WO2019242779A1 - 基于因果影响网络算法的极地海冰面积预测方法及装置 - Google Patents

基于因果影响网络算法的极地海冰面积预测方法及装置 Download PDF

Info

Publication number
WO2019242779A1
WO2019242779A1 PCT/CN2019/092781 CN2019092781W WO2019242779A1 WO 2019242779 A1 WO2019242779 A1 WO 2019242779A1 CN 2019092781 W CN2019092781 W CN 2019092781W WO 2019242779 A1 WO2019242779 A1 WO 2019242779A1
Authority
WO
WIPO (PCT)
Prior art keywords
sea ice
ice area
month
several months
predicted
Prior art date
Application number
PCT/CN2019/092781
Other languages
English (en)
French (fr)
Inventor
白玉琪
李莎
Original Assignee
清华大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 清华大学 filed Critical 清华大学
Publication of WO2019242779A1 publication Critical patent/WO2019242779A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Definitions

  • the present application relates to the field of computer technology, and in particular, to a method and device for predicting polar sea ice area based on a causal effect network algorithm.
  • polar sea ice As the main cold source of the earth, polar sea ice has an important impact on regional and global climate.
  • the surface of the sea ice reflects most of the short-wave solar radiation, blocking the exchange of heat and water vapor between the sea and the air; on the other hand, the energy absorbed and released by the change of sea ice changes the balance of energy balance in the atmosphere. Change the regional or global temperature, pressure, and wind field distribution from the above two aspects. Therefore, it is of great significance to accurately predict the area of polar sea ice.
  • a typical polar sea ice prediction system is the Earth System Model FIO-ESM developed by the First Ocean Research Institute of the National Oceanic Administration. This model is mainly composed of a climate system model and a carbon cycle model.
  • the polar sea ice prediction system has a complicated algorithm, and the implementation cost and maintenance cost of the system are relatively high.
  • the purpose of this application is to provide a method and device for predicting polar sea ice area based on a causal effect network algorithm, which solves the polar sea ice prediction system in the prior art.
  • the algorithm is more complex, and the system has higher implementation and maintenance costs. problem.
  • the present application provides a method for predicting polar sea ice area based on a causal effect network algorithm, including:
  • the present application provides a polar sea ice area prediction device based on a causal effect network algorithm, including:
  • An acquisition module configured to acquire the sea ice area in several months before the month to be predicted, and each climate variable at each observation point in each of the several months;
  • a prediction module for inputting the sea ice area of each of the months and each climate variable at each observation point of each of the months into a preset causal influence network model, and outputting The sea ice area in the month to be predicted.
  • the present application provides an electronic device for predicting polar sea ice area based on a causal effect network algorithm, including:
  • a memory and a processor where the processor and the memory communicate with each other through a bus; the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the foregoing method.
  • the present application provides a computer-readable storage medium on which a computer program is stored, and the computer program implements the foregoing method when executed by a processor.
  • the method and device for predicting polar sea ice area based on the causal effect network algorithm provided in the present application trains the causal effect network model through historical data, and then inputs the observation data of the first few months of the year to the trained causal effect network
  • the model can accurately predict the sea ice area of the target month. This method is easy to operate, the system complexity is low, and the prediction efficiency is high.
  • FIG. 1 is a schematic diagram of a method for predicting polar sea ice area based on a causal effect network algorithm according to an embodiment of the present application
  • FIG. 2 is a schematic diagram of a polar sea ice area prediction device based on a causal effect network algorithm according to an embodiment of the present application
  • FIG. 3 is a schematic structural diagram of an electronic device for predicting a polar sea ice area based on a causal effect network algorithm according to an embodiment of the present application.
  • FIG. 1 is a schematic diagram of a method for predicting polar sea ice area based on a causal effect network algorithm according to an embodiment of the present application. As shown in FIG. 1, an embodiment of the application provides a method for predicting polar sea ice area based on a causal effect network algorithm. Methods include:
  • Step S101 Obtain the sea ice area in several months before the month to be predicted, and each climate variable at each observation point in each of the several months;
  • Step S102 input the sea ice area of each of the several months and each climate variable at each observation point of each of the several months into a preset causal influence network model, and output the Sea ice area for the month to be predicted.
  • the sea ice area in several months before the month to be predicted and each climate variable at each observation point in each of the several months. For example, if we want to predict the sea ice area in September based on the data for the six months before September, then we need to obtain the sea ice area for each of the six months before September, and the monthly Every climate variable at every observation point.
  • the polar region can be divided into multiple grids (small areas), and each grid can be set with an observation point.
  • the number of observation points and the number of climate variables are determined according to the actual situation.
  • the output sea ice area for the predicted month includes the sea ice area of the entire polar region and also the sea ice area of each network. For the above example, you only need to input the sea ice area in each of the six months before September and each climate variable at each observation point in each month into a pre-trained causal influence network model. The predicted value of sea ice area in September can be output.
  • the method for predicting polar sea ice area based on the causal effect network algorithm trains the causal effect network model through historical data, and then inputs the observation data of the first few months of the year to the trained causal effect network
  • the model can accurately predict the sea ice area of the target month. This method is easy to operate, the system complexity is low, and the prediction efficiency is high.
  • Each year's historical data includes the sea ice area in several months before the target month of the year, and each climate variable at each observation point in each of the several months. Including sea ice area for the target month of the year;
  • a causal effect network algorithm is used to perform a fitting operation on the historical data to obtain the preset causal effect network model.
  • the causal effect network model needs to be trained.
  • the specific training method is as follows:
  • Each year's historical data includes the sea ice area in several months before the target month of the year, and each climate variable at each observation point in each of these months. , Also includes the sea ice area for the target month of the year.
  • the historical data of each of the 20 years of historical data includes the sea ice area for the six months before September of that year, as well as each climate variable at each observation point in each of the six months. Including the sea ice area in September of that year.
  • the causal effect network (CEN) algorithm is used to fit the obtained historical data to the sea ice area of each month in the months before the target month of the year, and each of these months.
  • the method for predicting polar sea ice area based on the causal effect network algorithm trains the causal effect network model through historical data, and then inputs the observation data of the first few months of the year to the trained causal effect network
  • the model can accurately predict the sea ice area of the target month. This method is easy to operate, the system complexity is low, and the prediction efficiency is high.
  • the preset causal influence network model is as follows:
  • Y is the sea ice area in the month to be predicted
  • Z -k represents the sea ice area in the k-th month before the month to be predicted
  • q -k is the regression coefficient of Z -k
  • O is the number of months
  • M is the observation point
  • N is the number of said climate variables
  • ⁇ is a constant.
  • the preset causal influence network model obtained is as follows:
  • Y is the sea ice area in the month to be predicted
  • Z -k represents the sea ice area in the k-th month before the month to be predicted
  • q -k is the regression coefficient of Z -k
  • O is the number of months
  • M is the observation point
  • N is the number of said climate variables
  • ⁇ is a constant.
  • the method for predicting polar sea ice area based on the causal effect network algorithm trains the causal effect network model through historical data, and then inputs the observation data of the first few months of the year to the trained causal effect network
  • the model can accurately predict the sea ice area of the target month. This method is easy to operate, the system complexity is low, and the prediction efficiency is high.
  • climate variables include at least sea level pressure, surface air temperature, sea surface temperature, near-surface zonal wind, near-surface meridional wind, near-surface downward short-wave radiation, and near-ground Any of the long-wave radiation facing downwards.
  • the number of observation points and the number of climate variables need to be determined according to the actual situation.
  • sea level pressure, surface air temperature, sea surface temperature, near-surface zonal wind, near-surface meridional wind, near-surface short-wave radiation, and near-surface downward are obtained.
  • Long-wave radiation has a significant impact on the sea ice area in the target month. Therefore, the climate variables include at least sea level pressure, surface air temperature, sea surface temperature, near-surface zonal wind, near-surface meridional wind, and near-surface downward short waves. Either radiation or near-ground downward long-wave radiation. To get more accurate predictions.
  • the method for predicting polar sea ice area based on the causal effect network algorithm trains the causal effect network model through historical data, and then inputs the observation data of the first few months of the year to the trained causal effect network
  • the model can accurately predict the sea ice area of the target month. This method is easy to operate, the system complexity is low, and the prediction efficiency is high.
  • the number of the several months is eight.
  • each impact factor has the most significant impact on the sea ice area in the target month.
  • the predicted values are most consistent with the observed values. Therefore, the number of the several months is set to 8 in order to obtain a more accurate prediction value.
  • the method for predicting polar sea ice area based on the causal effect network algorithm trains the causal effect network model through historical data, and then inputs the observation data of the first few months of the year to the trained causal effect network
  • the model can accurately predict the sea ice area of the target month. This method is easy to operate, the system complexity is low, and the prediction efficiency is high.
  • the month to be predicted is September.
  • the Arctic sea ice coverage area is the smallest in September each year, and the prediction of the smallest sea ice coverage area is of great significance. Therefore, the month to be predicted is set to September.
  • the method for predicting polar sea ice area based on the causal effect network algorithm trains the causal effect network model through historical data, and then inputs the observation data of the first few months of the year to the trained causal effect network
  • the model can accurately predict the sea ice area of the target month. This method is easy to operate, the system complexity is low, and the prediction efficiency is high.
  • the sea ice area in the several months is the sea ice area in August.
  • the Arctic sea ice coverage area is the smallest in September each year, and the prediction of the smallest sea ice coverage area is of great significance. Therefore, the month to be predicted is set to September.
  • the sea ice area in August had a strong autocorrelation. Therefore, in order to obtain a more accurate prediction of the sea ice area in September, the sea ice area in the several months is the sea ice area in August.
  • the method for predicting polar sea ice area based on the causal effect network algorithm trains the causal effect network model through historical data, and then inputs the observation data of the first few months of the year to the trained causal effect network
  • the model can accurately predict the sea ice area of the target month. This method is easy to operate, the system complexity is low, and the prediction efficiency is high.
  • FIG. 2 is a schematic diagram of a polar sea ice area prediction device based on a causal effect network algorithm according to an embodiment of the present application.
  • an embodiment of the present application provides a polar sea ice area prediction device based on a causal effect network algorithm.
  • the method for completing the method described in the foregoing embodiment specifically includes an obtaining module 201 and a prediction module 202, where:
  • the obtaining module 201 is configured to obtain the sea ice area in several months before the month to be predicted, and each climate variable at each observation point in each of the several months;
  • the prediction module 202 is configured to input the sea ice area in each of the months and each climate variable at each observation point in each of the months into a preset causal influence network model, and output The sea ice area in the month to be predicted.
  • the embodiment of the present application provides a polar sea ice area prediction device based on a causal effect network algorithm, which is used to complete the method described in the foregoing embodiment, and the device provided in this embodiment is used to complete the specific method of the method described in the foregoing embodiment.
  • the steps are the same as those in the above embodiment, and are not repeated here.
  • the polar sea ice area prediction device based on the causal effect network algorithm provided in the embodiment of the present application trains the causal effect network model through historical data, and then inputs the observation data of the first few months of the year to the trained causal effect network
  • the model can accurately predict the sea ice area of the target month. This method is easy to operate, the system complexity is low, and the prediction efficiency is high.
  • FIG. 3 is a schematic structural diagram of an electronic device for predicting a polar sea ice area based on a causal effect network algorithm according to an embodiment of the present application.
  • the device includes: a processor 301, a memory 302, and a bus 303;
  • the processor 301 and the memory 302 complete communication with each other through the bus 303.
  • the processor 301 is configured to call program instructions in the memory 302 to execute the methods provided by the foregoing method embodiments, for example, including:
  • An embodiment of the present application discloses a computer program product.
  • the computer program product includes a computer program stored on a non-transitory computer-readable storage medium.
  • the computer program includes program instructions.
  • the computer can execute the methods provided by the foregoing method embodiments, for example, including:
  • An embodiment of the present application provides a non-transitory computer-readable storage medium that stores computer instructions, and the computer instructions cause the computer to execute the methods provided by the foregoing method embodiments, for example, include:
  • the foregoing program may be stored in a computer-readable storage medium.
  • the program is executed, the program is executed.
  • the method includes the steps of the foregoing method embodiment.
  • the foregoing storage medium includes: a ROM, a RAM, a magnetic disk, or an optical disk, and other media that can store program codes.
  • the embodiments can be implemented by means of software plus a necessary universal hardware platform, and of course, they can also be implemented by hardware.
  • the above-mentioned technical solution essentially or part that contributes to the existing technology can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic A disc, an optical disc, and the like include instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种基于因果影响网络算法的极地海冰面积预测方法及装置，所述方法包括：获取待预测月份之前的若干个月份的海冰面积，以及所述若干个月份中每一月份的在每一观测点的每一气候变量（S101）；将所述若干个月份中每一月份的海冰面积和所述若干个月份中每一月份的在每一观测点的每一气候变量输入至预设因果影响网络模型，输出所述待预测月份的海冰面积（S102）。所提供的基于因果影响网络算法的极地海冰面积预测方法及装置，通过历史数据对因果影响网络模型进行训练，然后将当年中的前几个月份的观测数据输入至训练好的因果影响网络模型，即可准确的预测出目标月份的海冰面积，该方法操作简便，系统复杂度低，预测效率高。

Description

基于因果影响网络算法的极地海冰面积预测方法及装置
相关申请的交叉引用
本申请要求于2018年6月21日提交的申请号为2018106466573，发明名称为“基于因果影响网络算法的极地海冰面积预测方法及装置”的中国专利申请的优先权，其通过引用方式全部并入本申请。
技术领域
本申请涉及计算机技术领域，尤其涉及一种基于因果影响网络算法的极地海冰面积预测方法及装置。
背景技术
极地海冰作为地球的主要冷源，对区域乃至全球气候有着重要影响。一方面表现在海冰表面会反射大部分的太阳短波辐射，阻隔海一气之间的热量和水汽交换；另一方面，因为海冰消长吸收和释放的热量会改变大气的能量收支平衡关系，从以上两方面改变区域或全球的温度、气压和风场分布等。因此，准确预测极地海冰面积意义重大。
现有技术中，典型的极地海冰预测系统是由国家海洋局第一海洋研究所研发的地球系统模式FIO-ESM，该模式主要由气候系统模式和碳循环模式两部分组成。
但是，现有技术中的极地海冰预测系统，算法较为复杂，系统的实现成本和维护成本较高。
发明内容
本申请的目的是提供一种基于因果影响网络算法的极地海冰面积预测方法及装置，解决了现有技术中极地海冰预测系统，算法较为复杂，系统的实现成本和维护成本较高的技术问题。
为了解决上述技术问题，一方面，本申请提供一种基于因果影响网络算法的极地海冰面积预测方法，包括：
获取待预测月份之前的若干个月份的海冰面积，以及所述若干个月份中每一月份的在每一观测点的每一气候变量；
将所述若干个月份中每一月份的海冰面积和所述若干个月份中每一月份的在每一观测点的每一气候变量输入至预设因果影响网络模型，输出所述待预测月份的海冰面积。
另一方面，本申请提供一种基于因果影响网络算法的极地海冰面积预测装置，包括：
获取模块，用于获取待预测月份之前的若干个月份的海冰面积，以及所述若干个月份中每一月份的在每一观测点的每一气候变量；
预测模块，用于将所述若干个月份中每一月份的海冰面积和所述若干个月份中每一月份的在每一观测点的每一气候变量输入至预设因果影响网络模型，输出所述待预测月份的海冰面积。
再一方面，本申请提供一种用于基于因果影响网络算法的极地海冰面积预测的电子设备，包括：
存储器和处理器，所述处理器和所述存储器通过总线完成相互间的通信；所述存储器存储有可被所述处理器执行的程序指令，所述处理器调用所述程序指令能够执行上述的方法。
又一方面，本申请提供一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现上述的方法。
本申请提供的基于因果影响网络算法的极地海冰面积预测方法及装置，通过历史数据对因果影响网络模型进行训练，然后将当年中的前几个月份的观测数据输入至训练好的因果影响网络模型，即可准确的预测出目标月份的海冰面积，该方法操作简便，系统复杂度低，预测效率高。
附图说明
图1为依照本申请实施例的基于因果影响网络算法的极地海冰面积预测方法示意图；
图2为依照本申请实施例的基于因果影响网络算法的极地海冰面积预测装置示意图；
图3为本申请实施例提供的用于基于因果影响网络算法的极地海冰面积预测的电子设备的结构示意图。
具体实施方式
为了使本申请实施例的目的、技术方案和优点更加清楚，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
图1为依照本申请实施例的基于因果影响网络算法的极地海冰面积预测方法示意图，如图1所示，本申请实施例提供一种基于因果影响网络算法的极地海冰面积预测方法，该方法包括：
步骤S101、获取待预测月份之前的若干个月份的海冰面积，以及所述若干个月份中每一月份的在每一观测点的每一气候变量；
步骤S102、将所述若干个月份中每一月份的海冰面积和所述若干个月份中每一月份的在每一观测点的每一气候变量输入至预设因果影响网络模型，输出所述待预测月份的海冰面积。
具体的，首先获取待预测月份之前的若干个月份的海冰面积，以及所述若干个月份中每一月份的在每一观测点的每一气候变量。例如，我们想要根据9月份之前的6个月的数据预测9月份的海冰面积，那么就需要获取9月份之前的这6个月的每个月的海冰面积，以及每个月的在每一观测点的每一气候变量。
在数据采集的过程中，可以将极地划分成多个网格(小区域)，每个网格中可以设置一个观测点。观测点的数量和气候变量的数量，根据具体的实际情况来确定。
然后，将这若干个月份中每一月份的海冰面积和这若干个月份中每一月份的在每一观测点的每一气候变量输入至预先训练好的因果影响网络模型，输出待预测月份的海冰面积。输出的待预测月份的海冰面积，包括整个极地的海冰面积，还包括，每个网络的海冰面积。针对上述例子，只需要将9月份之前的这6个月的每个月的海冰面积，以及每个月的在每一观测点的每一气候变量输入至预先训练好的因果影响网络模型，即可输出9月份海冰面积的预测值。
本申请实施例提供的基于因果影响网络算法的极地海冰面积预测方法，通过历史数据对因果影响网络模型进行训练，然后将当年中的前几个月份的 观测数据输入至训练好的因果影响网络模型，即可准确的预测出目标月份的海冰面积，该方法操作简便，系统复杂度低，预测效率高。
在上述实施例的基础上，进一步地，获取所述预设因果影响网络模型的具体步骤如下：
获取若干年的历史数据，每一年的历史数据包括当年目标月份之前的若干个月份的海冰面积，以及所述若干个月份中每一月份的在每一观测点的每一气候变量，还包括当年目标月份的海冰面积；
利用因果影响网络算法对所述历史数据进行拟合运算，获取所述预设因果影响网络模型。
具体的，在根据当年中的前几个月份的观测数据预测目标月份的海冰面积之前，需要对因果影响网络模型进行训练，具体的训练方法如下：
首先，获取若干年的历史数据，每一年的历史数据都包括当年目标月份之前的若干个月份的海冰面积，以及这若干个月份中每一月份的在每一观测点的每一气候变量，还包括当年目标月份的海冰面积。
针对上述例子，我们想要根据9月份之前的6个月的数据预测9月份的海冰面积，在训练的时候，可以选用当年之前20年的历史数据对因果影响网络模型进行训练。这20年的历史数据中每一年的历史数据都包括当年9月份之前的6个月的海冰面积，以及这6个月份中每一月份的在每一观测点的每一气候变量，还包括当年9月份的海冰面积。
然后，利用因果影响网络(Causal Effect Network，CEN)算法对获取到的历史数据进行拟合运算，以当年目标月份之前的若干个月份中每个月的海冰面积，以及这若干个月份中每一月份的在每一观测点的每一气候变量，分别作为一个影响因子，计算在不同的最大超前时间下，每个影响因子与目标月份的海冰面积之间的偏相关系数，以排除某一超前时刻(例如，lag＝-1～-8months)不相关的影响因子(偏相关系数r＝0)或不显著相关的影响因子(p-value高于显著性水平α＝0.05)，保留显著相关的影响因子(p<0.05)。
再利用多元线性回归理论，进行迭代分析，确定最终的与目标月份的海冰面积显著直接相关的影响因子，以及每一影响因子的权重，和消除随机性的常数值，最终得到训练好的因果影响网络模型。
本申请实施例提供的基于因果影响网络算法的极地海冰面积预测方法， 通过历史数据对因果影响网络模型进行训练，然后将当年中的前几个月份的观测数据输入至训练好的因果影响网络模型，即可准确的预测出目标月份的海冰面积，该方法操作简便，系统复杂度低，预测效率高。
在以上各实施例的基础上，进一步地，所述预设因果影响网络模型如下：
Figure PCTCN2019092781-appb-000001
其中，Y表示所述待预测月份的海冰面积，
Figure PCTCN2019092781-appb-000002
表示所述待预测月份之前第k个月份的第i个观测点的第j个气候变量，
Figure PCTCN2019092781-appb-000003
为
Figure PCTCN2019092781-appb-000004
的回归系数，Z -k表示所述待预测月份之前第k个月份的海冰面积，q -k为Z -k的回归系数，O表示所述若干个月份的数量，M表示所述观测点的数量，N表示所述气候变量的数量，ε为常数。
具体的，利用CEN算法对获取到的历史数据进行拟合运算，得到的预设因果影响网络模型如下：
Figure PCTCN2019092781-appb-000005
其中，Y表示所述待预测月份的海冰面积，
Figure PCTCN2019092781-appb-000006
表示所述待预测月份之前第k个月份的第i个观测点的第j个气候变量，
Figure PCTCN2019092781-appb-000007
为
Figure PCTCN2019092781-appb-000008
的回归系数，Z -k表示所述待预测月份之前第k个月份的海冰面积，q -k为Z -k的回归系数，O表示所述若干个月份的数量，M表示所述观测点的数量，N表示所述气候变量的数量，ε为常数。
本申请实施例提供的基于因果影响网络算法的极地海冰面积预测方法，通过历史数据对因果影响网络模型进行训练，然后将当年中的前几个月份的观测数据输入至训练好的因果影响网络模型，即可准确的预测出目标月份的海冰面积，该方法操作简便，系统复杂度低，预测效率高。
在以上各实施例的基础上，进一步地，所述气候变量至少包括海平面气压、表面气温、海表温度、近地面纬向风、近地面经向风、近地面向下短波辐射和近地面向下长波辐射中的任一种。
具体的，在获取观测数据时，观测点的数量和气候变量数量需要根据具体的实际情况而定。根据CEN算法对获取到的历史数据进行拟合分析后，得到海平面气压、表面气温、海表温度、近地面纬向风、近地面经向风、近地 面向下短波辐射和近地面向下长波辐射这些气候变量对目标月份海冰面积的影响较为显著，所以，气候变量至少包括海平面气压、表面气温、海表温度、近地面纬向风、近地面经向风、近地面向下短波辐射和近地面向下长波辐射中的任一种。以便获取更加准确的预测值。
本申请实施例提供的基于因果影响网络算法的极地海冰面积预测方法，通过历史数据对因果影响网络模型进行训练，然后将当年中的前几个月份的观测数据输入至训练好的因果影响网络模型，即可准确的预测出目标月份的海冰面积，该方法操作简便，系统复杂度低，预测效率高。
在以上各实施例的基础上，进一步地，所述若干个月份的数量为8。
具体的，根据CEN算法对获取到的历史数据进行拟合分析后，得到最大超前时间为8个月时，各影响因子对目标月份海冰面积的影响最为显著，得到的目标月份海冰面积的预测值与观测值最为一致。因此，所述若干个月份的数量为设置为8，以便获取更加准确的预测值。
针对上述例子，我们想要预测9月份的海冰面积，那么我们选择1-8月份这8个月的观测数据输入至训练好的因果影响网络模型，得到的9月份的海冰面积的预测值与真实值最为接近。
本申请实施例提供的基于因果影响网络算法的极地海冰面积预测方法，通过历史数据对因果影响网络模型进行训练，然后将当年中的前几个月份的观测数据输入至训练好的因果影响网络模型，即可准确的预测出目标月份的海冰面积，该方法操作简便，系统复杂度低，预测效率高。
在以上各实施例的基础上，进一步地，所述待预测月份为9月份。
具体的，经过对历史观测数据的分析，可知，每年9月份北极海冰的覆盖面积最小，对最小海冰覆盖面积的预测意义重大。因此，设置待预测月份为9月份。
本申请实施例提供的基于因果影响网络算法的极地海冰面积预测方法，通过历史数据对因果影响网络模型进行训练，然后将当年中的前几个月份的观测数据输入至训练好的因果影响网络模型，即可准确的预测出目标月份的海冰面积，该方法操作简便，系统复杂度低，预测效率高。
在以上各实施例的基础上，进一步地，所述若干个月份的海冰面积为8月份的海冰面积。
具体的，经过对历史观测数据的分析，可知，每年9月份北极海冰的覆盖面积最小，对最小海冰覆盖面积的预测意义重大。因此，设置待预测月份为9月份。
根据CEN算法对获取到的历史数据进行拟合分析后，得到8月份的海冰面积与9月份海冰面积有强自相关性。因此，为了获得更加准确的9月份海冰面积的预测值，所述若干个月份的海冰面积为8月份的海冰面积。
本申请实施例提供的基于因果影响网络算法的极地海冰面积预测方法，通过历史数据对因果影响网络模型进行训练，然后将当年中的前几个月份的观测数据输入至训练好的因果影响网络模型，即可准确的预测出目标月份的海冰面积，该方法操作简便，系统复杂度低，预测效率高。
图2为依照本申请实施例的基于因果影响网络算法的极地海冰面积预测装置示意图，如图2所示，本申请实施例提供一种基基于因果影响网络算法的极地海冰面积预测装置，用于完成上述实施例中所述的方法，具体包括获取模块201和预测模块202，其中，
获取模块201用于获取待预测月份之前的若干个月份的海冰面积，以及所述若干个月份中每一月份的在每一观测点的每一气候变量；
预测模块202用于将所述若干个月份中每一月份的海冰面积和所述若干个月份中每一月份的在每一观测点的每一气候变量输入至预设因果影响网络模型，输出所述待预测月份的海冰面积。
本申请实施例提供一种基于因果影响网络算法的极地海冰面积预测装置，用于完成上述实施例中所述的方法，通过本实施例提供的装置完成上述实施例中所述的方法的具体步骤与上述实施例相同，此处不再赘述。
本申请实施例提供的基于因果影响网络算法的极地海冰面积预测装置，通过历史数据对因果影响网络模型进行训练，然后将当年中的前几个月份的观测数据输入至训练好的因果影响网络模型，即可准确的预测出目标月份的海冰面积，该方法操作简便，系统复杂度低，预测效率高。
图3为本申请实施例提供的用于基于因果影响网络算法的极地海冰面积预测的电子设备的结构示意图，如图3所示，所述设备包括：处理器301、存储器302和总线303；
其中，处理器301和存储器302通过所述总线303完成相互间的通信；
处理器301用于调用存储器302中的程序指令，以执行上述各方法实施例所提供的方法，例如包括：
获取待预测月份之前的若干个月份的海冰面积，以及所述若干个月份中每一月份的在每一观测点的每一气候变量；
将所述若干个月份中每一月份的海冰面积和所述若干个月份中每一月份的在每一观测点的每一气候变量输入至预设因果影响网络模型，输出所述待预测月份的海冰面积。
本申请实施例公开一种计算机程序产品，所述计算机程序产品包括存储在非暂态计算机可读存储介质上的计算机程序，所述计算机程序包括程序指令，当所述程序指令被计算机执行时，计算机能够执行上述各方法实施例所提供的方法，例如包括：
获取待预测月份之前的若干个月份的海冰面积，以及所述若干个月份中每一月份的在每一观测点的每一气候变量；
将所述若干个月份中每一月份的海冰面积和所述若干个月份中每一月份的在每一观测点的每一气候变量输入至预设因果影响网络模型，输出所述待预测月份的海冰面积。
本申请实施例提供一种非暂态计算机可读存储介质，所述非暂态计算机可读存储介质存储计算机指令，所述计算机指令使所述计算机执行上述各方法实施例所提供的方法，例如包括：
获取待预测月份之前的若干个月份的海冰面积，以及所述若干个月份中每一月份的在每一观测点的每一气候变量；
将所述若干个月份中每一月份的海冰面积和所述若干个月份中每一月份的在每一观测点的每一气候变量输入至预设因果影响网络模型，输出所述待预测月份的海冰面积。
本领域普通技术人员可以理解：实现上述方法实施例的全部或部分步骤可以通过程序指令相关的硬件来完成，前述的程序可以存储于一计算机可读取存储介质中，该程序在执行时，执行包括上述方法实施例的步骤；而前述的存储介质包括：ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
以上所描述的装置及设备等实施例仅仅是示意性的，其中所述作为分离 部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。本领域普通技术人员在不付出创造性的劳动的情况下，即可以理解并实施。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到各实施方式可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件。基于这样的理解，上述技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品可以存储在计算机可读存储介质中，如ROM/RAM、磁碟、光盘等，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行各个实施例或者实施例的某些部分所述的方法。
最后应说明的是：以上实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围。

Claims (10)

  1. 一种基于因果影响网络算法的极地海冰面积预测方法，其特征在于，包括：
    获取待预测月份之前的若干个月份的海冰面积，以及所述若干个月份中每一月份的在每一观测点的每一气候变量；
    将所述若干个月份中每一月份的海冰面积和所述若干个月份中每一月份的在每一观测点的每一气候变量输入至预设因果影响网络模型，输出所述待预测月份的海冰面积。
  2. 根据权利要求1所述的方法，其特征在于，获取所述预设因果影响网络模型的具体步骤如下：
    获取若干年的历史数据，每一年的历史数据包括当年目标月份之前的若干个月份的海冰面积，以及所述若干个月份中每一月份的在每一观测点的每一气候变量，还包括当年目标月份的海冰面积；
    利用因果影响网络算法对所述历史数据进行拟合运算，获取所述预设因果影响网络模型。
  3. 根据权利要求1所述的方法，其特征在于，所述预设因果影响网络模型如下：
    Figure PCTCN2019092781-appb-100001
    其中，Y表示所述待预测月份的海冰面积，
    Figure PCTCN2019092781-appb-100002
    表示所述待预测月份之前第k个月份的第i个观测点的第j个气候变量，
    Figure PCTCN2019092781-appb-100003
    为
    Figure PCTCN2019092781-appb-100004
    的回归系数，Z -k表示所述待预测月份之前第k个月份的海冰面积，q -k为Z -k的回归系数，O表示所述若干个月份的数量，M表示所述观测点的数量，N表示所述气候变量的数量，ε为常数。
  4. 根据权利要求1所述的方法，其特征在于，所述气候变量至少包括海平面气压、表面气温、海表温度、近地面纬向风、近地面经向风、近地面向下短波辐射和近地面向下长波辐射中的任一种。
  5. 根据权利要求1所述的方法，其特征在于，所述若干个月份的数量为8。
  6. 根据权利要求1所述的方法，其特征在于，所述待预测月份为9月份。
  7. 根据权利要求6所述的方法，其特征在于，所述若干个月份的海冰面积为8月份的海冰面积。
  8. 一种基于因果影响网络算法的极地海冰面积预测装置，其特征在于，包括：
    获取模块，用于获取待预测月份之前的若干个月份的海冰面积，以及所述若干个月份中每一月份的在每一观测点的每一气候变量；
    预测模块，用于将所述若干个月份中每一月份的海冰面积和所述若干个月份中每一月份的在每一观测点的每一气候变量输入至预设因果影响网络模型，输出所述待预测月份的海冰面积。
  9. 一种用于基于因果影响网络算法的极地海冰面积预测的电子设备，其特征在于，包括：
    存储器和处理器，所述处理器和所述存储器通过总线完成相互间的通信；所述存储器存储有可被所述处理器执行的程序指令，所述处理器调用所述程序指令能够执行如权利要求1至7任一所述的方法。
  10. 一种计算机可读存储介质，其上存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现如权利要求1至7任一所述的方法。
PCT/CN2019/092781 2018-06-21 2019-06-25 基于因果影响网络算法的极地海冰面积预测方法及装置 WO2019242779A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810646657.3 2018-06-21
CN201810646657.3A CN108876041B (zh) 2018-06-21 2018-06-21 基于因果影响网络算法的极地海冰面积预测方法及装置

Publications (1)

Publication Number Publication Date
WO2019242779A1 true WO2019242779A1 (zh) 2019-12-26

Family

ID=64340754

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/092781 WO2019242779A1 (zh) 2018-06-21 2019-06-25 基于因果影响网络算法的极地海冰面积预测方法及装置

Country Status (2)

Country Link
CN (1) CN108876041B (zh)
WO (1) WO2019242779A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108876041B (zh) * 2018-06-21 2020-07-28 清华大学 基于因果影响网络算法的极地海冰面积预测方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5704017A (en) * 1996-02-16 1997-12-30 Microsoft Corporation Collaborative filtering utilizing a belief network
CN106874686A (zh) * 2017-03-03 2017-06-20 北京北青厚泽数据科技有限公司 指标关系网构建方法
CN107590551A (zh) * 2017-08-16 2018-01-16 南京航空航天大学 一种大气pm2.5浓度变化的预测方法
CN108876041A (zh) * 2018-06-21 2018-11-23 清华大学 基于因果影响网络算法的极地海冰面积预测方法及装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5704017A (en) * 1996-02-16 1997-12-30 Microsoft Corporation Collaborative filtering utilizing a belief network
CN106874686A (zh) * 2017-03-03 2017-06-20 北京北青厚泽数据科技有限公司 指标关系网构建方法
CN107590551A (zh) * 2017-08-16 2018-01-16 南京航空航天大学 一种大气pm2.5浓度变化的预测方法
CN108876041A (zh) * 2018-06-21 2018-11-23 清华大学 基于因果影响网络算法的极地海冰面积预测方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LI, SHA ET AL.: "Precursors of September Arctic Sea-Ice Extent Based on Causal Effect Networks", ATMOSPHERE, vol. 9, no. 11, 9 November 2019 (2019-11-09), pages 3 *

Also Published As

Publication number Publication date
CN108876041B (zh) 2020-07-28
CN108876041A (zh) 2018-11-23

Similar Documents

Publication Publication Date Title
Engelbrecht et al. Multi-scale climate modelling over Southern Africa using a variable-resolution global model
CN113538391B (zh) 一种基于Yolov4和热红外图像的光伏缺陷检测方法
CN110929963A (zh) 风速预测方法、装置以及存储介质
CN103489038A (zh) 基于lm-bp神经网络的光伏超短期功率预测方法
CN104732296A (zh) 一种分布式光伏输出功率短期预测模型建模方法
CN112766600B (zh) 一种城市区域人群流量预测方法及系统
Saxena et al. A review study of weather forecasting using artificial neural network approach
CN105930635A (zh) 一种内陆河干旱绿洲区干旱评价系统
JP2017135964A (ja) 太陽光発電システムの動作
CN104598765A (zh) 一种基于弹性自适应神经网络的建筑物能耗预测方法
JP6086875B2 (ja) 発電量予測装置および発電量予測方法
CN117216526B (zh) 基于人工智能的光伏出力预测方法及系统
CN116595394A (zh) 风速修正模型的训练方法及风速预测方法、设备和介质
WO2019242779A1 (zh) 基于因果影响网络算法的极地海冰面积预测方法及装置
CN113984198B (zh) 一种基于卷积神经网络的短波辐射预测方法及系统
CN103778306B (zh) 一种基于ei和逐步消减法的传感器优化布设方法
Jin et al. Deep learning for seasonal precipitation prediction over China
CN105956708A (zh) 基于灰关联时间序列的短期风速预测方法
CN116706903B (zh) 光伏发电量预测方法、装置、设备及介质
CN107194507A (zh) 一种基于组合支持向量机的风电场短期风速预测方法
CN110046756A (zh) 基于小波去噪与Catboost的短时天气预报方法
CN116341391B (zh) 基于STPM-XGBoost模型的降水预测方法
CN116663727A (zh) 一种光伏功率预测方法及系统
CN114819257A (zh) 辐照计算方法、装置及设备
CN110991083B (zh) 一种光伏电站模型确定方法、装置、设备及储存介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19823538

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19823538

Country of ref document: EP

Kind code of ref document: A1