Waterlogging risk level prediction method and device, equipment, storage medium and product

Abstract

The invention discloses a waterlogging risk level prediction method and device, equipment, a storage medium and a product. The current grade of each target waterlogging condition index is determined, and the current grade of the target waterlogging condition index is determined based on the obtained current value of the target waterlogging condition index; the current grade of each target waterlogging condition index is input into a risk prediction model to obtain a current waterlogging condition risk grade, and the risk prediction model is a Bayesian network which takes each target waterlogging condition index as a node and is constructed based on conditional probability distribution among the target waterlogging condition indexes; the conditional probability distribution among the target waterlogging condition indexes is determined based on the historical grade of each target waterlogging condition index. And the accuracy and efficiency of waterlogging risk level prediction are effectively improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of power grid risk management, and in particular, to a method, device, equipment, storage medium and product for predicting the risk level of waterlogging Background Art

[0002] The substations of the power grid are characterized by a large number, wide distribution and significant climate differences, which reflect the key supporting role of the power grid in aspects such as industrial production, enterprise operation and residents' lives. Against the background of the intensification of global climate change, meteorological conditions have become one of the key factors affecting the safety of the power grid and power supply guarantee. In particular, meteorological factors such as rainfall are extremely likely to cause waterlogging in substations, which in turn threatens the safe and reliable operation of substations and the overall stability of the power grid. At present, it is difficult to quantitatively evaluate the risk of substation waterlogging caused by meteorological factors such as rainfall, which poses a great challenge to the flood control work of substations.

[0003] However, the existing methods for predicting the risk level of waterlogging have problems of low accuracy and efficiency, and it is difficult to meet the actual needs of substation flood control work. Summary of the Invention

[0004] The present invention provides a method, device, equipment, storage medium and product for predicting the risk level of waterlogging, so as to solve the problems of low accuracy and efficiency existing in the existing methods for predicting the risk level of waterlogging.

[0005] According to one aspect of the present invention, a method for predicting the risk level of waterlogging is provided, including:

[0006] Determine the current level of each target waterlogging index, wherein the current level of the target waterlogging index is determined based on the currently obtained value of the target waterlogging index;

[0007] Input the current levels of the target waterlogging indices into a risk prediction model to obtain the current waterlogging risk level, wherein the risk prediction model is a Bayesian network constructed with the target waterlogging indices as nodes based on the conditional probability distribution between the target waterlogging indices, and the conditional probability distribution between the target waterlogging indices is determined based on the historical levels of the target waterlogging indices.

[0008] According to another aspect of the present invention, a device for predicting the risk level of waterlogging is provided, including:

[0009] An index level determination module for determining the current level of each target waterlogging index, wherein the current level of the target waterlogging index is determined based on the currently obtained value of the target waterlogging index;

[0010] A risk level determination module for inputting the current levels of the respective target waterlogging condition indicators into a risk prediction model to obtain the current waterlogging risk level, where the risk prediction model is a Bayesian network constructed with the respective target waterlogging condition indicators as nodes based on the conditional probability distribution among the respective target waterlogging condition indicators, and the conditional probability distribution among the respective target waterlogging condition indicators is determined based on the historical levels of the respective target waterlogging condition indicators.

[0011] According to another aspect of the present invention, there is provided an electronic device, which includes:

[0012] At least one processor; and

[0013] A memory communicatively connected to the at least one processor; wherein,

[0014] The memory stores a computer program executable by the at least one processor, and when the computer program is executed by the at least one processor, the at least one processor is enabled to execute the waterlogging risk level prediction method according to any embodiment of the present invention.

[0015] According to another aspect of the present invention, there is provided a computer-readable storage medium storing computer instructions for causing a processor to implement the waterlogging risk level prediction method according to any embodiment of the present invention when executed.

[0016] According to another aspect of the present invention, there is provided a computer program product including a computer program, which implements the waterlogging risk level prediction method according to any embodiment of the present invention when executed by a processor.

[0017] The technical solution provided by the embodiments of the present invention determines the current levels of the respective target waterlogging condition indicators, where the current levels of the target waterlogging condition indicators are determined based on the currently obtained numerical values of the target waterlogging condition indicators; the current levels of the respective target waterlogging condition indicators are input into a risk prediction model to obtain the current waterlogging risk level, where the risk prediction model is a Bayesian network constructed with the respective target waterlogging condition indicators as nodes based on the conditional probability distribution among the respective target waterlogging condition indicators, and the conditional probability distribution among the respective target waterlogging condition indicators is determined based on the historical levels of the respective target waterlogging condition indicators. Through the above technical solution, the currently obtained numerical values of the respective target waterlogging condition indicators are quantified into current levels, ensuring the accuracy and processing efficiency of the input data; furthermore, a risk prediction model constructed based on a Bayesian network is used to predict the current waterlogging risk level. This model takes the respective target waterlogging condition indicators as nodes and is constructed based on the conditional probability distribution determined by the historical levels of these indicators, fully considering the causal relationship and mutual dependence among the respective target waterlogging condition indicators, effectively improving the accuracy and efficiency of the waterlogging risk level prediction.

[0018] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0020] Figure 1 is a flowchart of a method for predicting the risk level of waterlogging

[0021] Figure 2 is a schematic diagram of the process of constructing a target waterlogging network provided by an embodiment of the present invention;

[0022] Figure 3 is a schematic structural diagram of a device for predicting the risk level of waterlogging provided by Embodiment 2 of the present invention;

[0023] Figure 4 is a schematic structural diagram of an electronic device provided by Embodiment 3 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] In order to enable those skilled in the art to better understand the solutions of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, rather than all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the description, claims and above-mentioned drawings of the present invention are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0026] Embodiment 1

[0027] Figure 1 FIG. is a flowchart of a method for predicting the risk level of waterlogging provided in Embodiment 1 of the present invention. This embodiment is applicable to the situation of predicting the risk level of waterlogging in a substation. This method can be executed by a device for predicting the risk level of waterlogging. The device for predicting the risk level of waterlogging can be implemented in the form of hardware and / or software, and the device for predicting the risk level of waterlogging can be configured in an electronic device. As Figure 1 shown, the method includes:

[0028] S110. Determine the current level of each target waterlogging index, where the current level of the target waterlogging index is determined based on the currently obtained value of the target waterlogging index.

[0029] In this embodiment, the target waterlogging index can be understood as a key index for evaluating the current risk level of waterlogging. These indexes are determined by screening based on prior knowledge (such as historical data and expert experience, etc.). For example, the target waterlogging index can be rainfall, rainfall time, terrain, etc. The currently obtained value of the target waterlogging index can be understood as the specific value obtained in real time by a data acquisition device.

[0030] Specifically, the currently obtained values of each target waterlogging index are collected in real time through a data acquisition device installed in a waterlogging risk area such as a substation; furthermore, according to the preset mapping rules corresponding to each target waterlogging index, these currently obtained values are respectively mapped to the corresponding current levels. Among them, the preset mapping rule can be a rule that is preset and used to map the value of the waterlogging index to the corresponding level. Among them, one target waterlogging index corresponds to one preset mapping rule.

[0031] S120. Input the current levels of the respective target waterlogging situation indicators into a risk prediction model to obtain the current waterlogging risk level. Here, the risk prediction model is a Bayesian network constructed with the respective target waterlogging situation indicators as nodes, based on the conditional probability distribution among the respective target waterlogging situation indicators, and the conditional probability distribution among the respective target waterlogging situation indicators is determined based on the historical levels of the respective target waterlogging situation indicators.

[0032] In this embodiment, the risk prediction model can be understood as a model constructed based on a Bayesian network for predicting the waterlogging risk level. In this model, the respective target waterlogging situation indicators are set as nodes, and the directed edges between the nodes represent the conditional probability distribution among these indicators. The historical level of a target waterlogging situation indicator can be understood as the level determined according to the historical values of the obtained target waterlogging situation indicator.

[0033] Specifically, input the current levels of the respective target waterlogging situation indicators into a pre-constructed risk prediction model. The risk prediction model will calculate and output the current waterlogging risk level based on the current levels of the respective target waterlogging situation indicators and the conditional probability distribution among the respective target waterlogging situation indicators.

[0034] The technical solution provided in the first embodiment of the present invention determines the current levels of the respective target waterlogging situation indicators, where the current level of the target waterlogging situation indicator is determined based on the currently obtained value of the target waterlogging situation indicator; input the current levels of the respective target waterlogging situation indicators into a risk prediction model to obtain the current waterlogging risk level, where the risk prediction model is a Bayesian network constructed with the respective target waterlogging situation indicators as nodes, based on the conditional probability distribution among the respective target waterlogging situation indicators, and the conditional probability distribution among the respective target waterlogging situation indicators is determined based on the historical levels of the respective target waterlogging situation indicators. Through the above technical solution, the current values of the respective target waterlogging situation indicators are quantified into current levels, ensuring the accuracy and processing efficiency of the input data; furthermore, a risk prediction model constructed based on a Bayesian network is used to predict the current waterlogging risk level. This model is constructed with the respective target waterlogging situation indicators as nodes, based on the conditional probability distribution determined according to the historical levels of these indicators, fully considering the causal relationship and mutual dependence among the respective target waterlogging situation indicators, effectively improving the accuracy and efficiency of predicting the waterlogging risk level.

[0035] In some embodiments, determining the current levels of the target waterlogging condition indicators includes: obtaining the current values of the target waterlogging condition indicators; based on the level division criteria of the target waterlogging condition indicators, determining the target preset numerical interval in which the current values of the target waterlogging condition indicators are located, and determining the level corresponding to the target preset numerical interval as the current level of the target waterlogging condition indicators, where the level division criteria include multiple preset numerical intervals, and each preset numerical interval corresponds to one level. Through the above technical solution, the level division of the current values of the target waterlogging condition indicators is realized, the ambiguity caused by numerical continuity is reduced, a clear and quantitative basis is provided for waterlogging risk assessment, the accuracy of the data is ensured, and a foundation is laid for further improving the accuracy of waterlogging risk prediction.

[0036] In this embodiment, one target waterlogging condition indicator corresponds to one level division criterion. The level division criterion can be understood as a pre-set mapping rule for determining the current level corresponding to the current value of the target waterlogging condition indicator. The level division criterion includes multiple preset numerical intervals, and each preset numerical interval corresponds to one level. Among them, the preset numerical interval can be understood as the specific numerical range defined in the level division criterion. The target preset numerical interval can be understood as the preset numerical interval in which the current value of the target waterlogging condition indicator is located. It should be noted that the specific number of preset numerical intervals is not limited in this embodiment. For example, it can be 4, 5, 6 or more. This embodiment takes the level division criteria corresponding to each target waterlogging condition indicator as including 4 preset numerical intervals as an example for illustration. Among them, the 4 preset numerical intervals correspond to the four levels of low, medium, high and severe respectively.

[0037] Specifically, for each target waterlogging condition indicator among the target waterlogging condition indicators, the overall numerical range of the current target waterlogging condition indicator is divided into multiple preset numerical ranges through discretization processing, and each preset numerical range corresponds to a specific level; based on these preset numerical intervals and their corresponding levels, the level division criterion corresponding to the current target waterlogging condition indicator is determined.

[0038] Obtain the current values of the target waterlogging condition indicators; according to the level division criteria of the target waterlogging condition indicators, determine the target preset numerical interval in which the current values of the target waterlogging condition indicators are located, and further, determine the level corresponding to the target preset numerical interval as the current level of the target waterlogging condition indicators.

[0039] Exemplarily, assume that the target waterlogging condition indicator is rainfall, and its overall numerical range is from 0 mm / h to infinity. Through discretization processing, the overall numerical range can be divided into 4 preset numerical intervals:

[0040] Low level: 0 - 10 mm / h

[0041] Medium level: 10 - 20 mm / h

[0042] High level: 20 - 30 mm / hour

[0043] Severe level: above 30 mm / hour

[0044] These 4 preset numerical ranges and their corresponding levels together constitute the grading standard for rainfall levels.

[0045] Assume that the current value of rainfall is 25 mm / hour. According to the grading standard, 25 mm / hour falls within the preset numerical range of 20 - 30 mm / hour. Therefore, its corresponding current level is the high level.

[0046] In some embodiments, the risk prediction model is determined by the following method: determining historical waterlogging data corresponding to different time points, where the historical waterlogging data includes the historical levels and historical waterlogging risk levels of each target waterlogging index; based on the historical waterlogging data and the initial waterlogging network corresponding to different time points, determining the target waterlogging network, where the initial waterlogging network is determined based on each target waterlogging index and prior knowledge, and the nodes of the target waterlogging network are the target waterlogging indexes; calculating the conditional probability distribution of each node in the target waterlogging network given the values of the parent nodes based on the historical waterlogging data corresponding to different time points; and determining the risk prediction model based on the conditional probability distribution of each node in the target waterlogging network.

[0047] In this embodiment, the historical waterlogging data includes the historical levels and historical waterlogging risk levels of each target waterlogging index, where the historical waterlogging risk level can be understood as the waterlogging risk level corresponding to a certain past time point.

[0048] Specifically, use data acquisition devices to collect the original data related to waterlogging in waterlogging risk areas such as inside and around the substation at different time points, as well as the historical waterlogging risk levels corresponding to each time point. Among them, the original data contains multiple waterlogging indexes. For example, the original data can include meteorological data, geographical information data, and substation flood control data. Among them, the meteorological data includes waterlogging indexes such as rainfall, rainfall time, and wind speed. The geographical information data contains waterlogging indexes such as terrain, land use status, river distance, slope, and altitude. The substation flood control data contains waterlogging indexes such as the drainage pipe and drainage pump drainage capacity inside the substation; the original data can also include historical flood disaster information, inundation depth, inundation range, and inundation duration and other waterlogging indexes.

[0049] Considering that the impact of each waterlogging index on waterlogging is different, based on prior knowledge, the original data can be screened from three aspects: hazard-causing factors, disaster-bearing environments, and disaster-affecting bodies, so as to determine the waterlogging indices whose impact on waterlogging is greater than a preset threshold, and these waterlogging indices are determined as target waterlogging indices. A data set is established based on the historical levels of each target waterlogging index and the historical waterlogging risk levels corresponding to each time point. Among them, each column in this data set corresponds to a target waterlogging index or historical waterlogging risk, and each row records the historical level corresponding to each target waterlogging index and the corresponding historical waterlogging risk level at this time point. Among them, the historical level is obtained by discretizing the historical values corresponding to each target waterlogging index to reduce the complexity of model construction. Among them, the data set can be expressed as:

[0050] D = {D1, D2, D3, …, D N , Y}

[0051] Among them, D represents the data set, D N represents the Nth target waterlogging index. Among them, each target waterlogging index corresponds to the historical levels at each time point, and Y represents the historical waterlogging risk. Among them, the historical waterlogging risk corresponds to the risk levels at each time point.

[0052] Taking each target waterlogging index and historical waterlogging risk as nodes respectively, a complete edge-less graph is constructed; furthermore, based on prior knowledge, the associations between each node are added to form an initial waterlogging network; based on the historical waterlogging data corresponding to different time points, the causal relevance of each node pair in the initial waterlogging network under actual conditions is calculated, and the initial waterlogging network is cyclically updated based on the causal relevance between each node pair, and finally the target waterlogging network is obtained. It should be noted that the nodes of the target waterlogging network are target waterlogging indices.

[0053] Furthermore, based on the historical waterlogging data corresponding to different time points, the conditional probability distribution of each node pair in the target waterlogging network under the condition of the value of the given father node can be calculated by using a preset probability distribution algorithm, and based on the conditional probability distribution of each node in the target waterlogging network, a risk prediction model is determined. It should be noted that the specific form of the preset probability distribution algorithm is not limited in this embodiment. For example, it can be the Bayesian algorithm or the Markov chain Monte Carlo algorithm, etc.

[0054] Through the above technical solutions, by using the historical waterlogging data of different time points and combining the initial waterlogging network determined based on prior knowledge, a target waterlogging network that can better reflect the actual law is obtained; calculating the conditional probability distribution of each node in the target waterlogging network under the condition of the value of the given father node can accurately reflect the causal relationship and dependence between each waterlogging index. Based on the risk prediction model determined thereby, the accuracy and reliability of waterlogging risk prediction are significantly improved.

[0055] Exemplarily,Figure 2 It is a schematic diagram of a process for constructing a target waterlogging situation network provided by an embodiment of the present invention. As Figure 2 shown, m target waterlogging situation indicators (X1, X2, X3, …, X m ) and historical waterlogging risks Y are respectively used as nodes to construct a complete edge-free graph 210; furthermore, based on prior knowledge, associations between each node are added, such as the edge between X1 and Y, to form an initial waterlogging situation network 220; based on historical waterlogging situation data corresponding to different time points, the causal relevance of each node pair in the initial waterlogging situation network under actual conditions is calculated, and the initial waterlogging situation network is updated based on the causal relevance between each node pair to obtain an intermediate waterlogging situation network 230. In 230, edges between X1 and X3, and between X2 and Y are established. Furthermore, the intermediate waterlogging situation network 230 is used as the initial waterlogging situation network, and the causal relevance between each node pair is calculated cyclically. Finally, the target waterlogging situation network 240 is determined. In 240, the edges between X2 and X m , and between X3 and Y are determined.

[0056] In some embodiments, determining the target waterlogging situation network based on historical waterlogging situation data corresponding to different time points and the initial waterlogging situation network includes: based on historical waterlogging situation data corresponding to different time points, using a preset pointwise causal algorithm to determine alternative operations corresponding to each node pair in the initial waterlogging situation network, where the alternative operations include adding an edge, subtracting an edge, or reversing an edge; for each node pair among the node pairs, using a preset scoring function to calculate the scoring value of the alternative waterlogging situation network obtained after the current node pair performs the corresponding alternative operation; determining the maximum scoring value among the scoring values of the alternative waterlogging situation networks; returning to the relevant steps of using a preset pointwise causal algorithm based on historical waterlogging situation data corresponding to different time points to determine the alternative operations of each node pair in the initial waterlogging situation network until the maximum scoring value determined this time is less than or equal to the maximum scoring value determined last time, and determining the alternative waterlogging situation network corresponding to the maximum scoring value determined last time as the target waterlogging situation network; where, after each determination of the maximum scoring value, if the maximum scoring value determined this time is greater than the maximum scoring value determined last time, the initial waterlogging situation network is updated to the alternative waterlogging situation network corresponding to the maximum scoring value determined this time.

[0057] In this embodiment, the preset pointwise causal algorithm can be understood as an algorithm that is preset to determine alternative operations between node pairs. Among them, the preset pointwise causal algorithm can be expressed as:

[0058] PC(x i →x j ∣∣y) = H(x j ∣∣y) - H(x j ∣∣x i , y)

[0059] Among them, xi The node representing the i-th target waterlogging index; x j The node representing the j-th target waterlogging index; y represents the node of waterlogging risk; PC(x i →x j ∣∣y) represents the pointwise causality between x i and x j , that is, given the value of y, the causal influence of x i on x j . The larger this value is, the greater the causal influence of x i on x j ; H(x j ∣∣y) represents the uncertainty of x j given the value of y, and H(x j ∣∣x i ,y) represents the uncertainty of x i given the values of y and x j .

[0060] In this embodiment, the alternative operations include adding an edge, deleting an edge, or reversing an edge. Among them, adding an edge means adding a new edge in the current waterlogging network, representing the existence of a dependency relationship between two node pairs; deleting an edge means deleting an edge in the current waterlogging network, representing canceling the dependency relationship between two node pairs, and reversing an edge means performing a reverse operation on the edge between node pairs.

[0061] Among them, the preset scoring function is determined based on the Bayesian information criterion and is used to evaluate the quality of the waterlogging network structure. Among them, the preset scoring function is expressed as:

[0062]

[0063] Among them, G represents the current waterlogging network; Gn represents the set of possible waterlogging networks; k i represents the number of parameters of the i-th node in the current waterlogging network. Among them, the parameters of a node represent the combination number of the possible values of the current node and the possible values of its parent nodes. For a node without a parent node, the combination number is 0; N represents the number of target waterlogging indicators; D represents the data set; score(G) represents the scoring value of the current waterlogging network; log P(D∣G,Gn) represents the value of the log-likelihood function of the data set D under the current waterlogging network G and the set of possible waterlogging networks Gn, which is used to measure the fitting degree of the network G to the data set D. The larger the log-likelihood value is, the better the network fits the data set; X i represents the i-th node in the current waterlogging network.

[0064] Among them, the Bayesian information criterion is expressed as:

[0065] BIC(G, Gn | D) = -2 · log P(D | G, Gn) + k · log N

[0066] Among them, BIC(G, Gn | D) represents the value of the Bayesian Information Criterion, which is used to evaluate the pros and cons of the current waterlogging network G under the given dataset D and the set Gn of possible waterlogging networks. The smaller this value is, the better the network performs in balancing the goodness of fit and complexity.

[0067] Specifically, based on the historical waterlogging data corresponding to different time points, using the preset pointwise causality algorithm, calculate the pointwise causality of each node pair, and based on the pointwise causality, determine the alternative operations corresponding to each node pair in the initial waterlogging network. The alternative operations include adding an edge, deleting an edge, or reversing an edge; then, for each node pair among the node pairs, use the preset scoring function to calculate the scoring value of the alternative waterlogging network obtained after the corresponding alternative operation is performed on the current node pair; select the maximum scoring value from the scoring values of the alternative waterlogging networks; if this is the first iteration, or the maximum scoring value determined this time is greater than the maximum scoring value determined last time, then update the initial waterlogging network to the alternative waterlogging network corresponding to the maximum scoring value determined this time, and then, iteratively execute the above process; if the maximum scoring value determined this time is less than or equal to the maximum scoring value determined last time, then determine the alternative waterlogging network corresponding to the maximum scoring value determined last time as the target waterlogging network, and end the iteration. This target waterlogging network can be represented as: G * :[[]]

[0068]

[0069] Optionally, after determining the target waterlogging network, relevant personnel can determine the rationality of the structure, and then update the target waterlogging network.

[0070] Optionally, when using the preset pointwise causality algorithm to calculate the pointwise causality of each node pair and based on the pointwise causality to determine the alternative operations corresponding to each node pair in the initial waterlogging network, first determine whether there is an edge between the node pairs. If there is, first judge whether the pointwise causality is less than or equal to 0. If it is less than or equal to 0, then delete it. Otherwise, further calculate the reverse pointwise causality, and when the reverse pointwise causality is greater than the pointwise causality, perform the operation of reversing the edge; if there is no edge between the two nodes, then perform the operation of adding an edge when the pointwise causality is greater than 0. Among them, the pointwise causality represents the causal influence of x on x given the value of y, and the reverse pointwise causality represents the causal influence of x on x given the value of y. i on x j 's causal influence, and the reverse pointwise causality represents the causal influence of x on x given the value of y j on x i 's causal influence.

[0071] Through the above technical solution, based on the historical waterlogging data at different time points, the alternative operations of each node pair in the initial waterlogging network are determined by using a preset pointwise causal algorithm, and the network score value after each alternative operation is calculated by using a preset scoring function. The waterlogging network structure is dynamically adjusted according to the score value, and the network structure with the highest score is gradually selected, so as to realize the determination of the optimal network structure as the target waterlogging network, thereby further significantly improving the accuracy and reliability of waterlogging risk prediction.

[0072] In some embodiments, calculating the conditional probability distribution of each node under the condition of the given values of its parent nodes based on the historical waterlogging data corresponding to different time points includes: determining the conditional probability table of each node in the target waterlogging network based on the historical waterlogging data corresponding to different time points, where the conditional probability table is used to represent the probability that each node takes a specific level under the condition of the given values of its parent nodes; using the Bayesian algorithm to calculate the specific values of the probabilities in the conditional probability table of each node according to the historical waterlogging data corresponding to different time points; and determining the conditional probability distribution of each node based on the specific values of the probabilities in the conditional probability table of each node. Through the above technical solution, the calculation of the conditional probability distribution of each node is realized, laying a foundation for accurately constructing a risk prediction model.

[0073] Specifically, based on the historical waterlogging data corresponding to different time points, for each node in the target waterlogging network, the conditional probability table of the current node is determined, where the conditional probability table is used to represent the probability that each node takes a specific level under the condition of the given values of its parent nodes. Taking node X i , whose parent node is Pa(X i ) as an example for illustration, as shown in Table 1:

[0074] Table 1 Conditional Probability Table of Node X i

[0075]

[0076]

[0077] Among them, v1, v2, v3, v4 represent the levels that node X i can take, u1, u2, u3, and u4 represent the values of the given parent node Pa(X i ), and P(X i = v j || Pa(X i ) = u j ), j ∈ [1, 4], represents the probability that X i = v j under the condition that Pa(X i ) = u j .

[0078] Furthermore, the Bayesian algorithm is used to calculate the specific values of the probabilities in the conditional probability table of each node according to the historical waterlogging data corresponding to different time points; based on the specific values of the probabilities in the conditional probability table of each node, the conditional probability distribution of each node is determined.

[0079] In some embodiments, the Bayesian algorithm is expressed as:

[0080]

[0081] where X i represents the i-th node, x i represents the level of the i-th node, Pa(X i ) represents the parent node of the i-th node, pa represents the level of the parent node, P(X i =x i |Pa(X i )=pa) represents the conditional probability that the node X i takes the value of x i when the value of the parent node Pa(X i ) is pa, count(X i =x i ,Pa(X i )=pa) represents the number of X i =x i and Pa(X i )=pa, and count(Pa(X i )=pa) represents the number of Pa(X i )=pa. Through the above technical solution, the calculation of the conditional probability distribution of each node is effectively simplified, and the modeling efficiency of the risk prediction model is further improved.

[0082] Embodiment 2

[0083] Figure 3 is a schematic structural diagram of a waterlogging risk level prediction device provided by the second embodiment of the present invention. As Figure 3 shown, the device includes:

[0084] An index level determination module 21, configured to determine the current level of each target waterlogging index, where the current level of the target waterlogging index is determined based on the currently obtained value of the target waterlogging index;

[0085] A risk level determination module 22, configured to input the current levels of the respective target waterlogging condition indicators into a risk prediction model to obtain the current waterlogging risk level, where the risk prediction model is a Bayesian network constructed based on the conditional probability distribution among the respective target waterlogging condition indicators with the respective target waterlogging condition indicators as nodes, and the conditional probability distribution among the respective target waterlogging condition indicators is determined based on the historical levels of the respective target waterlogging condition indicators.

[0086] The technical solution provided in the second embodiment of the present invention effectively improves the accuracy and efficiency of predicting the waterlogging risk level.

[0087] Optionally, the indicator level determination module 21 includes:

[0088] A numerical value acquisition unit, configured to acquire the current numerical values of the respective target waterlogging condition indicators;

[0089] An indicator level determination unit, configured to determine the target preset numerical value interval in which the current numerical values of the respective target waterlogging condition indicators are located based on the level division standard of the respective target waterlogging condition indicators, and determine the level corresponding to the target preset numerical value interval as the current level of the respective target waterlogging condition indicators, where the level division standard includes multiple preset numerical value intervals, and each preset numerical value interval corresponds to a level.

[0090] Optionally, the risk prediction model is determined by the following method:

[0091] Determine the historical waterlogging data corresponding to different time points, where the historical waterlogging data includes the historical levels of the respective target waterlogging condition indicators and the historical waterlogging risk levels;

[0092] Based on the historical waterlogging data corresponding to different time points and the initial waterlogging network, determine the target waterlogging network, where the initial waterlogging network is determined based on the respective target waterlogging condition indicators and prior knowledge, and the nodes of the target waterlogging network are the target waterlogging condition indicators;

[0093] Based on the historical waterlogging data corresponding to different time points, calculate the conditional probability distribution of each node in the target waterlogging network under the condition of the values of the parent nodes;

[0094] Based on the conditional probability distribution of each node in the target waterlogging network, determine the risk prediction model.

[0095] Optionally, the determining the target waterlogging network based on the historical waterlogging data corresponding to different time points and the initial waterlogging network includes:

[0096] Based on the historical waterlogging data corresponding to different time points, use a preset pointwise causal algorithm to determine the alternative operations corresponding to each node pair in the initial waterlogging network, where the alternative operations include adding an edge, deleting an edge, or reversing an edge;

[0097] For each node pair among the various node pairs, use a preset scoring function to calculate the scoring value of the alternative waterlogging network obtained after the current node pair performs the corresponding alternative operation.

[0098] Determine the maximum scoring value among the scoring values of the alternative waterlogging networks.

[0099] Return to the relevant steps of determining the alternative operations of each node pair in the initial waterlogging network by using the preset point - to - point causality algorithm based on the historical waterlogging data corresponding to different time points until the maximum scoring value determined this time is less than or equal to the maximum scoring value determined last time, and determine the alternative waterlogging network corresponding to the maximum scoring value determined last time as the target waterlogging network; where, after each determination of the maximum scoring value, if the maximum scoring value determined this time is greater than the maximum scoring value determined last time, update the initial waterlogging network to the alternative waterlogging network corresponding to the maximum scoring value determined this time.

[0100] Optionally, calculating the conditional probability distribution of each node under the given values of the parent nodes based on the historical waterlogging data corresponding to different time points includes:

[0101] Based on the historical waterlogging data corresponding to different time points, determine the conditional probability table of each node in the target waterlogging network, where the conditional probability table is used to represent the probability that each node takes a specific level under the given values of the parent nodes;

[0102] Use the Bayesian algorithm to calculate the specific values of the probabilities in the conditional probability table of each node according to the historical waterlogging data corresponding to different time points;

[0103] Based on the specific values of the probabilities in the conditional probability table of each node, determine the conditional probability distribution of each node.

[0104] Optionally, the Bayesian algorithm is expressed as:

[0105]

[0106] where, X i represents the i - th node, x i represents the level of the i - th node, Pa(X i ) represents the parent node of the i - th node, pa represents the level of the parent node, P(X i =x i ∣Pa(X i ) = pa) represents the conditional probability that the node X i takes the value x i under the condition that the value of the parent node Pa(X i ) is pa, count(X i =xi , Pa(X i ) = pa) represents X i = x i and the number of Pa(X i ) = pa, count(Pa(X i ) = pa) represents the number of Pa(X i ) = pa.

[0107] The waterlogging risk level prediction device provided by the embodiment of the present invention can execute the waterlogging risk level prediction method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0108] Embodiment III

[0109] Figure 4 is a schematic structural diagram of an electronic device provided by Embodiment III of the present invention. The electronic device is intended to represent various forms of digital computers, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described herein and / or claimed.

[0110] As Figure 4 shown, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. Among them, the memory stores a computer program executable by at least one processor. The processor 11 can execute various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0111] Multiple components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disc, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0112] The processor 11 can be various general and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the flood risk level prediction method.

[0113] In some embodiments, the flood risk level prediction method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the flood risk level prediction method described above can be executed. Alternatively, in other embodiments, the processor 11 can be configured to execute the flood risk level prediction method in any other suitable manner (e.g., by means of firmware).

[0114] The various embodiments of the systems and technologies described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special or general programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0115] A computer program for implementing the method of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, a special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The computer program may be executed entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0116] In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium may be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0117] In order to provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0118] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0119] A computing system can include a client and a server. The client and the server are generally remote from each other and typically interact through a communication network. The client-server relationship is created by computer programs running on respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system and solves the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0120] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is imposed herein.

[0121] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

[0122] An embodiment of the present invention also provides a computer program product, including a computer program and / or instructions, which when executed by a processor, implement the flood risk level prediction method provided in any embodiment of the present application.

[0123] In the process of implementing the computer program product, computer program code for performing the operations of the present invention can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network - including a local area network (LAN) or a wide area network (WAN) - or, it can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0124] Note that the above is only a preferred embodiment of the present invention and the technical principles applied. Those skilled in the art will understand that the present invention is not limited to the specific embodiments here, and various obvious changes, re-adjustments, and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments only. Without departing from the concept of the present invention, more other equivalent embodiments can be included, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for predicting the risk level of waterlogging, characterized in that, Including: Determine the current levels of each target waterlogging index, where the current level of the target waterlogging index is determined based on the obtained current value of the target waterlogging index; Input the current levels of the respective target waterlogging indices into a risk prediction model to obtain the current waterlogging risk level, where the risk prediction model is a Bayesian network constructed with the respective target waterlogging indices as nodes based on the conditional probability distribution among the respective target waterlogging indices, and the conditional probability distribution among the respective target waterlogging indices is determined based on the historical levels of the respective target waterlogging indices.

2. The method according to claim 1, wherein The determining the current levels of the respective target waterlogging indices includes: Obtain the current values of the respective target waterlogging indices; Based on the level division criteria of the respective target waterlogging indices, determine the target preset value interval in which the current values of the respective target waterlogging indices are located, and determine the level corresponding to the target preset value interval as the current level of the respective target waterlogging indices, where the level division criteria include multiple preset value intervals, and each preset value interval corresponds to a level.

3. The method according to claim 1, wherein The risk prediction model is determined by the following method: Determine the historical waterlogging data corresponding to different time points, where the historical waterlogging data includes the historical levels of the respective target waterlogging indices and the historical waterlogging risk levels; Based on the historical waterlogging data corresponding to different time points and the initial waterlogging network, determine the target waterlogging network, where the initial waterlogging network is determined based on the respective target waterlogging indices and prior knowledge, and the nodes of the target waterlogging network are the target waterlogging indices; Based on the historical waterlogging data corresponding to different time points, calculate the conditional probability distribution of each node in the target waterlogging network given the values of the parent nodes; Based on the conditional probability distribution of each node in the target waterlogging network, determine the risk prediction model.

4. The method according to claim 3, wherein The determining the target waterlogging network based on the historical waterlogging data corresponding to different time points and the initial waterlogging network includes: Based on the historical waterlogging data corresponding to different time points, use a preset point - to - point causal algorithm to determine the alternative operations corresponding to each pair of nodes in the initial waterlogging network, and the alternative operations include adding an edge, subtracting an edge, or reversing an edge; For each pair of nodes among the pairs of nodes, use a preset scoring function to calculate the scoring value of the alternative waterlogging network obtained after performing the corresponding alternative operation on the current pair of nodes; Determine the maximum scoring value among the scoring values of the alternative waterlogging networks; Return to the relevant steps of using a preset point - to - point causal algorithm based on the historical waterlogging data corresponding to different time points to determine the alternative operations of each pair of nodes in the initial waterlogging network until the maximum scoring value determined this time is less than or equal to the maximum scoring value determined last time, and determine the alternative waterlogging network corresponding to the maximum scoring value determined last time as the target waterlogging network; where, after each determination of the maximum scoring value, if the maximum scoring value determined this time is greater than the maximum scoring value determined last time, then update the initial waterlogging network to the alternative waterlogging network corresponding to the maximum scoring value determined this time.

5. The method according to claim 3, characterized in that The calculating the conditional probability distribution of each node given the values of the parent nodes based on the historical waterlogging data corresponding to different time points includes: Based on the historical waterlogging situation data corresponding to different time points, determine the conditional probability table of each node in the target waterlogging situation network, where the conditional probability table is used to represent the probability that each node takes a specific level given the values of its parent nodes; Using the Bayesian algorithm, calculate the specific values of the probabilities in the conditional probability table of each node according to the historical waterlogging situation data corresponding to different time points; Based on the specific values of the probabilities in the conditional probability table of each node, determine the conditional probability distribution of each node.

6. The method according to claim 5, characterized in that The Bayesian algorithm is expressed as: Among them, X i represents the i-th node, x i represents the level of the i-th node, Pa(X i ) represents the parent node of the i-th node, pa represents the level of the parent node, P(X i =x i |Pa(X i )=pa) represents the conditional probability that the node X i takes the value of x i when the value of the given parent node Pa(X i ) is pa, count(X i =x i ,Pa(X i )=pa) represents the number of X i =x i and Pa(X i )=pa, count(Pa(X i )=pa) represents the number of Pa(X i )=pa.

7. A device for predicting the risk level of waterlogging, characterized in that, Including: An index level determination module for determining the current level of each target waterlogging situation index, where the current level of the target waterlogging situation index is determined based on the currently obtained value of the target waterlogging situation index; A risk level determination module for inputting the current levels of the various target waterlogging situation indexes into a risk prediction model to obtain the current waterlogging situation risk level, where the risk prediction model is a Bayesian network constructed with the various target waterlogging situation indexes as nodes and based on the conditional probability distribution between the various target waterlogging situation indexes, and the conditional probability distribution between the various target waterlogging situation indexes is determined based on the historical levels of the various target waterlogging situation indexes.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; where The memory stores a computer program executable by the at least one processor, and when the computer program is executed by the at least one processor, the at least one processor is enabled to execute the waterlogging situation risk level prediction method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, they are used to implement the waterlogging situation risk level prediction method according to any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product includes a computer program, and when the computer program is executed by a processor, it is used to implement the waterlogging situation risk level prediction method according to any one of claims 1-6.

Images