2022数维杯赛题

2022_“ShuWei Cup” Problem A：

Automatic seismic horizon tracking

With the economic and social development of our country, the importance of geological work is also increasing. Seismic data interpretation is an important stage of seismic exploration engineering, which can clarify subsurface tectonic features for oil and gas exploration and can provide good and favorable reservoirs for oil and gas exploration; accurate stratigraphic information is the basis of seismic data interpretation and is an important basis for storage prediction. Seismic horizon tracking is one of the key technologies in seismic data interpretation, a good seismic horizon tracking method can greatly improve the efficiency and accuracy of seismic data interpretation. It is the main goal of seismic exploration to obtain the information of underground structure lithology and reservoir, because the main formation interface is generally a good wave impedance interface, the seismic wave is affected by the formation interface when it propagates in the underground medium, and finally shows different seismic reflection characteristics, such as the morphology, intensity, frequency and continuity of the homogeneous axis of seismic reflection. Structural information such as the shape and burial depth of the stratigraphic interface can be obtained directly from seismic data. Since this kind of structural information is the most intuitive and easily used information of seismic data, it has become one of the most important targets of seismic exploration to extract structural information from seismic data since the birth of seismic exploration technology. In reflected seismic data, the seismic wave impedance interface usually corresponds to the formation interface or lithologic interface, but the lithologic interface can not always form wave impedance interface, only in those adjacent formations with large enough wave impedance difference can form wave impedance interface. Although the lithology of strata formed in different geological ages is usually different, only through the alternation of sedimentary compaction and sedimentary hiatus in millions of years can the differences in rock physical properties (density, porosity, etc.) between adjacent strata be revealed, the combination of lithology and rock physical properties (differences) will form significant wave impedance differences, therefore, seismic reflection events axis on seismic profiles usually correspond to sedimentary isochronous surfaces rather than macroscopic lithological interfaces. According to this theory, the stratigraphic interface indicated by seismic events axis is the discontinuity of the stratigraphic deposition process, because of its relative isochronism, this sedimentary discontinuity is basically consistent with the structural characteristics of the stratum, therefore, seismic events axis is the main signs to identify the stratigraphic interface. The spatial distribution characteristics and time domain variation characteristics of seismic events axis are the main basis for horizon interpretation. Seismic events axis can also be used to obtain information such as stratigraphic dip and azimuth. In the era of two-dimensional seismic exploration and the early stage of three-dimensional seismic exploration, the horizon interpretation of seismic data is mainly single-layer, that is, several seismic events axes with good continuity corresponding to the strong stratigraphic reflection interface are selected from the seismic profile for tracking. Because of the low efficiency of this horizon interpretation method and the small number of seismic event axis that can be easily traced on the seismic profile, the number of horizons that can be obtained is limited, resulting in the traditional seismic structure interpretation model unable to obtain detailed geological structure information, so the detailed description of geological structure characteristics is not clear enough. In other words, the traditional seismic horizon interpretation method ignores or wastes a lot of seismic information, and it has been unable to meet the requirements of modern seismic structure interpretation and geological comprehensive research in terms of accuracy and efficiency. With the development of three-dimensional seismic exploration, especially high-density seismic exploration technology, the accuracy of seismic data obtained is getting higher and higher, and the number of seismic data is increasing, automatic extraction of structural, lithological, fluid and other information from seismic data has become the key to the progress of modern seismic data interpretation, it is also the goal that geophysicists and geologists are striving for. The existing seismic horizon tracking methods are usually done manually by seismic horizon interpreters. In the interpretation of seismic data, the tracking of the event axis is very important. Interpreters are mainly based on seismic wave dynamics and kinematics characteristics, namely amplitude, in-phase or continuity, waveform similarity three criteria, and artificial contrast tracking. The artificial horizon tracking is to manually track the continuous reflection events axes of the bottom layer on the two-dimensional seismic profile by using the waveform similarity to obtain the horizon line (stratigraphic interface), and then interpolate all the horizon lines to form the horizon surface. However, artificial horizon tracking labor cost demand is large, not only time-consuming, but also has a great impact on the efficiency of seismic exploration. In order to overcome the problems of low tracking time efficiency and poor reliability of results, researchers have begun to attach great importance to the automatic horizon tracking method in recent years. The automatic horizon tracking method is to search for ' seed points ' with similar characteristics on seismic traces, search through these characteristics, and search the next region repeatedly after meeting the conditions. This method solves the problem that it is difficult to obtain artificial horizon information when the terrain is more complex, and the information obtained is more accurate than that obtained manually. At present, there are two better automatic horizon tracking criteria, namely automatic tracking based on waveform characteristics and automatic tracking based on correlation. Automatic tracking based on waveform features is to find only similar waveform structures (crests, troughs, zero crossings, etc.) of feature points in the search time window, but no correlation calculations are performed between the seismic traces, and the defined troughs, crests, and crossings are searched one by one. Because the continuity and stability between the local areas of the underground are reflected in the seismic time profile, it is the similarity and continuity of the seismic wave reflection layer in the amplitude of the seismic wave reflection layer on the adjacent seismic channel. Therefore, based on the relevant horizon automatic tracking algorithm, the seed point is taken as the center, according to the relevant time window range, a seismic channel is selected, and the seismic data of this section of seismic data is correlated with the seismic data in the search time window of the adjacent channel, if the characteristic point that meets the conditions is found in the search time window, the point is fixed as a new seed point, and then the next trace is picked up. Please establish a mathematical model based on the attached data to solve the following problems: （1）There are often a lot of noise in seismic data, please use effective methods to denoise the accessory data. （2）Establish the correlation of seismic strata automatic tracking model or design the corresponding new tracking algorithm, and track the attachment data. （3）Establish an automatic tracking model based on waveform features or design a corresponding new tracking algorithm, and track the attachment data. （4）Evaluate the results of two automatic tracking models (or algorithms), verify the rationality of the model, analyze the error between the data obtained from the experiment and the actual data, and make a reasonable explanation. （5）Establish a three-dimensional horizon automatic tracking model based on correlation and waveform, and an algorithm is implemented on the data given in the annex to realize horizon tracking and identify and analyze the fault data. Notes: A profile is made up of a set of curves, horizon tracking, is to trace the event axis. An event axis on a profile is a curve, and multiple event axis bar curves, make up the horizon

 2022_“ShuWei Cup” Problem B：

Red VS. Blue

In modern war, both offensive and defensive sides need to introduce efficient war strategies to increase war threats and reduce losses. Only by forming a relatively stable and balanced war dynamics can the ultimate goal of reaching consensus be realized as soon as possible. In view of the above war problems, consider the following simplification of the Red VS. Blue war problem: assuming that the Red and the Blue are engaged in the battle as shown in Figure 1, the two parties can only conduct the initial platoon in the position with the same color, and each node has its own attack difficulty. The more difficult the attack is, the larger the circle radius in Figure 1, you need to provide the optimal battle strategy for each party based on the actual number and characteristics of the two parties' military weapons. The main fighting units on both sides are infantry, and the main weapons are light tanks with mobility and concealment, medium tanks with balanced firepower and mobility, heavy tanks with heavy armor and powerful firepower, self-propelled artillery with ultra-long-range striking ability and powerful fire support, strategic bombers (not too many units should be deployed to prevent bombing) and anti-aircraft artillery (each side can set up 10 anti-aircraft points). The Red has 1.25 million infantry, 500 drones, 180 heavy tanks, 300 medium tanks, 420 light tanks and 7000 self-propelled guns. The Blue has 1 million infantry, 300 drones, 340 heavy tanks, 570 medium tanks, 800 light tanks, and 14,000 self-propelled guns. See Attachment 2 for the specific parameters of the Red and the Blue weapons. Please solve the following three problems through appropriate simplified assumptions and mathematical modeling methods: Figure 1 Assignable nodes for the Red and Blue Question 1: Based on the data in Annex 1 and Annex 2, and considering the attack difficulty, march distance, weapon range and air defense deployment of each node, please work out the assigned positions and quantity scale of infantry, tanks, self-propelled artillery and air defense artillery of both sides, as well as the optimal command positions and several alternative positions of both sides. Question 2: Based on the optimization results of question 1, you need to build the optimization model of medical supplies, military supplies and daily supplies distribution and supply for both the Red and Blue. At the same time, on the basis of fully considering the potential attack strategy from the other side, the key information such as the total number of workers and vehicles required in the non-supply mode is provided during the modeling. Finally, you need to provide the optimal supply plan of the Red and the Blue in the form of tables or graphs in the text. Question 3: In combination with the previous two questions and in the case of the Red attacking and the Blue defending, please propose the Red’s better attack plan and the Blue's better retreat plan. Given that the retreat node of the Blue is [37,140,378], what are the different overall retreat plans of the Blue in the case of good communication and communication interruption?

2022_“ShuWei Cup” Problem C：

How to Diagnose Alzheimer's Disease Using Brain Structural Features and Cognitive Behavioral Features

Alzheimer's disease (AD) is a progressive neurodegenerative disease with an insidious onset. It is characterized clinically by a full spectrum of dementia, including memory impairment, aphasia, dysfluency, agnosia, impairment of visuospatial skills, executive dysfunction, and personality and behavioral changes, the cause of which is still unknown. It is characterized by a progressive decline in the ability to perform activities of daily living, with various neuropsychiatric symptoms and behavioral disturbances. The disease is usually progressive in the elderly, with progressive loss of independent living skills and death from complications 10 to 20 years after the onset of the disease. The preclinical stage of Alzheimer's disease, also known as mild cognitive impairment (MCI), is a transitional state between normal and severe. Due to the limited cognition of the disease by patients and their families, 67% of patients were diagnosed as moderate to severe and had missed the best intervention stage. Therefore, early and accurate diagnosis of Alzheimer's disease and mild cognitive impairment is of great significance. The attached data contain specific information characteristics of 4850 cognitive normal elderly (CN), 1416 patients with subjective memory complaint (SMC), 2968 patients with early mild cognitive impairment (EMCI), 5236 patients with late mild cognitive impairment (LMCI) and 1738 patients with Alzheimer's disease (AD) collected at different time points (one time point is a quantity). Please use the brain structural characteristics and cognitive behavioral characteristics of the different categories of people provided in the Appendix to construct an Alzheimer's disease identification model and design an intelligent diagnostic method to accurately diagnose Alzheimer's disease. （1）Preprocess the characteristic indicators of the attached data to investigate the correlation between data characteristics and the diagnosis of Alzheimer's disease. （2）Use the attached structural brain features and cognitive behavioral features to design an intelligent diagnosis of Alzheimer's disease. （3）First, cluster CN, MCI and AD into three major classes. Then, for the three subclasses contained in MCI (SMC, EMCI, and LMCI), the clustering was continued to be refined into three subclasses. （4）The same sample in the annex contains features collected at different time points, please analyze them in relation to the time points to uncover patterns in the evolution of different categories of diseases over time. （5）Please consult the relevant literature to describe the early intervention and diagnostic criteria for the five categories of CN, SMC, EMCI, LMCI, and AD. 

2022_“ShuWei Cup” Problem D：

Research on the loss evaluation and coping strategies of extreme climate disasters under the Triple La Niña Event

From July to August 2022, many cities in the south of China experienced many days of hot weather, while in some parts of the north there were also large-scale heavy precipitation. In addition, many European countries have also experienced historically rare drought disasters. Whether it is high temperature weather in the south, heavy precipitation in the north, and dry weather in Europe, it is unprecedented for decades, and even the highest temperature, heavy precipitation and drought disasters have been recorded since meteorological data. The high temperature weather has caused economic losses and casualties to a certain scale in many cities in the south and European countries. Similarly, the heavy rainfall has caused a significant reduction in agricultural production or even no harvest in some areas of the north. The meteorological department attributed this high temperature phenomenon and heavy precipitation event to the Triple La Niña event. The latest data from the World Meteorological Organization shows that the La Niña event, which has lasted for a long time, is likely to continue until the end of this year or beyond. This will be the first Triple La Niña event in the 21st century, meaning three consecutive La Niña winters in the northern hemisphere. The La Niña event is a phenomenon in which the sea surface temperature in the eastern and central equatorial Pacific continues to be abnormally cold. The British "Nature" magazine issued a warning in June that more La Niña events will have multiple impacts, such as increasing the probability of flooding in Southeast Asia, increasing the risk of drought and wildfires in the southwestern United States, forming multiple hurricane, cyclone and monsoon patterns in the Pacific and Atlantic Oceans, and triggering weather changes in other regions. Please complete the following four questions in combination with international meteorological data free download platforms such as https://www.ncei.noaa.gov/maps/daily/ and their related optimization modeling methods: （1）Conduct statistical analysis of the major countries and regions involved in the global Triple La Niña event, and predict the possibility of the Triple La Niña events in the future; （2）Taking a country as an example, evaluate and analyze the various types of disaster losses caused by heat and drought under the Triple La Niña event, and provide targeted coping strategies. （3）Taking a country as an example，evaluate and analyze various disaster losses caused by floods under the action of the Triple La Niña event, and provide targeted coping strategies; （4）Please write a report of no more than 2,000 words for the relevant management in response to the Triple La Niña Event. 

关注我，后续分享数学建模思路~