence of the corresponding time nodes of other operation links is constructed; Calculate the sequence similarity between the historical comprehensive evaluation value difference sequence of each operating link and the historical comprehensive evaluation value difference sequence of the target link. If the sequence similarity is greater than the preset similarity threshold, determine the corresponding operating link as the associated link of the target link, and calculate the correlation coefficient between the associated link and the target link. Generate a priority coefficient of a target link according to the number of associated links, the associated coefficient and the weight coefficient of the corresponding associated link; Generate the priority coefficient of each conflicting prediction link to be optimized in turn.

10. The port operation optimization method based on the Internet of Things and big data intelligent analysis as claimed in claim 9, characterized in that: Generate a port operation optimization strategy based on the priority coefficient and the second prediction optimization strategy, including: Setting the optimization order of the corresponding second prediction optimization strategy according to the priority coefficient of the conflicting prediction link to be optimized; The port operation optimization strategy of the current monitoring node is generated according to the optimization sequence and the second prediction optimization strategy for predicting the link to be optimized.

Images