A dock early warning system and method based on meteorological forecast model

Abstract

The present invention relates to the field of port management technology, and in particular to a terminal early warning system and method based on a meteorological forecast model. The content includes: collecting meteorological data, cleaning and normalizing the meteorological data; constructing a meteorological forecast model to obtain meteorological forecast results; evaluating the impact of meteorological forecast results on terminal production operations, and generating corresponding early warning information; based on the early warning information, adjusting the terminal's production operation plan, and communicating the adjusted terminal's production operation plan to relevant personnel; and storing and visualizing meteorological data, meteorological forecast results, and the adjusted terminal's production operation plan. It solves the technical problems that the existing technology lacks timeliness and accuracy in responding to meteorological changes, the acquisition and processing efficiency of data is low, there is a lack of effective data processing and forecasting models, and there is a lack of automation means in information communication and execution, and there is a risk of information lag and transmission errors.

Description

Technical Field

[0001] The present invention relates to the technical field of port management, and in particular to a dock early warning system and method based on a meteorological forecast model. Background Art

[0002] In recent years, with the rapid development of the global economy and the continuous growth of international trade, as an important logistics hub, the production management and operation efficiency of ports are crucial to the smooth operation of the entire supply chain. As one of the important ports in northern my country, Yantai Port has realized digital management in production management and operation to meet the needs of port business management and internal operations. However, with the continuous increase in port business volume and the complexity of the operating environment, traditional production scheduling and supervision methods have gradually revealed their limitations, especially when dealing with changing weather conditions, there is a lack of efficient and intelligent solutions.

[0003] The port operation environment is complex and changeable. Meteorological factors such as strong winds, heavy rains, and high temperatures have a direct impact on the safety and efficiency of terminal operations. At present, production managers at various terminals mainly rely on video surveillance, on-site inspections, and shift workload analysis to discover and solve on-site operation problems. This method is not only highly dependent on people, but also has the problem of poor timeliness. It is unable to respond to sudden weather changes in a timely and accurate manner, resulting in reduced operation safety and production efficiency. In addition, some terminals still use manual accounting and manual drawing to collect and record data, which is cumbersome and complex, inefficient, and difficult to achieve real-time sharing and efficient use of information.

[0004] In order to solve these problems and improve the intelligence and information level of terminal operations, the Production Business Department proposed the goal of building "one map for production efficiency, one map for equipment management, one map for personnel management, one map for warehouse management, and one map for video surveillance". This construction goal aims to achieve the dispatching, commanding and supervision of on-site production operations at the terminal through information and intelligent means, and comprehensively improve the safety, efficiency and management level of terminal operations.

[0005] There are at least the following technical problems in the existing technology: the existing technology lacks timeliness and accuracy in responding to meteorological changes, and it is difficult to respond to sudden meteorological changes in a timely manner, resulting in reduced operational safety and production efficiency, low efficiency in data acquisition and processing, and unable to meet the needs of modern production scheduling. There is a lack of effective data processing and prediction models, resulting in delayed or inaccurate early warning information, and unable to provide a scientific basis for production scheduling. There is a lack of automated means in information communication and execution, and there is a risk of information lag and transmission errors, which affects the timeliness and accuracy of operations. Summary of the invention

[0006] The present invention provides a terminal early warning system and method based on a meteorological forecast model to solve the problems that the existing technology lacks timeliness and accuracy in responding to meteorological changes, makes it difficult to respond to sudden meteorological changes in a timely manner, leads to reduced operation safety and production efficiency, low efficiency in data acquisition and processing, cannot meet the needs of modern production scheduling, lacks effective data processing and forecasting models, leads to delayed or inaccurate early warning information, cannot provide a scientific basis for production scheduling, lacks automated means in information communication and execution, and has the risk of information lag and transmission errors, affecting the timeliness and accuracy of operations.

[0007] The present invention provides a dock early warning system and method based on a meteorological forecast model, which specifically includes the following technical solutions:

[0008] A dock early warning method based on a meteorological forecast model comprises the following steps:

[0009] S1. Collect meteorological data, clean and normalize the meteorological data, build a meteorological forecast model, and obtain meteorological forecast results;

[0010] S2. Evaluate the impact of weather forecast results on terminal production operations and generate corresponding warning information; adjust the terminal's production operation plan based on the warning information, and communicate the adjusted terminal's production operation plan to relevant personnel; at the same time, store and visualize meteorological data, weather forecast results and the adjusted terminal's production operation plan.

[0011] Preferably, the S1 specifically includes:

[0012] The weather forecast model adopts a dual attention mechanism long short-term memory neural network, which adds time attention mechanism and feature attention mechanism on the basis of the long short-term memory neural network.

[0013] Preferably, the S1 specifically includes:

[0014] In the implementation process of the dual attention mechanism long short-term memory neural network, time attention weights and feature attention weights are introduced to dynamically adjust the weights of different time steps and different features.

[0015] Preferably, the S1 specifically includes:

[0016] Based on the dual attention mechanism of temporal attention mechanism and feature attention mechanism, nonlinear transformation and noise terms are introduced to obtain the weather forecast results.

[0017] Preferably, the S2 specifically includes:

[0018] An early warning information generation algorithm is introduced to determine whether an early warning is needed based on the meteorological forecast results, and generate corresponding early warning information.

[0019] Preferably, the S2 specifically includes:

[0020] An operation plan adjustment algorithm is introduced to adjust the production operation plan of the terminal according to the early warning information to obtain the adjusted production operation plan.

[0021] Preferably, the S2 specifically includes:

[0022] Through data processing and storage algorithms, meteorological data, meteorological forecast results and adjusted production operation plans are processed and stored, and visual data reports are generated for management personnel's reference.

[0023] A dock early warning system based on a meteorological forecast model includes the following parts:

[0024] Data acquisition module, weather forecast module, early warning module, production scheduling module, data display module, database;

[0025] The data acquisition module collects meteorological data through meteorological sensors and cleans and normalizes the meteorological data; the data acquisition module is connected to the meteorological forecast module and the database;

[0026] The meteorological forecast module builds a meteorological forecast model and obtains meteorological forecast results. The meteorological forecast module is connected with the early warning module and the database;

[0027] The early warning module evaluates the impact on the terminal's production operations based on the weather forecast results and generates corresponding early warning information. The early warning module is connected to the production scheduling module;

[0028] The production scheduling module receives early warning information, adjusts the production operation plan of the terminal, and obtains the adjusted production operation plan of the terminal. The production scheduling module is connected with the data display module and the database;

[0029] The data display module is responsible for processing and storing meteorological data, meteorological forecast results and the adjusted production operation plan of the terminal, and forming a visual data report for the reference of management personnel. The data display module is connected to the database;

[0030] Database, storing meteorological data, meteorological forecast results and adjusted production operation plans of the terminal.

[0031] The beneficial effects of the technical solution of the present invention are:

[0032] 1. By deploying multiple meteorological sensors in the dock area, real-time data such as temperature, humidity, wind speed, wind direction and rainfall are collected. The sensors transmit the data to the central control system through the wireless network. The meteorological forecast module uses the dual attention mechanism long short-term memory neural network (DA-LSTM) for meteorological forecasting, which effectively improves the accuracy and timeliness of meteorological forecasts, so that production scheduling can respond to meteorological changes more promptly and accurately.

[0033] 2. The early warning system evaluates the impact of weather forecast results on terminal production operations and generates corresponding early warning information. Early warning information includes strong wind warning and rainstorm warning, etc. The production scheduling module adjusts the operation plan according to the early warning information, such as suspending high-altitude operations and hoisting operations, and adjusting the stacking position of containers to prevent water accumulation from damaging the goods, which greatly improves the safety of operations and avoids operational accidents caused by weather changes.

[0034] 3. The early warning system reduces reliance on manual monitoring and decision-making by automatically generating early warning information and adjusting work plans. The production scheduling module automatically generates new work plans based on early warning information and communicates them to relevant personnel to ensure implementation, effectively improving the efficiency of production scheduling and supervision and reducing labor costs.

[0035] 4. The data processing and storage module is responsible for processing and storing the collected meteorological data, meteorological forecast results and production operation plans, forming visual data reports for reference by managers, and using charts, dashboards, etc. to intuitively display data to managers to help them make scientific decisions, which not only improves the efficiency of management decisions, but also enhances the transparency and traceability of data.

[0036] 5. By adopting wireless transmission technologies such as LoRa, Zigbee or NB-IoT, the stability and real-time performance of data transmission are ensured; at the same time, the data acquisition module cleans and normalizes the meteorological data to remove noise and outliers to ensure the accuracy of the data; the early warning system is designed with emphasis on the stability and reliability of each link and can operate stably under various meteorological conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a structural diagram of a dock early warning system based on a meteorological forecast model according to the present invention;

[0038] Figure 2 The present invention provides a flow chart of a dock early warning method based on a meteorological forecast model. DETAILED DESCRIPTION

[0039] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0041] The specific scheme of a dock early warning system and method based on a meteorological forecast model provided by the present invention is described in detail below with reference to the accompanying drawings.

[0042] See attached Figure 1 , which shows a structural diagram of a dock warning system based on a meteorological forecast model provided by an embodiment of the present invention, the system includes the following parts:

[0043] Data acquisition module, weather forecast module, early warning module, production scheduling module, data display module, database;

[0044] The data acquisition module receives data from meteorological sensors, obtains meteorological data, cleans and normalizes the meteorological data, removes noise and outliers, and ensures the accuracy of the data; the data acquisition module is connected to the meteorological forecast module and the database;

[0045] The meteorological forecast module builds a meteorological forecast model and obtains meteorological forecast results. The meteorological forecast module is connected with the early warning module and the database;

[0046] The early warning module evaluates the impact of the generated weather forecast results on the terminal's production operations and generates corresponding early warning information. The early warning module is connected to the production scheduling module;

[0047] The production scheduling module receives early warning information, adjusts the production operation plan of the terminal, obtains the adjusted production operation plan of the terminal, and ensures the safety of operations under adverse weather conditions. The production scheduling module is connected with the data display module and the database;

[0048] The data display module is responsible for processing and storing the collected meteorological data, meteorological forecast results and the adjusted production operation plan of the terminal, and forming a visual data report for the reference of the management personnel. The data display module is connected to the database;

[0049] The database stores the collected meteorological data, meteorological forecast results and the adjusted production operation plan of the terminal.

[0050] See attached Figure 2 , which shows a flow chart of a dock warning method based on a meteorological forecast model provided by an embodiment of the present invention, the method comprising the following steps:

[0051] S1. Collect meteorological data, clean and normalize the meteorological data, build a meteorological forecast model, and obtain meteorological forecast results;

[0052] Multiple meteorological sensors are arranged in the dock area to collect meteorological data such as temperature, humidity, wind speed, wind direction and rainfall in real time. The sensors transmit the data to the central control system through the wireless network, using wireless transmission technologies such as LoRa, Zigbee or NB-IoT to ensure the stability and real-time performance of data transmission. The sensors include temperature sensors, humidity sensors, anemometers, wind vanes and rain gauges. The data acquisition module receives the data from the meteorological sensors, cleans and normalizes the meteorological data, removes noise and outliers, and ensures the accuracy of the data; data cleaning includes removing outliers and filling missing values, and normalization converts the data to the same scale to ensure the consistency of the input data of the meteorological forecast model.

[0053] The meteorological forecast module constructs a meteorological forecast model. The meteorological forecast model adopts a dual attention mechanism long short-term memory neural network (DA-LSTM), which adds time attention mechanism and feature attention mechanism on the basis of LSTM to improve the prediction accuracy.

[0054] Introduce temporal attention weights to dynamically adjust the weights of different time steps. The formula is as follows:

[0055]

[0056] e t =tanh(W t h t +b t )

[0057] Among them, α t is the attention weight at time step t; e t is the attention scoring function at time step t; W t and b t are trainable parameters, which are the weight and bias of the temporal attention mechanism in the dual attention mechanism long short-term memory neural network; h tIt is the output obtained after inputting the input feature vector of time step t into the long short-term memory neural network; T is the prediction time range; the LSTM model controls the information flow through the input gate, forget gate and output gate. The input gate determines how much of the current input information is saved to the cell state, the forget gate determines how much information in the cell state is retained, and the output gate determines which parts of the cell state affect the final output.

[0058] Introduce feature attention weights to dynamically adjust the weights of different features. The formula is as follows:

[0059]

[0060] e f =tanh(W f x f +b f )

[0061] Among them, β f is the attention weight of feature f; e f is the attention scoring function of feature f; W f and b f are the weights and biases of the feature attention mechanism in the dual attention mechanism long short-term memory neural network; x f is the fth feature in the input feature vector; F represents the total number of features.

[0062] Taking into account time attention and feature attention, in order to further improve the accuracy of prediction, nonlinear transformation and noise terms are introduced on the basis of the dual attention mechanism to obtain the weather forecast results:

[0063]

[0064] in, It is the weather forecast result, including temperature, humidity, wind speed, wind direction, rainfall forecast value, etc.; h t，f is the LSTM output of time step t and fth feature; ∈ t is the noise term, which is used to simulate unpredictable random factors; Used to capture nonlinear relationships in input data; log(h t，f +1) is used to smooth the input data; sin(h t，f ) is used to introduce periodic components.

[0065] The derivation process of the above weather forecast formula is as follows: define the output h of LSTM t，f , represents the hidden state of time step t and the fth feature, and introduces a noise term ∈ t To simulate random disturbances, assume ∈ t Follows a Gaussian distribution with mean zero: The square root function and logarithmic function are used to perform nonlinear transformation on the LSTM output to obtain the LSTM output g(h t，f ):

[0066]

[0067] Finally, the sine function is introduced to capture the periodic component and the weather forecast formula is obtained:

[0068]

[0069] S2. Evaluate the impact of weather forecast results on terminal production operations and generate corresponding warning information; adjust the terminal's production operation plan based on the warning information, and communicate the adjusted terminal's production operation plan to relevant personnel; at the same time, store and visualize meteorological data, weather forecast results and the adjusted terminal's production operation plan.

[0070] The early warning module generates weather forecast results Evaluate its impact on the terminal's production operations and generate corresponding warning information. For example, a strong wind warning is issued when the wind speed exceeds 30km / h, and a heavy rain warning is issued when the rainfall exceeds 50mm. The warning information generation algorithm is based on the meteorological forecast results. Determine whether an early warning is needed and generate corresponding early warning information. Early warning information includes the type of warning (such as strong wind warning, rainstorm warning), warning level (such as general warning, severe warning) and response measures. Early warning rule formula:

[0071]

[0072] Among them, P wind and P rain Respectively represent the triggering status of wind speed and rainfall warnings; is the predicted wind speed value; is the predicted rainfall.

[0073] The production scheduling module receives warning information and adjusts the terminal's production operation plan to ensure operation safety under adverse weather conditions. For example, when a strong wind warning is issued, high-altitude operations and lifting operations are suspended; when a rainstorm warning is issued, the container stacking position is adjusted to prevent water accumulation from damaging the goods. The production scheduling module automatically generates a new operation plan based on the warning information and communicates the new plan to relevant personnel to ensure its implementation. The operation plan adjustment algorithm automatically adjusts the terminal's production operation plan based on the generated warning information. Operation plan adjustment formula:

[0074] S adjusted =S original ·(1-P wind ·W high -Prain ·W rain )

[0075] Among them, S adjusted is the adjusted work plan; S original is the original operation plan; W high and W rain They represent the weight coefficients of high-altitude work and rainy day work respectively.

[0076] The adjusted work plan will be notified to relevant personnel via SMS, email or other instant messaging methods to ensure the timeliness and accuracy of information transmission; electronic display screens and broadcasting systems will be set up at the terminal site to broadcast early warning information in real time to ensure that all personnel are aware of it in a timely manner and take countermeasures.

[0077] The data display module is responsible for processing and storing the collected meteorological data, meteorological forecast results, and the adjusted production operation plan of the terminal, and forming a visual data report for the reference of managers; using charts, dashboards, etc., the data is intuitively displayed to managers to help them make decisions. The data processing and storage algorithm is responsible for processing and storing the collected meteorological data, meteorological forecast results, and the adjusted production operation plan of the terminal, and generating a visual data report for the reference of managers. Data storage and visualization formula:

[0078]

[0079] Among them, D stored is the amount of data stored; is the predicted value of time step t and the fth feature; λ is the time decay coefficient; T is the prediction time range; F is the total number of features.

[0080] Through the steps of meteorological data collection, data preprocessing, meteorological forecasting, warning information generation, terminal production operation plan adjustment, operation plan communication and execution, data display and storage, a closed-loop intelligent early warning system is formed, which improves the efficiency and safety of terminal production scheduling and supervision, and ensures the safety and timeliness of terminal operations.

[0081] In summary, a dock early warning system and method based on the meteorological forecast model have been completed.

[0082] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0083] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0084] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A dock early warning method based on a meteorological forecast model, characterized in that: The following steps are involved: S1. Collect meteorological data, clean and normalize them; build a meteorological forecast model, which uses a dual attention mechanism long short-term memory neural network. On the basis of the long short-term memory neural network, the time attention mechanism and feature attention mechanism are added; in the implementation process of the dual attention mechanism long short-term memory neural network, the time attention weight is introduced to dynamically adjust the weights of different time steps. The formula is as follows: e t =tanh(W t h t +b t ) Among them, α t is the attention weight at time step t; e t is the attention scoring function at time step t; W t and b t are trainable parameters, which are the weight and bias of the temporal attention mechanism in the dual attention mechanism long short-term memory neural network; h t It is the output obtained after inputting the input feature vector of time step t into the long short-term memory neural network; T is the prediction time range; Introduce feature attention weights to dynamically adjust the weights of different features. The formula is as follows: e f =tanh(W f x f +b f ) Among them, β f is the attention weight of feature f; e f is the attention scoring function of feature f; W f and b f are the weights and biases of the feature attention mechanism in the dual attention mechanism long short-term memory neural network; x f is the fth feature in the input feature vector; F represents the total number of features; Based on the dual attention mechanism of time attention mechanism and feature attention mechanism, nonlinear transformation and noise terms are introduced to obtain the weather forecast results: in, is the meteorological forecast result; h t,f is the LSTM output of time step t and fth feature; ∈ t is the noise term; Used to capture nonlinear relationships in input data; log(h t,f +1) is used to smooth the input data; sin(h t,f ) is used to introduce periodic components; S2. Evaluate the impact of weather forecast results on terminal production operations and generate corresponding warning information; adjust the terminal's production operation plan based on the warning information, and communicate the adjusted terminal's production operation plan to relevant personnel; at the same time, store and visualize meteorological data, weather forecast results and the adjusted terminal's production operation plan.

2. The dock early warning method based on the meteorological forecast model according to claim 1 is characterized in that: The S2 specifically includes: An early warning information generation algorithm is introduced to determine whether an early warning is needed based on the meteorological forecast results, and generate corresponding early warning information.

3. The dock early warning method based on the meteorological forecast model according to claim 1 is characterized in that: The S2 specifically includes: An operation plan adjustment algorithm is introduced to adjust the production operation plan of the terminal according to the early warning information to obtain the adjusted production operation plan.

4. The dock early warning method based on the meteorological forecast model according to claim 3 is characterized in that: The S2 specifically includes: Through data processing and storage algorithms, meteorological data, meteorological forecast results and adjusted production operation plans are processed and stored, and visual data reports are generated for management personnel's reference.

5. A dock early warning system based on a meteorological forecast model, applied to the dock early warning method based on a meteorological forecast model as claimed in claim 1, characterized in that: Includes the following parts: Data acquisition module, weather forecast module, early warning module, production scheduling module, data display module, database; The data acquisition module collects meteorological data through meteorological sensors and cleans and normalizes the meteorological data; the data acquisition module is connected to the meteorological forecast module and the database; The meteorological forecast module builds a meteorological forecast model and obtains meteorological forecast results. The meteorological forecast module is connected with the early warning module and the database; The early warning module evaluates the impact on the terminal's production operations based on the weather forecast results and generates corresponding early warning information. The early warning module is connected to the production scheduling module; The production scheduling module receives early warning information, adjusts the production operation plan of the terminal, and obtains the adjusted production operation plan of the terminal. The production scheduling module is connected with the data display module and the database; The data display module is responsible for processing and storing meteorological data, meteorological forecast results and the adjusted production operation plan of the terminal, and forming a visual data report for the reference of management personnel. The data display module is connected to the database; Database, storing meteorological data, meteorological forecast results and adjusted production operation plans of the terminal.

Images