CN-122023095-A - SARIMA prediction-based inland river navigation draft optimization control method for draft overrun ship

CN122023095ACN 122023095 ACN122023095 ACN 122023095ACN-122023095-A

Abstract

According to the SARIMA prediction-based draught overrun inland navigation draught optimization management and control method, a multisource time sequence data set is obtained and preprocessed to obtain an input time sequence set, the input time sequence set is processed through a constructed SARIMA model to obtain a water level prediction sequence, the constructed ship dynamic draught model is processed based on the water level prediction sequence to obtain a ship dynamic draught prediction sequence, a draught safety margin judgment operator is constructed to evaluate draught safety margin and classify risks, finally a draught optimization management strategy is solved through a draught optimization management model and issued to form a prediction-calculation-evaluation-optimization-feedback ship draught control closed loop chain, and forward recognition and dynamic regulation of the draught overrun draught safety margin of the ship are achieved through coupling of the SARIMA time sequence prediction with the ship dynamic draught calculation and the draught optimization management depth, so that the utilization efficiency and the ship loading efficiency of inland navigation channels are greatly improved on the premise of guaranteeing navigation safety.

Inventors

LIU CHAO
LIU JINGXIAN
TANG CHENGGANG
ZHAO YUAN

Assignees

武汉理工大学三亚科教创新园

Dates

Publication Date: 20260512
Application Date: 20260326

Claims (10)

1. The method for optimizing and controlling the inland river navigation draft of the draught overrun ship based on SARIMA prediction is characterized by comprising the following steps of: step S1, acquiring hydrological data, channel condition data and navigation data of a target inland navigation section, and constructing a multisource time sequence data set of inland navigation of the draught overrun ship; S2, preprocessing a multisource time sequence data set to obtain an input time sequence set; S3, constructing an SARIMA model for predicting the water level of the target inland river navigation section, and obtaining a water level prediction sequence in a future time domain based on an input time sequence set; S4, decomposing the dynamic rich water depth into a plurality of measurable sub-items, constructing a ship dynamic draft model based on the water level prediction sequence and the dynamic rich water depth, and predicting the ship dynamic draft prediction sequence in the time domain by the ship dynamic draft model; S5, defining a comprehensive minimum control limit value of a target inland navigation section at a certain prediction moment, constructing a draft safety margin judging operator with a ship dynamic draft predicted value at a corresponding moment, and obtaining a ship draft safety margin assessment result and a risk classification result based on the draft full margin judging operator; And S6, constructing a draft optimization management model, solving the draft optimization management model based on a ship dynamic draft prediction sequence and a ship draft safety margin evaluation result to obtain a draft optimization management strategy, and transmitting the draft optimization management strategy to a ship and shore-based dispatching system.
2. The SARIMA prediction-based inland navigation draft optimization management and control method for a draft overrun ship according to claim 1, wherein the hydrographic weather data comprise river reach water level, flow, water depth, flow velocity, wind speed and wind direction, the channel condition data comprise channel maintenance water depth, riverbed annual accumulation, bridge zone control water depth and navigation building scale, and the navigation data of the draft overrun ship comprise ship loading capacity, ballast water configuration, fore draft, aft draft and/or middle draft, trim and trim parameters, speed, course, track and ship size parameters.
3. The method for optimizing and controlling inland river navigation draft of a draught overrun ship based on SARIMA prediction according to claim 1, wherein the specific steps of the step S2 include: Denoising hydrological data, channel condition data and navigation data of the draught overrun ship, and deleting meaningless records, abnormal values and repeated data; Carrying out error correction processing on data with symbol errors, coding errors and time reversal; format standardization is carried out on various data through unified units and decimal places; Adding a time stamp and a space stamp to each record, identifying repeated records by taking the time stamp and the space stamp as joint indexes, and deleting redundant records; interpolation and alignment are carried out on the missing data by utilizing adjacent time points or adjacent observation values, and a sequence with continuous time and complete data is obtained; And performing normalization or standardization processing on the key variables to construct an input time sequence set under a unified time axis.
4. The method for optimizing and controlling inland river navigation draft of a draught overrun ship based on SARIMA prediction according to claim 1, wherein the step of constructing the SARIMA model in the step S3 is as follows: taking a month-by-month minimum water level sequence of a water level station along a ship sailing plan as a modeling object, removing abnormal values and repeated records, and unifying water level units and time scales to form a water level time sequence which is ordered according to months; STL decomposition is adopted to carry out seasonal trend decomposition on the water level time sequence, so as to obtain seasonal components, trend components and residual components; ADF test is used for judging the stability of the water level time sequence, and first-order constant difference and necessary season difference are carried out on the non-stable sequence until the stability test is passed under the preset significance level; on the basis of obtaining a stable sequence, drawing an auto-correlation function ACF diagram and a partial auto-correlation function PACF diagram, determining an auto-regression order p according to the PACF significant tail cutting position, determining a moving average order q according to the ACF significant tail cutting position, and determining SARIMA model parameters by combining a seasonal period s and a seasonal difference order D; Introducing flow related exogenous variables into the SARIMA model, and describing the influence direction and strength of the flow on the water level predicted value by using the month-by-month average flow and the variation thereof for the last years as exogenous input through regression coefficients; and comparing the combinations of different orders based on AIC/BIC and residual error test, and selecting an SARIMA model with optimal fitting goodness and generalization performance as a target water level prediction model.
5. The method for optimizing and controlling inland river navigation draft of a draught overrun ship based on SARIMA prediction according to claim 4, wherein the flow-related exogenous variables comprise a current month average flow, a difference value between the current month average flow and a previous month average flow, and a difference value between the current month average flow and a previous two month average flow, and regression coefficients of the exogenous variables are used for reflecting positive or negative influence and influence intensity of corresponding flow characteristics on the corrected water level.
6. The method for optimizing and controlling inland river navigation draft of a draught overrun ship based on SARIMA prediction according to claim 1, wherein the calculation formula of the ship dynamic draft model is as follows: Wherein the method comprises the steps of For the dynamic draft value of the vessel, Is the water depth of the channel and is equal to the water depth of the channel, Is a predicted value of the water level, The dynamic abundant water depth of the ship is decomposed into a plurality of measurable sub-items, and the specific calculation formula is as follows: Wherein the method comprises the steps of For the draft variation caused by the consumption of the fuel and fresh water of the ship, The draft change amount caused by the ship attitude change is used, The draft change amount caused by the sinking amount of the ship, The silt is prepared for the river bed.
7. The optimization and control method for inland navigation draft of a draught overrun ship based on SARIMA prediction according to claim 6, wherein draft variation caused by ship fuel and fresh water consumption is obtained by calculating draft variation rates corresponding to fuel consumption in unit time and unit speed according to ship speed and draft records at current time and historical time and combining future planned speeds; The draft variable quantity caused by the ship attitude change is obtained through calculation of the difference value between the transverse inclination angle and the longitudinal inclination angle of the current moment and the future predicted moment; the draft variation caused by the ship sinking amount is calculated based on the current water depth draft ratio, the ship square coefficient and the future planned navigational speed by using a shallow water navigational speed sinking underwater sinking empirical formula; the river bed dredging amount is converted into the predicted moment through the annual river bed dredging thickness acquired by the investigation channel management department, and the river bed dredging thickness at the future moment is obtained.
8. The method for optimizing and controlling inland river navigation draft of a draught overrun ship based on SARIMA prediction according to claim 1, wherein the calculation formula of the comprehensive minimum control limit value is as follows: Wherein the method comprises the steps of At the prediction time for the target inland navigation section Is used to control the overall minimum control limit value of (c), For maximum allowable draft of the ship structure, For maximum allowable draft for a navigable channel, The draft is limited for the bridge region, The draft is safely controlled; the expression of the draft safety margin judging operator is as follows: Wherein the method comprises the steps of To predict time of day Is a draft safety margin determination value of (c), To predict time of day Is a predicted value of the dynamic draft of the ship; and obtaining a minimum draft safety margin judgment value in a given period based on the draft safety margin judgment operator, and comparing the minimum draft safety margin judgment value with a preset threshold value to obtain a risk classification result comprising safety, attention, early warning and high risk.
9. The method for optimizing and controlling the inland river navigation draft of the over-draft ship based on SARIMA prediction according to claim 1, wherein the specific steps of the step S6 are as follows: taking cargo hold loading capacity, ballast water amount of each ballast tank, ship attitude control, and partial river reach ship navigational speed and limited river reach passing time window as decision variables; setting constraint conditions, wherein the constraint conditions comprise draft safety constraint that the dynamic draft value of the ship does not exceed the comprehensive minimum control limit value at each prediction moment, and hold capacity and loading constraint that the capacities of each hold and ballast tank are not exceeded; The method comprises the steps of constructing an objective function by taking the optimization objective of maximizing the effective cargo capacity and considering the fuel consumption cost under the premise of meeting the draft safety constraint; And solving the optimization problem by adopting a mixed integer linear programming or heuristic algorithm, obtaining a draft optimization management strategy, and transmitting the draft optimization management strategy to the ship and shore-based dispatching system.
10. The method for optimizing and controlling inland river navigation draft of a draught overrun ship based on SARIMA prediction according to claim 9, wherein the expression of the objective function is: Wherein the method comprises the steps of Is a navigation speed vector of a ship in a river segment, For the cargo hold loading capacity, For a cargo hold load of number n, For each amount of ballast tank ballast water, The ballast water amount of the m-size ballast tank, As the pitch control parameter, a control parameter of the pitch, In order to trade-off the coefficients, For a fuel consumption estimation function associated with a cruise configuration, Is the total amount of ballast water.

Description

SARIMA prediction-based inland river navigation draft optimization control method for draft overrun ship Technical Field The invention relates to the technical field of water traffic safety, in particular to a method for optimizing and controlling inland river navigation draft of a draught overrun ship based on SARIMA prediction. Background The "inland navigation" refers to navigation activities performed by a ship on inland waters such as rivers, water networks, lakes and the like in a country or region, and relative to offshore navigation, inland navigation mostly occurs in water areas upstream and downstream of Yangtze river, zhujiang river, xijiang river, beijing Hangzhou canal, harbor basin, ship lock and the like in China. The term "draft overrun ship" refers to a ship whose actual draft exceeds the maximum allowable draft specified by the corresponding safety control limit or regulations under the conditions of a given channel or section, and generally includes, but is not limited to, 1) when the ship is loaded with cargo and/or ballast water, the actual draft exceeds the maximum allowable draft specified by the ship itself or the facilities of the channel, bridge, gate dam, etc., 2) when the actual draft of the ship does not reach the rigid maximum allowable draft limit, but exceeds the safety control draft set by the inland river or the specific section, and 3) the ship draft distribution is abnormal, including at least the partial overrun of the bow draft, stern draft, or middle draft, resulting in the shortage of the overall or partial safety margin of the ship. With the continuous increase of domestic shipping quantity and the continuous increase of the number of large and heavy load ships, the occurrence frequency of the over-draft ships in the inland waterways is in a rising trend year by year. On one hand, in order to reduce the transportation cost and improve the income per voyage, part of voyage enterprises have heavy-load operation tendency in loading management, so that the draft of the ship is easy to approach or even exceed the safety control limit value in a specific water level or shallow-groove river section, and on the other hand, the safety draft margin of the ship is further compressed under different river sections and different working conditions under the influence of factors such as seasonal change of the water level, channel siltation, bridge area control water depth, navigation building scale and the like, so that the draft overrun risk is continuously accumulated and exposed. In addition, the management level of some ships in ballast water allocation, trim and heel control, cargo allocation and the like is insufficient, the occurrence probability of excessive draft and abnormal draft distribution is objectively increased, and the factors act together, so that the excessive draft ships present development situations of increased quantity, expanded distribution range, increased supervision difficulty and increased potential safety hazards in a inland river shipping system in China. At present, certain researches are carried out on the perception and analysis of inland water levels and navigation environments, for example, the inland multi-site daily water level prediction method improves the multi-site water level prediction precision in a clustering-modeling-first mode, the inland navigation water area classification method achieves classification of water areas with different navigation environments through clustering analysis, and the bridge monitoring and early warning method and the ship navigation safety control method respectively improve the water traffic safety from the aspects of bridge area safety and multi-source data fusion early warning. However, the above-mentioned techniques focus on the recognition, classification and early warning of the water level or the navigation environment itself, and are not yet coupled with the draft control depth of a specific ship, and are difficult to be directly used for the fine dynamic management of the over-draft ship. Conventional methods of ship draft management are typically based on "water depthThe simple relation of the fixed rich water depth is that the geometric water depth from the water surface to the riverbed is taken as the available water depth, one 'rich water depth' is given by an empirical fixed value or according to the draft proportion, and the allowable draft value is reversely deduced. In this method, the rich water depth is generally not subdivided into constituent items such as ship sinking amount, attitude change, and riverbed fluctuation uncertainty, and is basically not changed in real time with flow rate, draft, loading condition, and time. Accordingly, draft checking is completed statically in the loading or off-berthing stage, so that the influence of future water level change on draft safety margin is difficult to reflect in time, and the time-varying characteristics of dynamic draft of the ship in