Method and device for determining sea ice and snow disaster resonance probability

Abstract

The present application relates to the technical field of sea ice and snow disaster resonance event research, in particular to a method and device for determining sea ice and snow disaster resonance probability. The method comprises: obtaining sea ice condition data of a target sea area, snow disaster data of a target region and port data in the target sea area; obtaining sea ice resonance indicators and snow disaster resonance indicators according to the sea ice condition data, the snow disaster data and the port data; obtaining sea ice resonance comprehensive indexes and snow disaster resonance comprehensive indexes according to the sea ice resonance indicators and the snow disaster resonance indicators; determining a resonance event set according to the sea ice resonance indicators, the snow disaster resonance indicators, the sea ice resonance comprehensive indexes, the snow disaster resonance comprehensive indexes and the port data; and determining a resonance probability according to the resonance event set. The present application can provide more refined data support for marine transportation and operation by determining the resonance probability of sea ice and snow disaster.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sea ice and snow disaster resonance event research, and particularly relates to a method and device for determining sea ice and snow disaster resonance probability. BACKGROUND

[0002] Affected by winter low temperature and cold wave, sea ice and snow disaster show spatial regional heterogeneity, but there is a possibility of simultaneous or successive occurrence in time, and both of them will affect different modes of transport when they occur simultaneously or successively. At present, cross-regional resource transportation, such as coal transportation, has the special nature of transporting from north to south. When the transportation is blocked or interrupted due to sea ice in a sea area and snow disaster in a region, the disaster effect will be superimposed or magnified. This disaster effect is called disaster resonance. It is of great significance to help disaster system theory research to find out whether there is a disaster resonance event between sea ice and snow disaster in history, and if there is a disaster resonance event between sea ice and snow disaster, how the resonance probability is distributed. At present, the analysis of the resonance event of sea ice and snow disaster is not sufficient. SUMMARY

[0003] The technical problem to be solved by the present application is to provide a method and device for determining sea ice and snow disaster resonance probability, so as to solve the problem that the analysis of the resonance event of sea ice and snow disaster in the prior art is not sufficient.

[0004] To solve the above technical problem, the technical scheme of the present application is as follows:

[0005] In one aspect of the present application, a method for determining sea ice and snow disaster resonance probability is provided, comprising:

[0006] obtaining sea ice ice condition data of a target sea area, snow disaster data of a target region and port data in the target sea area;

[0007] obtaining sea ice resonance indicators and snow disaster resonance indicators according to the sea ice ice condition data, the snow disaster data and the port data;

[0008] obtaining sea ice resonance comprehensive indexes and snow disaster resonance comprehensive indexes according to the sea ice resonance indicators and the snow disaster resonance indicators;

[0009] determining a resonance event set according to the sea ice resonance indicators, the snow disaster resonance indicators, the sea ice resonance comprehensive indexes, the snow disaster resonance comprehensive indexes and the port data;

[0010] determining a resonance probability according to the resonance event set.

[0011] Further, according to the sea ice resonance index and the snow disaster resonance index, a sea ice resonance comprehensive index and a snow disaster resonance comprehensive index are obtained, comprising:

[0012] The weight of the sea ice resonance index and the weight of the snow disaster resonance index are determined by using an entropy weight coefficient method and the sea ice resonance index and the snow disaster resonance index.

[0013] According to the weight of the sea ice resonance index and the weight of the snow disaster resonance index, a sea ice resonance comprehensive index and a snow disaster resonance comprehensive index are obtained.

[0014] Further, according to the sea ice resonance index and the snow disaster resonance index, a sea ice resonance comprehensive index and a snow disaster resonance comprehensive index are obtained, further comprising:

[0015] The sea ice resonance comprehensive index and the snow disaster resonance comprehensive index are subjected to edge distribution fitting calculation to obtain an optimal sea ice resonance comprehensive index and an optimal snow disaster resonance comprehensive index.

[0016] Further, according to the resonance event set, a resonance probability is determined, comprising:

[0017] According to the resonance event set, an average sea ice comprehensive resonance index and an average snow disaster comprehensive resonance index are obtained.

[0018] The resonance probability is determined by using a connection function and the average sea ice comprehensive resonance index and the average snow disaster comprehensive resonance index.

[0019] Further, the method further comprises:

[0020] According to the resonance probability and the port data, a resonance risk distribution map is obtained.

[0021] Further, the sea ice condition data comprises sea ice area and sea ice thickness, the snow disaster data comprises cumulative precipitation, negative accumulated temperature and duration, and the port data comprises name, channel, longitude and latitude.

[0022] Further, the time range in the sea ice condition data and the snow disaster data is consistent.

[0023] Further, the sea ice resonance index is average sea ice thickness and sea ice event duration, and the snow disaster resonance index is cumulative precipitation, negative accumulated temperature and event duration.

[0024] According to another aspect of the present application, a device for determining sea ice and snow disaster resonance probability is provided, comprising:

[0025] An acquisition module is configured to acquire sea ice condition data of a target sea area, snow disaster data of a target region, and port data in the target sea area, and send the data to a resonance index calculation module and an event set determination module;

[0026] The resonance index calculation module is configured to obtain sea ice resonance indexes and snow disaster resonance indexes according to the sea ice condition data, the snow disaster data, and the port data, and send the indexes to a resonance comprehensive index calculation module and the event set determination module;

[0027] The resonance comprehensive index calculation module is configured to obtain sea ice resonance comprehensive indexes and snow disaster resonance comprehensive indexes according to the sea ice resonance indexes and the snow disaster resonance indexes, and send the indexes to the event set determination module;

[0028] The event set determination module is configured to determine a resonance event set according to the sea ice resonance indexes, the snow disaster resonance indexes, the sea ice resonance comprehensive indexes, the snow disaster resonance comprehensive indexes, and the port data, and send the event set to a probability determination module;

[0029] The probability determination module is configured to determine a resonance probability according to the resonance event set.

[0030] The above scheme of the present application at least has the following beneficial effects:

[0031] The above scheme of the present application determines a resonance event set of sea ice and snow disaster according to sea ice condition data of a target sea area, snow disaster data of a target region, and port data in the target sea area, and obtains a resonance probability of sea ice and snow disaster, thereby solving the problem that the resonance event of sea ice and snow disaster is not sufficiently studied in the prior art, and providing a data basis and technical support for reasonable layout planning of a port and smooth navigation of coal across regions. BRIEF DESCRIPTION OF DRAWINGS

[0032] Figure 1 is a flowchart of a method for determining a resonance probability of sea ice and snow disaster of the present application;

[0033] Figure 2 is a main port and channel buffer zone of a target sea area;

[0034] Figure 3a is a Q-Q test chart of optimal distribution fitting of a resonance event sea ice resonance comprehensive index;

[0035] Figure 3b is a Q-Q test chart of optimal distribution fitting of a snow disaster resonance comprehensive index;

[0036] Figure 4a is a probability density chart in a Cupola function;

[0037] Figure 4b is a joint cumulative probability distribution chart in a Cupola function;

[0038] Figure 5 is the average joint return period of the resonance event of sea ice and snow disaster in the target sea area and the main port of the target sea area;

[0039] Figure 6 is the device connection diagram of the resonance probability determination device of sea ice and snow disaster. DETAILED DESCRIPTION

[0040] Exemplary embodiments of the present application will be described in greater detail below with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it is understood that the present application can be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the application to those skilled in the art.

[0041] As shown in Figure 1 , the embodiment of the present application proposes a resonance probability determination method of sea ice and snow disaster, comprising:

[0042] S1, obtaining sea ice ice condition data of a target sea area, snow disaster data of a target area and port data in the target sea area;

[0043] S2, obtaining sea ice resonance indicators and snow disaster resonance indicators according to the sea ice ice condition data, the snow disaster data and the port data;

[0044] S3, obtaining sea ice resonance comprehensive indexes and snow disaster resonance comprehensive indexes according to the sea ice resonance indicators and the snow disaster resonance indicators;

[0045] S4, determining a resonance event set according to the sea ice resonance indicators, the snow disaster resonance indicators, the sea ice resonance comprehensive indexes, the snow disaster resonance comprehensive indexes and the port data;

[0046] S5, determining a resonance probability according to the resonance event set.

[0047] The above scheme of the embodiment determines a resonance event set of sea ice and snow disaster according to the sea ice ice condition data of a target sea area, the snow disaster data of a target area and the port data in the target sea area, and obtains a resonance probability of sea ice and snow disaster from the resonance event set, which solves the problem that the resonance event of sea ice and snow disaster is not sufficient in the prior art, and can provide data basis and technical support for reasonable layout planning of the port and smooth navigation of coal across regions.

[0048] In an optional embodiment of the present application, step S3 comprises:

[0049] S31, determining the weight of the sea ice resonance index and the weight of the snow disaster resonance index by using the entropy weight coefficient method and the sea ice resonance index and the snow disaster resonance index.

[0050] S32, obtaining a sea ice resonance comprehensive index and a snow disaster resonance comprehensive index according to the weight of the sea ice resonance index and the weight of the snow disaster resonance index.

[0051] In an optional embodiment of the present application, step S3 further comprises:

[0052] S33, performing edge distribution fitting calculation on the sea ice resonance comprehensive index and the snow disaster resonance comprehensive index to obtain an optimal sea ice resonance comprehensive index and an optimal snow disaster resonance comprehensive index.

[0053] In an optional embodiment of the present application, step S5 comprises:

[0054] S51, obtaining an average sea ice comprehensive resonance index and an average snow disaster comprehensive resonance index according to the resonance event set.

[0055] S52, determining a resonance probability by using a connection function and the average sea ice comprehensive resonance index and the average snow disaster comprehensive resonance index.

[0056] In an optional embodiment of the present application, the method further comprises:

[0057] S6, obtaining a resonance risk distribution map according to the resonance probability and the port data.

[0058] In an optional embodiment of the present application, the sea ice condition data comprises sea ice area and sea ice thickness, the snow disaster data comprises cumulative precipitation, negative accumulated temperature and duration, and the port data comprises name, channel, longitude and latitude.

[0059] In an optional embodiment of the present application, the time range of the sea ice condition data and the snow disaster data is consistent.

[0060] In an optional embodiment of the present application, the sea ice resonance index comprises average sea ice thickness and sea ice event duration, and the snow disaster resonance index comprises cumulative precipitation, negative accumulated temperature and event duration.

[0061] In an optional embodiment of the present application, a specific embodiment of a method for determining a sea ice and snow disaster resonance probability is as follows:

[0062] The sea ice condition data of the target sea area, snow disaster data of the target area, and port data in the target sea area are obtained. The sea ice condition data is a historical continuous spatial distribution daily scale data set (nearly N years), including two data types of sea ice area and sea ice thickness. The simulation is carried out in units of kilometer grid, that is, the spatial resolution is 1 km. The snow disaster data is a snow disaster encounter event set of the target area in nearly M years, and the data type is the process cumulative precipitation, negative accumulated temperature and event duration of each freezing and snowfall encounter event. Considering that the time spans of the two data sets are different, the overlapping time (from December of each winter to February of the next year) is selected as the time range for constructing the historical resonance events of sea ice and snow disaster. Considering that sea ice and snow disaster occur at the same time or successively, they mainly affect the cross-regional transportation of coal and other energy materials, and aggravate disaster losses. The sea transportation of energy materials such as coal mainly relies on the main ports and their wharfs in the target sea area, which affects the ship's entry and exit and route. Considering the availability of data, the main channel of the port is selected as the research focus. According to the port construction scale, throughput and other data, considering the main transportation routes in the target sea area, the port data of seven ports in the sea area are selected: port a, port b, port c, port d, port e, port f and port g. The distribution of longitude and latitude coordinates and the length of the main channel of the seven ports are shown in Table 1.

[0063] Table 1 Distribution of main ports in the target sea area and length of main channel

[0064]

[0065]

[0066] From the perspective of port sea transportation, the target sea area is prone to sea ice conditions affected by cold waves. When there is a certain range and thickness of sea ice in the main channel that the ship must pass through when entering and leaving the port, it will affect the ship's entry and exit, operation, and whether the ship can smoothly enter and exit the port is the basic guarantee for subsequent navigation and transportation. In addition, the existence of sea ice in the port wharf and dock will also affect the ship's berthing. At present, ships mainly rely on the channels constructed by each port when entering and leaving the port, and berth near the shore wharf. The main channel of each port is different in length due to differences in natural conditions and functional zoning. Among them, the length of the channel of port e is relatively short, about 15.2 km, the main channel of port a is the longest, about 46.48 km, and the main channels of port g and port f are also relatively long, both more than 40 km. Accordingly, the main channel that plays an important role in the entry and exit of ships is taken as the object, and the distribution and length of the channel are used to define the range of the area affected by sea ice in the port. Further, taking the location of the port center as the circle point and the length of the main channel as the radius, a circular buffer zone is drawn, and the fan-shaped range of the buffer zone exposed to the target sea area is intercepted, which is taken as the main influence area of the ship entering and leaving the port of each port, such as Figure 2The triangle in the figure represents the port area, and the fan-shaped area represents the buffer area.

[0067] After determining the buffer area of the port affected by sea ice, the historical sea ice grid data (sea ice area and sea ice thickness) in each buffer area is counted. When defining the parameter threshold of sea ice affecting ship navigation, the sea ice is less dangerous when the ice thickness is <10 cm, and there is a certain danger when the sea ice is ≥10 cm. Combined with the mechanism of the influence of port sea ice on ship navigation, the grid points with sea ice thickness ≥10 cm in the buffer area of the port channel are selected, and it is considered that the sea ice grid point has an impact on the navigation of the port ship. In addition, considering that in addition to the thickness of sea ice, the distribution range of sea ice in the buffer area of the port channel will also have different effects on ship navigation. The number of grid points in each port buffer area that meet the sea ice thickness ≥10 cm is counted every winter day, and the 50th percentile of the above grid point number value sequence is selected as the threshold. As shown in Table 2, when the number of grid points with an average ice thickness ≥10 cm in the buffer area is greater than or equal to the threshold value, it is considered that the sea ice has an impact on the ship entering and leaving the port. For example, the g port channel buffer area has 2617 kilometer grid points, and when the number of grid points with an average ice thickness ≥2153 is met, it meets the selection requirements. Finally, for the sea ice kilometer grid points that meet the conditions, their average sea ice thickness is calculated for backup. Specifically, the average thickness of the sea ice in the buffer area of a port is ≥10 cm, and the number of grid points on the same day is greater than 50% of the quantile, then it is considered that the sea ice on that day will affect the ship entering and leaving the port, and the average value of the sea ice thickness that meets the conditions is one of the sensitive indicators of the sea ice resonance event.

[0068] Table 2: Number of grid points in the buffer area of the main port and its 50th percentile

[0069] Serial number Port name Buffer grid number Grid point threshold 1 Port a 3458 238 2 Port b 1806 898 3 Port c 1590 1465 4 Port d 480 94 5 Port e 386 25.5 6 Port f 2598 40 7 Port g 2617 2153

[0070] The process negative accumulated temperature, process precipitation and event duration were selected as resonance indicators, which were brought into the subsequent resonance probability calculation, considering the disaster-causing and influencing mechanism of snow disaster. It was assumed that the average transportation time of coal from the port of departure to the target port was about 3 days. Therefore, when determining the resonance event, the time of the snow disaster event was taken as the benchmark, and 3 days of transportation time was added to the 1st day of the snow disaster event. Within this time range, the time when the sea ice and snow disaster overlapped was selected as a resonance event. The sea ice in the buffer zone of the main port and the snow disaster event at the target region station were matched to select the resonance indicators that met the conditions of affecting ship transportation. Finally, the sea ice resonance indicators involved the average sea ice thickness and the duration of the sea ice event, and the snow disaster resonance indicators involved the cumulative precipitation, negative accumulated temperature and event duration. The weights of each indicator were determined by the entropy weight coefficient method, and the sea ice resonance comprehensive index and the snow disaster resonance comprehensive index were calculated for the subsequent joint probability calculation. The entropy weight coefficient method is an effective method for multi-objective decision-making, which can objectively evaluate by considering multiple factors and making full use of the inherent information of the evaluation object. This method is widely used in the field of natural disaster indicator definition and other fields, and is an objective statistical weight assignment method that reflects the law and principle of data indicators.

[0071] The connection function can be expressed as: for continuous random variables X, Y, whose marginal distributions are represented by F X (x) and F Y (y) respectively, the joint distribution is F(x, y), and if each marginal distribution function is continuous, there is a unique function connection function C θ (u, v) that makes F(x, y) = C θ (F X (x), F Y (y)) = u + vθ, x, y, where C θ (u, v) is the connection function and θ is the undetermined parameter.

[0072] The connection function Copula can be divided into elliptical type, Archimedean type and quadratic type, etc. Among them, the Archimedean type connection function has the advantages of simple function model and simple calculation process, and is widely used in the field of disaster. At present, the most commonly used in multi-disaster or multi-variable evaluation is the Archimedean type connection function, including Clayton Copula, Frank Copula and Gumbel Copula, as shown in Table 3. For the selection of these three Copula functions, the principle of minimum root mean square error and Akaike information criterion is adopted to determine the optimal Copula function.

[0073] Table 3 Copula function types and their expressions

[0074]

[0075] Both the sea ice composite index and the snow disaster composite index are continuous variables. In order to select the optimal distribution function, the generalized extreme value distribution, extreme value distribution, normal distribution, Poisson distribution, gamma distribution, exponential distribution, lognormal distribution, generalized Pareto distribution, and Weibull distribution are selected for the calculation of the edge distribution fitting. The maximum likelihood method is used to solve the parameters, and the Kolmogorov and Smirnov test methods are used for goodness-of-fit test to determine the optimal fitting distribution of each site. In the calculation of the bivariate joint return period, N is defined as the time series length of the ice and snow events, and n is the number of events. Therefore, M t is the average time interval between two events,

[0076] The bivariate joint return period based on the connection function can be expressed as:

[0077]

[0078] where x and y represent the sea ice composite index and the snow disaster composite index, respectively, T is the joint return period, F(X) is the cumulative probability, and M t represents the average time of two disasters.

[0079] The entropy weight coefficient method is used to calculate the weight of the sea ice characteristic index and the snow disaster characteristic index of each port data. The weight coefficient results are shown in Table 4. Among the sea ice resonance composite index, the weight coefficient of the duration is higher, with an average value of 0.91, followed by the sea ice thickness of 0.09. In the snow disaster encounter event resonance composite index, the weight coefficient of the negative temperature is higher, with an average of 0.32, followed by the process cumulative precipitation with an average of 0.51, and finally the duration with an average of 0.17.

[0080] Table 4 Weight coefficient of resonance composite index

[0081]

[0082] In the edge distribution fitting results, the optimal distribution of each site fitting is passed through the significance test. Among the sea ice and snow disaster in the resonance index near port g, the edge distribution of the sea ice composite index corresponding to each site, the optimal distribution of 75 sites is the generalized extreme value distribution, the most, followed by the lognormal distribution, the number is 57; the fitting average R 2 0.8201; for the snow disaster composite index, the most is the generalized extreme value distribution and the lognormal distribution, 90 and 41 respectively; the fitting average R 20.9119. Among the two indexes of port c, the generalized extreme value distribution was 61 and 91, and the lognormal distribution was 59 and 40, respectively; the fitting average R 2 of each station was 0.8128 and 0.9064, respectively. Among the two indexes of port b, the generalized extreme value distribution was 105 and 94, and the lognormal distribution was 34 and 45, respectively; the fitting average R 2 of each station was 0.8741 and 0.9117, respectively. Among the two indexes of port d, the generalized extreme value distribution was 31 and 22, and the lognormal distribution was 25 and 30, respectively; the fitting average R 2 of each station was 0.8649 and 0.9113, respectively. Among the two indexes of port f, the generalized extreme value distribution was 61 and 82, and the lognormal distribution was 60 and 33, respectively; the fitting average R 2 of each station was 0.8814 and 0.9176, respectively. Among the two indexes of port a, the generalized extreme value distribution was 43 and 79, and the lognormal distribution was 72 and 52, respectively; the fitting average R 2 of each station was 0.8785 and 0.9186, respectively. Among the two indexes of port e, the number of optimal distributions was slightly different, and the number of optimal distributions of the sea ice comprehensive index was the largest, with 36 lognormal distributions and 31 generalized extreme value distributions, and the fitting average R 2 of each station was 0.8776; in the snow disaster comprehensive index, there were 71 lognormal distributions and 53 normal distributions, and the fitting average R 2 of each station was 0.8989. The remaining stations selected their optimal distributions from the lognormal, Weibull, and other distributions. The sea ice resonance comprehensive index and the snow disaster resonance comprehensive index corresponding to port g and the national weather station were selected, and the fitting distribution was displayed by Q-Q diagram, as shown in FIG. 3. At the 0.05 significance level, the fitting R 2 was 0.9952 and 0.9901, respectively, and the fitting result was ideal.

[0083] The advantage of the connection function is that it does not require the marginal distribution of each variable in the joint distribution to follow the same distribution type, and it can be more flexible to solve practical problems. Based on the resonance of the port and the target area of the meteorological station, the optimal connection function of the resonance index is selected, and then the cumulative probability calculation and joint distribution model construction are carried out. When the bivariate joint distribution model is optimized, the parameters of the three connection functions are calculated and the goodness of fit is evaluated, and the optimal model of each port and its corresponding station is selected for subsequent return period calculation. All calculation processes can be realized in Matlab software. Taking the resonance calculation of port g and meteorological station snow disaster events as an example, the probability density and joint cumulative probability distribution in the connection function are shown in Figure 4, the calculation of the parameters of the three connection functions and the evaluation of the fitting effect of the model are shown in Table 5, and the Gumbel connection function is finally selected as the probability calculation model of the resonance of port g and the target area meteorological station.

[0084] Table 5 Copula function parameter calculation and model fitting effect evaluation

[0085]

[0086] The optimal Copula model of the resonance events of the 7 main ports in the target sea area and the 257 meteorological stations in the target area is selected, and the final result is shown in Table 6. Among them, the number of Gumbel Copula in the fitting results of port g, port c, port b, port a and port f is the most, and the number of Clayton Copula selected by port d and port e is the most.

[0087] Table 6 Number of optimal Copula functions of resonance events of each port and target area station

[0088]

[0089]

[0090] The 7 main ports of the target sea area are respectively established with 257 weather stations in the target area to form a data set. The resonance comprehensive index is obtained according to the selected resonance characteristic index and its weight, and the historical resonance event screening of "port-station" point-to-point is carried out. The number of resonance events of each port and the target area in N years and the average number of events per year are obtained, as shown in Table 7. Among them, port g and port b have the most resonance years, a total of 31 years. Followed by port c 29 years, port a 28 years, port f 27 years, port e and port d have relatively fewer historical resonance years, 10 years and 7 years respectively. The total number of historical resonance events shows that the number of events is the most in port b, port g and port c. Among them, port c has the most average number of events, about 149.84, that is, every 1.6 stations will have a resonance with port c every year from December to March of the next year. The total number of historical resonance events of port d and port e is 656 and 761 respectively, and the average number of resonance events per year is lower, which is 93.71 and 76.10 respectively.

[0091] Table 7 Statistics of historical resonance event occurrence

[0092] Port name Resonance years Historical resonance events Port a 28 3222 Port b 31 4645 Port c 29 4104 Port d 7 656 Port e 10 761 Port f 27 2439 Port g 31 4301

[0093] By counting the number of resonance events of each port and the target area weather station, the three stations with the most resonance events of each port are selected. Port a, port a, port a and port g have the most resonance events with the weather stations in the target area, and the number of events is between 50-173. Port d, port e and port f have the most resonance events with the weather stations in the target area, and the number of events is much smaller than the above ports, only 14-85. Therefore, "ice and snow" resonance events do exist, and different degrees of "ice and snow" resonance events will occur every winter.

[0094] According to the target sea area 7 port and its corresponding target area site index weight, the optimal edge distribution, resonance model and parameter results, the average sea ice comprehensive resonance index and the average snow disaster comprehensive resonance index of the historical "ice and snow" resonance event are taken as the input data, the numerical value is 0.1797 and 0.0795, and the preferred Cupola function of each port and each site is brought in, and the corresponding "ice and snow" resonance event joint return period of each "port-site" under the same resonance index is obtained, which is taken as the ice and snow resonance event risk evaluation index. With the help of ArcGIS10.6, the spatial distribution of joint return period is represented. The larger the joint return period is, the lower the exceeding probability of the same intensity event at the site is, and vice versa. According to the calculation of joint return period, the risk map is drawn. The average sea ice resonance index and snow disaster resonance of all historical resonance events are substituted into the resonance model of each port and site, that is, the same disaster resonance event intensity is selected, the joint return period is calculated to represent the risk. In some areas of the target area, due to the lower latitude and relatively warm temperature, there is basically no resonance of port ice event and snow disaster event of each site in the target area. In contrast, in some areas of the target area, the return period is small and the risk level is high, which is basically consistent with the spatial distribution of snow disaster event risk in the target area. In addition, the resonance of different ports and target areas is also different. The spatial distribution of port a, port b, port c, port f and port g is similar, and in winter, sea ice freezing is easy to resonate with snow disaster events in some places of the target area, and "ice and snow" resonance phenomenon occurs. Port d only has resonance events with some places in the target area, and the overall occurrence probability is low, mainly relying on the natural geographical advantage, the sea ice is less in winter, the water area is open, and it has the advantage condition of forming a non-freezing port. The overall high-risk site of port e is also less.

[0095] In the risk distribution of each port and target area site, the minimum joint return period exists between a site in the target area and port a, port b, port c, port d and port g, that is, the resonance risk is the largest; the resonance risk of port e and a site in the target area and port f and a site in the target area is larger. By calculating the average joint return period of each main port ice and target area site snow disaster event resonance, Figure 5 It can be found that the average joint return period fluctuates in the interval of 6.87 years to 15.22 years, among which the return period of port a is the smallest, followed by port c, port b and port g, then port f and port d, and finally port e, which has the largest return period. The reason is that port a has the longest main channel distribution, about 46.48 km, so the channel buffer zone has more possibility of sea ice distribution.

[0096] At the same time, the target sea area of 10 cm above sea ice, mainly concentrated in the southwest coastline, that is, the port a is located The probability of ice thickness affecting transportation is larger. Therefore, the average joint return period is lower, and the danger is high. In addition, the joint return period of port c, port b and port g is also relatively low. The reason is that the probability of large-scale sea ice and thick sea ice is the largest, so it also leads to a higher risk. In contrast, the risk of port f, port d and port e is lower.

[0097] Through the daily scale data set of the spatial distribution of sea ice in the target sea area and the snow disaster encounter event set in the target area, a historical resonance event set (near N years) of "sea ice" and "snow disaster" is further established. Considering that the resonance events of the two will directly affect transportation, thereby causing the cross-regional transportation of coal to be blocked. Therefore, taking the main transportation ports and their main navigation channels in the target sea area as the starting point, the navigation buffer zones of each port are drawn, and the comprehensive resonance characteristic indexes of sea ice and snow disaster encounter events are respectively optimized and calculated. With the help of Copula model theory, the optimal resonance model type is selected, the resonance probability of each port and snow disaster is obtained, and the risk distribution is characterized by spatial expression. In the future, the coal transportation capacity planning of the target sea area port can refer to the resonance risk distribution diagram of sea ice and snow disaster events in each port, so as to provide data basis and technical support for reasonable layout planning of the port and smooth navigation of coal.

[0098] When sea ice and snow disaster resonance occurs, it has a certain impact on various transportation modes of cross-regional coal transportation. When the overall ice condition of the target sea area is serious, and there is more sea ice or floating ice in the navigation buffer zone of each port, the port capacity can be coordinated and planned in a timely manner. According to the resonance risk distribution diagram of sea ice and snow disaster in the buffer zone of each port, a certain reference can be provided for policy planning such as allocation of port transportation capacity and transportation destination coordination.

[0099] As shown in Figure 6 The embodiment of the present application proposes a sea ice and snow disaster resonance probability determination device, which comprises:

[0100] The acquisition module is used for obtaining the sea ice condition data of the target sea area, the snow disaster data of the target area and the port data in the target sea area, and sending them to the resonance index calculation module and the event set determination module;

[0101] The resonance index calculation module is used for obtaining the sea ice resonance index and the snow disaster resonance index according to the sea ice condition data, the snow disaster data and the port data, and sending them to the resonance comprehensive index calculation module and the event set determination module;

[0102] The resonance comprehensive index calculation module is used for obtaining the sea ice resonance comprehensive index and the snow disaster resonance comprehensive index according to the sea ice resonance index and the snow disaster resonance index, and sending them to the event set determination module;

[0103] an event set determination module configured to determine a resonance event set according to the sea ice resonance indicator, the snow disaster resonance indicator, the sea ice resonance comprehensive index, the snow disaster resonance comprehensive index, and the port data, and send the resonance event set to a probability determination module;

[0104] the probability determination module is configured to determine a resonance probability according to the resonance event set.

[0105] The above scheme of the embodiment determines a resonance event set of sea ice and snow disaster according to sea ice condition data of a target sea area, snow disaster data of a target region, and port data in the target sea area, and obtains a resonance probability of sea ice and snow disaster, thereby solving the problem that the resonance event of sea ice and snow disaster is not sufficiently studied in the prior art, and providing data basis and technical support for reasonable layout planning of a port and smooth navigation of coal across regions.

[0106] In an optional embodiment of the present application, the resonance comprehensive index calculation module is specifically configured to:

[0107] determine the weight of the sea ice resonance indicator and the weight of the snow disaster resonance indicator by using an entropy weight coefficient method and the sea ice resonance indicator and the snow disaster resonance indicator;

[0108] obtain a sea ice resonance comprehensive index and a snow disaster resonance comprehensive index according to the weight of the sea ice resonance indicator and the weight of the snow disaster resonance indicator.

[0109] In an optional embodiment of the present application, the resonance comprehensive index calculation module is further configured to:

[0110] perform edge distribution fitting calculation on the sea ice resonance comprehensive index and the snow disaster resonance comprehensive index to obtain an optimal sea ice resonance comprehensive index and an optimal snow disaster resonance comprehensive index.

[0111] In an optional embodiment of the present application, the probability determination module is specifically configured to:

[0112] obtain an average sea ice comprehensive resonance index and an average snow disaster comprehensive resonance index according to the resonance event set;

[0113] determine a resonance probability by using a connection function and the average sea ice comprehensive resonance index and the average snow disaster comprehensive resonance index.

[0114] In an optional embodiment of the present application, the device further comprises:

[0115] a distribution map drawing module configured to obtain a resonance danger distribution map according to the resonance probability and the port data.

[0116] In an optional embodiment of the present application, the sea ice condition data comprises sea ice area and sea ice thickness, the snow disaster data comprises accumulated precipitation, negative accumulated temperature and duration, and the port data comprises name, channel, longitude and latitude.

[0117] In an optional embodiment of the present application, the time range of the sea ice condition data and the snow disaster data is consistent.

[0118] In an optional embodiment of the present application, the sea ice resonance index comprises average sea ice thickness and sea ice event duration, and the snow disaster resonance index comprises accumulated precipitation, negative accumulated temperature and event duration.

[0119] An embodiment of the present application also provides a computer readable storage medium, which stores instructions, and the instructions, when executed on a computer, cause the computer to perform the method as Figure 1 The steps of the method are not repeated here.

[0120] The above describes the preferred embodiments of the present application. It should be noted that, for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A method for determining the probability of resonance between sea ice and snow disasters, characterized in that, include: Acquire sea ice and ice condition data for the target sea area, snow disaster data for the target area, and port data within the target sea area; Based on the sea ice condition data, the snow disaster data, and the port data, the sea ice resonance index and the snow disaster resonance index are obtained. Based on the sea ice resonance index and the snow disaster resonance index, the comprehensive sea ice resonance index and the comprehensive snow disaster resonance index are obtained. Based on the sea ice resonance index, the snow disaster resonance index, the sea ice resonance comprehensive index, the snow disaster resonance comprehensive index, and the port data, a set of resonance events is determined; Determine the resonance probability based on the set of resonance events; The sea ice data includes sea ice area and sea ice thickness; the snow disaster data includes cumulative precipitation, negative accumulated temperature and duration; and the port data includes name, channel, longitude and latitude. The time ranges of the sea ice condition data and the snow disaster data are consistent. Based on the occurrence time of the snow disaster, the first day after the snow disaster event is added to the three days required for initial transportation. Within this time frame, the time when the sea ice and snow disaster events overlap is selected as a resonance event. The sea ice conditions in the buffer zone of the main port are matched with the snow disaster events at the target area stations to select resonance indicators that meet the conditions affecting ship transportation. The sea ice resonance indicators are sea ice thickness and sea ice event duration, while the snow disaster resonance indicators are cumulative precipitation, negative accumulated temperature, and event duration. The main port buffer zone is a circular buffer zone with the port center as the center and the main channel length as the radius. The fan-shaped area of ​​the circular buffer zone exposed in the target sea area is then selected. Among them, based on the sea ice resonance index and the snow disaster resonance index, the comprehensive sea ice resonance index and the comprehensive snow disaster resonance index are obtained, including: The weights of the sea ice resonance index and the snow disaster resonance index are determined using the entropy weight coefficient method and the sea ice resonance index. Based on the weights of the sea ice resonance index and the snow disaster resonance index, the comprehensive sea ice resonance index and the comprehensive snow disaster resonance index are obtained. The comprehensive index of sea ice resonance and the comprehensive index of snow disaster resonance are obtained based on the sea ice resonance index and the snow disaster resonance index, and also include: The optimal sea ice resonance index and the optimal snow disaster resonance index are obtained by edge distribution fitting calculation on the sea ice resonance composite index and the snow disaster resonance composite index.

2. The method for determining the resonance probability of sea ice and snow disasters according to claim 1, characterized in that, Based on the set of resonance events, the resonance probability is determined, including: Based on the set of resonance events, the mean sea ice comprehensive resonance index and the mean snow disaster comprehensive resonance index are obtained; The resonance probability is determined using the connection function, the mean sea ice composite resonance index, and the mean snow disaster composite resonance index.

3. The method for determining the resonance probability of sea ice and snow disasters according to claim 1, characterized in that, The method further includes: Based on the resonance probability and the port data, a resonance hazard distribution map is obtained.

4. A device for determining the probability of resonance between sea ice and snow disasters, characterized in that, include: The acquisition module is used to acquire sea ice condition data, snow disaster data, and port data within the target sea area, and send them to the resonance index calculation module and the event set determination module. The resonance index calculation module is used to obtain the sea ice resonance index and the snow disaster resonance index based on the sea ice condition data, the snow disaster data and the port data, and send them to the resonance comprehensive index calculation module and the event set determination module. The resonance composite index calculation module is used to obtain the sea ice resonance composite index and the snow disaster resonance composite index based on the sea ice resonance index and the snow disaster resonance index, and send them to the event set determination module. The event set determination module is used to determine the resonance event set based on the sea ice resonance index, the snow disaster resonance index, the sea ice resonance comprehensive index, the snow disaster resonance comprehensive index, and the port data, and send it to the probability determination module. The probability determination module is used to determine the resonance probability based on the set of resonance events. The sea ice data includes sea ice area and sea ice thickness; the snow disaster data includes cumulative precipitation, negative accumulated temperature and duration; and the port data includes name, channel, longitude and latitude. The time ranges of the sea ice condition data and the snow disaster data are consistent. Based on the occurrence time of the snow disaster, the first day after the snow disaster event is added to the three days required for initial transportation. Within this time frame, the time when the sea ice and snow disaster events overlap is selected as a resonance event. The sea ice conditions in the buffer zone of the main port are matched with the snow disaster events at the target area stations to select resonance indicators that meet the conditions affecting ship transportation. The sea ice resonance indicators are sea ice thickness and sea ice event duration, while the snow disaster resonance indicators are cumulative precipitation, negative accumulated temperature, and event duration. The main port buffer zone is a circular buffer zone with the port center as the center and the main channel length as the radius. The fan-shaped area of ​​the circular buffer zone exposed in the target sea area is then selected. Among them, based on the sea ice resonance index and the snow disaster resonance index, the comprehensive sea ice resonance index and the comprehensive snow disaster resonance index are obtained, including: The weights of the sea ice resonance index and the snow disaster resonance index are determined using the entropy weight coefficient method and the sea ice resonance index. Based on the weights of the sea ice resonance index and the snow disaster resonance index, the comprehensive sea ice resonance index and the comprehensive snow disaster resonance index are obtained. The comprehensive index of sea ice resonance and the comprehensive index of snow disaster resonance are obtained based on the sea ice resonance index and the snow disaster resonance index, and also include: The optimal sea ice resonance index and the optimal snow disaster resonance index are obtained by edge distribution fitting calculation on the sea ice resonance composite index and the snow disaster resonance composite index.

5. A computer-readable storage medium, characterized in that, The computer stores instructions that, when executed on the computer, cause the computer to perform the method as described in any one of claims 1 to 3.

Images