Artificial intelligence-based power engineering survey design risk identification method and system

Abstract

The present application relates to the technical field of electric power engineering survey design, and proposes a power engineering survey design risk identification method and system based on artificial intelligence. The method obtains multi-round design scheme data formed in the survey design stage, and divides the design stage according to the design iteration order. Key design parameters that have an impact on engineering safety are screened, and their parameter evolution track is constructed. Based on the parameter evolution track, track feature indexes such as parameter change amplitude, parameter approximation to safety boundary degree, parameter change shock characteristics and parameter modification density characteristics are extracted. Combined with the preset normal design evolution mode, the track feature index benchmark range is constructed, the deviation of each design stage is analyzed, and whether there is an abnormal evolution mode in the parameter evolution process is identified. When an abnormal evolution mode is identified, it is determined that the corresponding survey design scheme has potential risks, and the relevant design stage and key design parameter information are output. The present application can identify potential risks from the perspective of design process evolution.

Description

Technical Field

[0001] This invention relates to the field of power engineering survey and design technology, specifically to a risk identification method and control system for power engineering survey and design based on artificial intelligence. Background Technology

[0002] In the process of power engineering survey and design, the design scheme is usually not completed in one go, but goes through multiple iterations and adjustments. For example, in the process of designing transmission line routes, selecting tower types, determining foundation types, and configuring electrical parameters, designers often modify the design parameters multiple times based on survey results, specification constraints, and review comments, and finally form a design scheme that meets the specification requirements.

[0003] Existing risk identification methods mainly focus on verifying the final design results, including determining whether design parameters meet the standard thresholds and whether stability calculations have been passed. However, in actual engineering projects, it has been found that some projects, although meeting the standard requirements in their final design results, still expose significant risks under operation or extreme conditions. The root cause is often not the final parameters themselves, but rather hidden problems in the evolution path that led to the formation of those parameters during the design process.

[0004] For example, during the design process, certain key parameters may be repeatedly pulled back after approaching the safety boundary, or be repeatedly manually corrected under conditions of multi-parameter coupling; also, parameter adjustments made to meet local constraints may introduce systemic vulnerabilities at the overall design level. These problems are difficult to detect using traditional verification methods based on the final result.

[0005] Currently, there is a lack of technical means to identify risks in the design process itself, especially a lack of methods for systematically modeling and analyzing the evolution trajectory of design parameters. Therefore, it is necessary to propose a new technical solution to identify potential engineering risks in advance by analyzing the evolution trajectory of design schemes during the survey and design phase. Summary of the Invention

[0006] In view of the technical problems mentioned in the background, the present invention provides a method and system for identifying risks in power engineering survey and design based on artificial intelligence.

[0007] This invention provides an artificial intelligence-based method for risk identification in power engineering survey and design, comprising the following steps:

[0008] S1, acquire the multi-round design scheme data formed during the survey and design stage of the power project, and divide the design scheme data into multiple design stages according to the design iteration order, wherein each design stage corresponds to a set of design parameters;

[0009] S2, Select at least one key design parameter that has an impact on the safety of the project from the design parameters of each design stage, and construct the parameter evolution trajectory of the key design parameter according to the order of the design stages;

[0010] S3. For the parameter evolution trajectory, extract trajectory feature indicators that reflect the change characteristics of key design parameters during the design iteration process, including at least parameter change amplitude characteristics, parameter approximation characteristics relative to the safety boundary, parameter change oscillation characteristics, and parameter modification density characteristics.

[0011] S4. Analyze the trajectory feature indicators based on the preset normal design evolution mode to identify whether there is an abnormal evolution mode in the parameter evolution trajectory, including at least one design parameter that is adjusted back to the compliance range after repeatedly approaching the safety boundary in multiple design stages.

[0012] S5. When the abnormal evolution pattern is identified, it is determined that the survey and design scheme represented by the corresponding design scheme data has potential risks, and the design stage and key design parameter information corresponding to the potential risks are output.

[0013] This invention also provides an artificial intelligence-based risk identification system for power engineering survey and design, the system comprising:

[0014] The design scheme data acquisition module is used to acquire multiple rounds of design scheme data formed during the survey and design phase of power engineering, and divide the design scheme data into multiple design stages according to the design iteration order, wherein each design stage corresponds to a set of design parameters.

[0015] The parameter evolution trajectory construction module is used to select at least one key design parameter that has an impact on engineering safety from the design parameters of each design stage, and construct the parameter evolution trajectory of the key design parameter according to the order of the design stages.

[0016] The trajectory feature index extraction module is used to extract trajectory feature indices that reflect the changes in key design parameters during the design iteration process for the parameter evolution trajectory. These indices include at least parameter change amplitude features, parameter approximation features relative to the safety boundary, parameter change oscillation features, and parameter modification density features.

[0017] An abnormal evolution pattern identification module is used to analyze the trajectory feature indicators based on a preset normal design evolution pattern, and identify whether there is an abnormal evolution pattern in the parameter evolution trajectory, including at least one design parameter that is adjusted back to the compliance range after repeatedly approaching the safety boundary in multiple design stages.

[0018] The risk assessment and output module is used to determine that the survey and design scheme represented by the corresponding design scheme data has potential risks when the abnormal evolution pattern is identified, and to output the design stage and key design parameter information corresponding to the potential risks.

[0019] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0020] This invention performs parameter evolution modeling and trajectory analysis on data from multiple design schemes during the survey and design process of power engineering projects. It introduces multi-dimensional trajectory characteristic indicators such as parameter change amplitude, approach to safety boundaries, oscillations, and parameter modification density. By comparing and analyzing the design process with normal design evolution patterns, it identifies potential risks where the design results appear compliant but the process exhibits abnormal evolutionary behavior. Compared to methods that rely solely on verification based on results from a single design stage, this invention can uncover hidden risk sources at the design iteration level, improving the comprehensiveness and relevance of risk identification during the survey and design stage, and providing more reliable data support for design review and risk management. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating a risk identification method for power engineering survey and design based on artificial intelligence, as disclosed in an embodiment of the present invention.

[0022] Figure 2 This is a schematic diagram of the structure of an artificial intelligence-based power engineering survey and design risk identification system disclosed in an embodiment of the present invention;

[0023] Figure 3 This is a schematic diagram of the parameter evolution trajectory construction module disclosed in an embodiment of the present invention;

[0024] Figure 4 This is a schematic diagram of the trajectory generation unit disclosed in an embodiment of the present invention;

[0025] Figure 5 This is a schematic diagram of the structure of the abnormal evolution pattern recognition module disclosed in an embodiment of the present invention;

[0026] Figure 6 This is a schematic diagram of the deviation calculation unit disclosed in an embodiment of the present invention. Detailed Implementation

[0027] The following specific embodiments illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0028] The risk identification method described in this embodiment can be deployed in business systems or technical platforms related to power engineering survey and design, such as survey and design management systems, digital design platforms for power engineering, engineering data platforms, or cloud-based design collaboration platforms deployed within power design institutes. These platforms can be used to store multi-round design outcome data generated during the survey and design phase of power engineering projects, including but not limited to scheme documents, parameter lists, verification results, and manual adjustment records generated at different design stages for line engineering, substation engineering, distribution network engineering, or new energy grid connection engineering.

[0029] For ease of understanding, the following examples use multi-round design scheme data generated during the survey and design process of power engineering as the subject of illustration. It should be noted that this invention is not limited to the specific type of power engineering, nor does it require a direct physical coupling relationship between design parameters; this invention focuses on the evolutionary behavior of design parameters and their implicit risk characteristics during the design iteration process, and its essence belongs to the intelligent analysis and inference of the evolution process of design data.

[0030] Please see Figure 1 This embodiment discloses an artificial intelligence-based method for risk identification in power engineering survey and design, including the following steps:

[0031] S1, acquire the multi-round design scheme data formed during the survey and design stage of the power project, and divide the design scheme data into multiple design stages according to the design iteration order, wherein each design stage corresponds to a set of design parameters;

[0032] In the actual survey and design process of power engineering, it typically involves multiple stages, including scheme comparison, preliminary design, construction drawing design, and multiple rounds of verification and adjustment. At each stage, designers revise the design scheme based on the results of on-site surveys, specification requirements, or review comments, thus forming multiple rounds of design scheme data.

[0033] In this embodiment, by parsing the design document version number, submission time, review node, or design process identifier, the data of multiple rounds of design schemes generated during the survey and design process of the same project can be organized in chronological or logical order, and divided into multiple design stages accordingly. For example, the first submission of the preliminary design, the modification after the preliminary design review, the first submission of the construction drawing design, and the verification and modification of the construction drawing design can be regarded as different design stages. The design parameters corresponding to each design stage may include line corridor parameters, conductor type parameters, pole or frame parameters, foundation parameters, current carrying capacity parameters, short-circuit check parameters, insulation coordination parameters, safety distance parameters, etc., which are used to characterize the technical status of the design scheme at that stage from different dimensions.

[0034] S2, Select at least one key design parameter that has an impact on the safety of the project from the design parameters of each design stage, and construct the parameter evolution trajectory of the key design parameter according to the order of the design stages;

[0035] Design parameters generated during the survey and design phase of power engineering projects are typically numerous, diverse in origin, inconsistent in scope, and vary significantly across stages. For example, the same project may generate parameter lists repeatedly during the scheme comparison, preliminary design, and construction drawing stages; the same parameter may appear at different stages with different names or units (e.g., icing thickness recorded in mm / cm, wind speed recorded as a 10-minute average or instantaneous value); simultaneously, different disciplines (electrical, structural, line, civil engineering) may have different sets of parameters of concern for the same risk. Therefore, directly performing evolutionary analysis on all parameters would not only be computationally intensive but also prone to non-critical parameters masking key risk signals. Based on this, this step first requires establishing an executable mechanism for key parameter screening and trajectory construction.

[0036] In this embodiment, the design parameters at each design stage are not limited to a single data format and may include, but are not limited to: 1) parameter tables in the design documents (such as line parameter tables, tower parameter tables, foundation parameter tables, and primary electrical parameter tables); 2) key intermediate quantities or verification result values ​​output from the verification calculation sheet (such as strength utilization coefficient, stability coefficient, sag control value, insulation coordination verification margin, etc.); 3) structured data exported from the design software (such as XML / JSON / Excel exported from CAD / BIM / electrical design tools); 4) structured parameter items extracted from the review modification records or change records (such as changing the conductor type from X to Y, or changing the foundation burial depth from a to b).

[0037] The above-mentioned parameters from multiple sources need to be normalized. For example, name normalization: map synonymous parameters to a unified parameter ID through a preset parameter dictionary (e.g., foundation depth, burial depth, and foundation depth are unified to the same parameter ID); unit normalization: unify units and perform conversions (e.g., kN↔N, mm↔m); object normalization: when a parameter has an object dimension (e.g., "per tower / per span / per loop"), it needs to be bound to an object identifier (e.g., tower number, span number, loop number) to avoid confusion; stage normalization: each design stage corresponds to a stage identifier (e.g., D1, D2, ...), and establish a corresponding relationship with the version number, timestamp, and review node.

[0038] In one optional embodiment, at least one key design parameter that affects engineering safety is selected from the design parameters of each design stage, and the parameter evolution trajectory of the key design parameter is constructed according to the order of the design stages, including:

[0039] Based on the preset safety impact criteria, design parameters related to the stability of the engineering structure, electrical safety, or operational reliability are selected from the design parameters of each design stage as key design parameters.

[0040] In practical implementation, to ensure the enforceability of the aforementioned preset security impact criteria, the security impact criteria can be constructed as a set of rules consisting of several screening rules, wherein the screening rules include at least one or more of the following:

[0041] 1) Standard Clause Mapping Rules: Map parameters explicitly used for safety verification in key standard clauses to candidate key parameters, such as parameters related to safety distance, conductor mechanical strength, tower stability, foundation bearing capacity, insulation coordination, lightning withstand level, and short-circuit thermal stability;

[0042] 2) Verification link dependency rule: Extract the input parameters used to calculate the safety margin / verification criterion from the calculation link of the verification calculation sheet or design software, and use them as key parameters; for example, if a certain parameter is included in the strength verification formula, stability verification formula or electrical clearance verification criterion, it is marked as a key parameter;

[0043] 3) Review focus rules: Analyze frequently occurring parameters from historical review comments (such as unmet sag control, insufficient safety distance, insufficient foundation depth, unreasonable tension section configuration, etc.) and mark parameters that appear frequently enough as key parameters.

[0044] 4) Sensitive rules for engineering objects: When an engineering object belongs to a specific high-risk object (such as a large span, repeated icing area, mountain strong wind area, soft foundation area, compact passage, etc.), the parameters that are strongly related to the object are promoted to key parameters; this rule can be implemented through the mapping between engineering object labels and parameter dictionaries.

[0045] The key design parameters can be organized by professional category, for example:

[0046] Structural stability related factors include: tower component cross-section, stress combination control values, stability coefficient, and strength utilization coefficient.

[0047] Civil engineering foundation related: bearing capacity of foundation, foundation type, burial depth, bottom area, parameters related to pull-out / overturning resistance calculation, etc.

[0048] Mechanical aspects of the power line: conductor type, sag, tension, span, icing thickness, wind speed, etc.

[0049] Electrical safety related parameters include: electrical clearance, safety distance, insulator string configuration, lightning parameters, and short-circuit thermal stability parameters.

[0050] For the same project, a layered selection strategy can be adopted, first selecting a small number of strongly indicative parameters (such as safety distance, sag control, and foundation bearing capacity) and then gradually expanding them, in order to control the calculation scale and highlight risk signals.

[0051] According to the chronological order of the design stages, the parameter values ​​of the key design parameters in each design stage are obtained sequentially, and the parameter evolution trajectory of the key design parameters is constructed based on the sequential relationship of the parameter values ​​changing with the design stages.

[0052] In practice, constructing the parameter evolution trajectory according to the design phase sequence is not simply a matter of piecing together a list of values. Instead, a structured trajectory object can be used to facilitate subsequent calculation of trajectory characteristic indicators. For example, a key design parameter can be... The evolutionary trajectory is represented as an ordered sequence:

[0053] Track nodes: ,in For the design phase identification, The parameter values ​​are assigned for this stage. This includes metadata related to the values ​​taken at this stage (such as unit, object ID, data source, version number, etc.).

[0054] Node sequence: .

[0055] To avoid trajectory breakage caused by missing stage values, one or more of the following methods can be adopted: If a parameter is missing in a certain stage, record it as an empty value and retain the stage node for subsequent feature extraction to handle it as missing; if a parameter is renamed in a certain stage, and alignment can still be achieved after mapping through a parameter dictionary, then write the mapped unified parameter ID into the stage value; if there are multiple version values ​​in the same stage (e.g., multiple internal iterations), then use the final committed version value of that stage as the value. and in Record the number of versions or the frequency of modifications.

[0056] Furthermore, the design process of power engineering often involves multiple revisions within a single stage. This means that the same design stage may undergo multiple rounds of internal verification, cross-disciplinary review, and feedback processing before formal submission. These adjustments are often not reflected in a single value at the end of the stage. Therefore, this embodiment introduces parameter modification density information into the parameter evolution trajectory, ensuring that the same stage not only has a value but also reflects the intensity of modifications made around that parameter within that stage.

[0057] In another optional implementation of this embodiment, the parameter evolution trajectory of the key design parameters is constructed based on the sequential relationship of the parameter values ​​changing with the design stage, including:

[0058] The number of times key design parameters were modified in each design stage was statistically analyzed to form parameter modification density information for the corresponding design stage, and then correlated with the values ​​of key design parameters for the corresponding design stage.

[0059] Following the design phase sequence, based on the values ​​of key design parameters and their corresponding parameter modification density information, a parameter evolution trajectory containing the key design parameter value sequence and the parameter modification density sequence is constructed.

[0060] In practice, the number of modifications can be obtained based on one or more of the following data sources:

[0061] 1) Version Difference Statistics: Compare parameters across multiple versions of files within the same design phase to count the number of times key parameters have changed; for example, when a parameter changes in version... If a value changes, it is recorded as a modification.

[0062] 2) Log event statistics: If the design platform provides parameter modification logs (including parameter ID, value before modification, value after modification, time, and operator), then the count can be directly based on log events;

[0063] 3) Review Comment Mapping Statistics: When a review comment item can be mapped to a parameter ID (e.g., adjusting sag, reviewing safety distance, modifying foundation depth), it can be counted according to the number of closed comment items and merged with the version difference statistics to remove duplicates;

[0064] 4) Workflow node statistics: For systems lacking fine-grained logs, the number of modifications can be estimated based on the differences between workflow nodes (commit - rollback - recommit) and parameter table snapshots.

[0065] To avoid the same modification being counted repeatedly across multiple channels, a deduplication rule can be adopted: if a version difference event with the same parameter matches a log event within the same time window, it will only be counted once; if the modification caused by the review comments has already been reflected in the version difference, it will not be counted again.

[0066] The parameter modification density information can be represented in any of the following forms: modification count, normalized density, or hierarchical density label. The modification count... This represents the count of modifications to this parameter during this phase; normalized density. This represents the density value obtained by dividing the number of modifications by the number of versions in that phase or the phase duration, for example... (in (This refers to the version number at this stage); hierarchical density tags: for example, […]. Mapping can be done in three or more levels: low, medium, and high.

[0067] Subsequently, the parameter modification density information is associated with the values ​​of key design parameters for the corresponding design stage. For example, a certain stage node can be expanded as follows: or ,in or This refers to the modified density information at this stage.

[0068] Furthermore, following the design phase sequence, and based on the values ​​of key design parameters and their corresponding parameter modification density information, a parameter evolution trajectory is constructed, comprising a sequence of key design parameter values ​​and a sequence of parameter modification densities. For example, this parameter evolution trajectory can be represented as a binary sequence pair:

[0069] Value sequence: ;

[0070] Density sequence: or .

[0071] In some embodiments, when key design parameters simultaneously have an object dimension (e.g., per tower / per span), an independent trajectory can be constructed for each object, or object identifiers can be introduced into the trajectory nodes to form a multi-object trajectory set, for example: Furthermore, the statistics on modification density should be performed at the object level to avoid mixing modification events between different objects.

[0072] S3. For the parameter evolution trajectory, extract trajectory feature indicators that reflect the change characteristics of key design parameters during the design iteration process, including at least parameter change amplitude characteristics, parameter approximation characteristics relative to the safety boundary, parameter change oscillation characteristics, and parameter modification density characteristics.

[0073] The parameter evolution trajectory itself only describes the value changes of key design parameters at each design stage. However, design risks are often not directly reflected in whether the values ​​at a certain stage are compliant, but rather in the changing behavior patterns of parameters throughout the entire design iteration process. Therefore, in this embodiment, the parameter evolution trajectory needs to be further characterized to transform the original trajectory data into trajectory feature indicators that can characterize the design evolution behavior.

[0074] In this embodiment, the extraction of trajectory feature indicators uses the design stage as the basic time granularity. That is, for each design stage in the parameter evolution trajectory, the corresponding feature indicator value is calculated by combining information from adjacent stages and within each stage. The extraction methods for various trajectory feature indicators are described in detail below.

[0075] Parameter variation amplitude characteristics; used to characterize the degree of change in the values ​​of key design parameters between adjacent design stages.

[0076] Based on the parameter values ​​of adjacent design stages in the parameter evolution trajectory, key design parameters at each design stage can be calculated. Compared to the previous design phase The magnitude of change. For example, the magnitude of change can be expressed in one or more of the following forms:

[0077] 1) Absolute range of change: .in, The parameter values ​​are for the current design phase. The parameter values ​​are from the previous design phase;

[0078] 2) Relative range of change:

[0079]

[0080] in, To prevent extremely small positive numbers with a denominator of zero;

[0081] 3) Normalized variation range: The variation range is normalized to the span of the parameter within the allowable range of historical projects or specifications to obtain a comparable variation range index across parameters.

[0082] In some embodiments, when a key design parameter has missing values ​​or has not been updated in an adjacent design stage, the corresponding change magnitude feature can be marked as a null value, or processed according to a preset rule (such as setting it to zero or using the previous value), and then distinguished in subsequent analysis.

[0083] Understandably, the parameter change amplitude feature is used to reflect the drastic degree of parameter adjustment during the design process. Its numerical value is not directly used to determine risk, but rather serves as one of the inputs for subsequent abnormal evolution pattern recognition.

[0084] Approximation characteristics of parameters relative to safety boundaries: used to describe the relationship between key design parameters and their corresponding safety constraint boundaries at each design stage.

[0085] A corresponding safety boundary value or safety range can be pre-determined for each key design parameter. The safety boundary can be derived from current power industry technical specifications, design manuals, review standards, or engineering experience rules. For example, for safety distance parameters, the minimum safety distance specified in the specifications can be used as the lower safety limit; for strength or stability verification parameters, the upper limit of the allowable utilization coefficient can be used as the upper safety limit.

[0086] In the parameter evolution trajectory, for the design phase Parameter values This allows for the calculation of how close the feature is to the safety boundary. For example, the approximation feature can take one or more of the following forms:

[0087] 1) Boundary distance value: .in, For the corresponding safety boundary value;

[0088] 2) Normalized approximation: Normalizes the position of the parameter value within the safe interval, for example, mapping it to... The proportion within the interval;

[0089] 3) Hierarchical approximation state: based on The distance to the safety boundary is used to classify the approaching state of this stage into levels such as far away, approaching, close to, and touching the boundary.

[0090] In some embodiments, when the safety boundary is in the form of an interval (such as upper and lower limits), the distance between the parameter value and the upper and lower boundaries can be calculated respectively, and the minimum distance can be selected as the approximation feature value.

[0091] Understandably, this approximation feature is used to reflect whether the parameters remain near the safety boundary for a long period during the design iteration process, rather than whether they exceed the boundary at a certain stage.

[0092] Parameter variation oscillation characteristics: used to characterize whether key design parameters are frequently adjusted in opposite directions or fluctuate back and forth between multiple design stages.

[0093] Understandably, during normal design evolution, key parameters often show a trend of monotonically converging or slowly adjusting; however, when parameters are repeatedly adjusted up and down in multiple design stages, especially oscillating near the safety boundary, it often reflects that the design scheme is repeatedly balancing between constraints and has potential instability.

[0094] In some embodiments, oscillation features can be extracted based on the direction of change of adjacent stages in the parameter evolution trajectory. For example:

[0095] 1) Construction of the change direction sequence: Encode the parameter change direction of adjacent stages, for example: if Marked as a positive change; if If there is no change, it is marked as a reverse change; if there is no change, it is marked as stable.

[0096] 2) Count of Direction Reversals: The number of times the direction reverses in the sequence of changing directions is counted as a quantitative indicator of the intensity of oscillation.

[0097] 3) Local oscillation window analysis: Within the sliding phase window (e.g., 3 or 5 consecutive design phases), the density or amplitude of direction reversal changes are statistically analyzed to characterize the local oscillation features.

[0098] Oscillation characteristics can also be used in conjunction with the state where parameters approach the safety boundary. For example, oscillation behavior can only be statistically analyzed when the parameter value is within a certain threshold range of the safety boundary.

[0099] Parameter modification density characteristics: The modification density information in the parameter evolution trajectory constructed in step S2 is used to reflect the frequency of manual adjustment of key design parameters in each design stage.

[0100] For the design phase The density value can be directly modified according to the corresponding parameter. or This serves as a characteristic of the modification density at this stage. To enhance comparability across projects and stages, the modification density characteristic can also be normalized, for example: normalized by the number of version versions at each stage; normalized by the duration of each stage; or normalized by the average modification density of similar parameters in historical projects.

[0101] When the same key design parameter exists on multiple objects (such as multiple towers or multiple spans), the modification density characteristics of each object can be calculated separately, or the statistics can be summarized at the stage granularity (e.g., taking the maximum value, average value, or weighted value).

[0102] Understandably, the parameter modification density characteristic does not directly reflect whether the parameter value is reasonable, but rather reflects the intensity of the designer's repeated adjustments around the parameter, and is an important auxiliary characteristic for identifying risks in the design process.

[0103] In this embodiment, for each key design parameter, a set of trajectory characteristic index values ​​can be generated at each design stage, for example... ,in, Indicates the characteristics of the range of change. Indicates approximation features, Indicates oscillation characteristics. This indicates a modification to the density feature.

[0104] In some embodiments, the trajectory feature indicators can be stored in a structured form, for example, using "parameter ID + design stage ID" as an index to store the corresponding feature vectors; or they can be organized into a time series feature matrix according to the parameter dimension for subsequent comparative analysis based on normal design evolution patterns.

[0105] S4. Analyze the trajectory feature indicators based on the preset normal design evolution mode to identify whether there is an abnormal evolution mode in the parameter evolution trajectory, including at least one design parameter that is adjusted back to the compliance range after repeatedly approaching the safety boundary in multiple design stages.

[0106] Understandably, in the process of power engineering survey and design, even if the design results formed at each design stage meet the requirements of current technical specifications, the evolution of design parameters during the design iteration process may still contain potential risks. For example, if a key design parameter repeatedly approaches the safety boundary and is then adjusted back to the compliance range in multiple design stages, although this kind of evolutionary behavior does not result in an out-of-bounds outcome in a single stage, from the perspective of the overall design process, it reflects that the design scheme is repeatedly tested and modified near the safety constraints, which has a high degree of uncertainty.

[0107] Therefore, in this embodiment, the judgment is not based solely on the parameter status of a single design stage, but rather a normal design evolution mode is introduced. By comparing and analyzing the overall characteristics of the parameter evolution trajectory, abnormal evolution modes are identified.

[0108] An optional embodiment analyzes the trajectory feature indicators based on a preset normal design evolution pattern to identify whether there are abnormal evolution patterns in the parameter evolution trajectory, including:

[0109] Obtain the baseline range of trajectory characteristic indicators corresponding to the normal design evolution mode, including at least the range of parameter change amplitude distribution, the range of parameter approaching the safety boundary stage distribution, the range of parameter change oscillation distribution, and the range of parameter modification density distribution.

[0110] In practice, the baseline range of the trajectory characteristic index is used to characterize the characteristic distribution range of key design parameters during the design iteration process under normal design evolution, rather than a fixed value or a fixed trajectory.

[0111] The baseline range of trajectory characteristic indicators can be obtained through statistical analysis of historically completed or approved power engineering design data. Specifically, for sample projects that match the current project type, voltage level, and environmental conditions, the parameter evolution trajectory of their key design parameters can be extracted, and the distribution of corresponding parameter change amplitude characteristics, approaching safety boundary characteristics, change oscillation characteristics, and parameter modification density characteristics at each design stage can be calculated.

[0112] To reflect the characteristic that design freedom gradually converges as the design stage progresses, the reference range of the trajectory feature index can be divided according to the design stage. For example, the distribution range of feature indices corresponding to the preliminary design stage and the construction drawing design stage can be constructed separately, so that the same index has different reference intervals at different stages.

[0113] In addition, the baseline range of trajectory characteristic indicators can be constructed in layers according to the type of project, regional environmental conditions or design conditions. For example, corresponding baseline ranges can be established for projects in icy areas, strong wind areas or high-altitude areas.

[0114] The trajectory feature index corresponding to the parameter evolution trajectory is compared with the baseline range of the trajectory feature index to calculate the deviation of the parameter evolution trajectory at each design stage;

[0115] In practice, deviation calculations are performed at the design stage as the basic granularity. Specifically, for each design stage in the parameter evolution trajectory, the trajectory characteristic index corresponding to that stage is compared with the corresponding baseline range of the trajectory characteristic index. For example, when the parameter change amplitude characteristic of a certain design stage exceeds the normal change amplitude distribution range, the deviation value of that stage in that index dimension can be recorded; when the parameter approach safety boundary characteristic of a certain design stage falls into the corresponding high approach interval within the baseline range, the approach deviation state of that stage can be recorded; when the parameter change oscillation characteristic or parameter modification density characteristic is significantly higher than the upper limit of the corresponding baseline range, it can also be recorded as the deviation performance of that stage.

[0116] It is understandable that deviations can be represented numerically, such as recording the distance between the characteristic index value and the boundary of the benchmark range; or they can be represented by grades or labels, such as marked as no deviation, slight deviation, significant deviation, etc.

[0117] Based on the deviation, identify whether there is a combination of design stages in the parameter evolution trajectory that meets the preset anomaly judgment conditions. When there is a combination of design stages that meets the anomaly judgment conditions, determine that the parameter evolution trajectory has an abnormal evolution mode.

[0118] In practice, design risks are often not caused by deviations in a single design phase, but rather by the combined behavior of multiple design phases over time. Therefore, in this embodiment, the combination of design phases is introduced as the basic analytical unit for anomaly evolution pattern recognition.

[0119] The design phase combination may include a combination of consecutive design phases, a combination of discontinuous design phases, or a combination of phases within a specified phase interval. For example, two or more consecutive design phases can be combined as a single unit to identify cumulative deviation behavior in consecutive phases; or multiple discontinuous design phases that all exhibit deviations can be combined to identify intermittent abnormal behavior.

[0120] The preset anomaly detection criteria may include one or more of the following rules: In the design stage combination, multiple design stages exhibit parameter deviations approaching the safety boundary; in the design stage combination, the parameter modification density remains above the upper limit of the baseline range in multiple stages; in the design stage combination, parameter change oscillation characteristics and safety boundary approach characteristics appear simultaneously; in the design stage combination, there is an evolutionary behavior where parameters are adjusted back to the compliant range after approaching the safety boundary. When the design stage combination meets the preset anomaly detection criteria, it can be determined that the corresponding parameter evolution trajectory exhibits an abnormal evolution pattern.

[0121] In one optional embodiment, the trajectory feature index corresponding to the parameter evolution trajectory is compared with the baseline range of the trajectory feature index to calculate the deviation of the parameter evolution trajectory at each design stage, including:

[0122] For the trajectory feature indicators within the same design phase, the deviation value of each trajectory feature indicator relative to the corresponding trajectory feature indicator reference range is determined, and the deviation value is marked with the corresponding deviation direction;

[0123] In practice, when comparing the trajectory feature index corresponding to the parameter evolution trajectory with the baseline range of the trajectory feature index, and calculating the deviation of the parameter evolution trajectory at each design stage, the multiple trajectory feature indices are not directly merged, but are calculated separately according to the deviation direction of the trajectory feature indices within the same design stage.

[0124] Specifically, in any design stage, different trajectory characteristic indicators reflect different focuses of the design state. For example, the parameter change amplitude characteristic reflects the intensity of adjustments between stages, the parameter approaching the safety boundary characteristic reflects the relationship between the parameter value and the safety constraint, and the parameter modification density characteristic reflects the frequency of adjustment behaviors around the parameter within a stage. Since the above trajectory characteristic indicators are not consistent in numerical meaning and deviation semantics, when calculating the deviation of the design stage, if the deviation values ​​of different indicators are directly merged, it is impossible to distinguish the direction of design state change indicated by the deviation of each indicator.

[0125] Based on the semantic inconsistency of multiple indicators deviating within the aforementioned design phase, this embodiment first determines the deviation value of each trajectory feature indicator relative to its corresponding baseline range for each trajectory feature indicator within the same design phase. Simultaneously, the deviation value is labeled with its corresponding deviation direction. This deviation direction is used to distinguish the deviation trend of the trajectory feature indicator value relative to its baseline range, enabling the deviation behavior of different trajectory feature indicators within the same design phase to be differentiated along the directional dimension.

[0126] Based on the deviation direction, the deviation values ​​within the same design stage are grouped to form at least one set of deviation values ​​with the same deviation direction.

[0127] In practice, after determining the deviation value and direction, the deviation values ​​within the same design stage are grouped according to the deviation direction. This grouping method ensures that deviation values ​​with consistent deviation directions are grouped into the same set, thus avoiding the simultaneous inclusion of trajectory feature indicators with opposite deviation trends in subsequent calculations. Therefore, even if multiple trajectory feature indicators deviate within the same design stage, the structural independence of various deviation behaviors can be maintained.

[0128] The deviation values ​​of each deviation direction are combined and calculated to obtain the deviation values ​​of the corresponding design stage. Based on the deviation values ​​of the deviations, the deviation of the parameter evolution trajectory in the corresponding design stage is determined.

[0129] In practice, the deviation value sets consistent in each deviation direction are combined and calculated separately. This combined calculation is performed within each set of deviation values ​​consistent in each deviation direction, ensuring that the resulting directional deviation values ​​only reflect the overall degree of deviation in the corresponding deviation direction for that design stage, without being affected by indicators from other deviation directions. In this way, the directional deviation values ​​can form a stage-level deviation expression that can be used for subsequent analysis while maintaining the integrity of multi-indicator information.

[0130] Finally, based on the directional deviation values, the deviation of the parameter evolution trajectory at the corresponding design stage is determined. This deviation is no longer a single numerical value, but rather a stage-specific deviation state characterized by the directional deviation values, thus providing structured input for subsequent abnormal evolution pattern recognition based on design stage combinations.

[0131] For example: Suppose at a certain design stage In this study, for a specific key design parameter, the following deviation values ​​and directions of trajectory feature indicators were extracted and calculated: the deviation value of the parameter change amplitude feature was +0.35, and the deviation direction was positive; the deviation value of the parameter change oscillation feature was +0.2, and the deviation direction was positive; the deviation value of the parameter approaching the safety boundary feature was -0.15, and the deviation direction was negative; the deviation value of the parameter modification density feature was +0.4, and the deviation direction was positive.

[0132] It can be seen that within the same design phase, the deviation directions of different trajectory characteristic indicators are not consistent. Among them, the parameter change amplitude characteristic, parameter change oscillation characteristic, and parameter modification density characteristic all show positive deviation, while the parameter approaching the safety boundary characteristic shows negative deviation.

[0133] Based on the differences in the direction of deviation mentioned above, in this embodiment, the deviation values ​​are first grouped according to the direction of deviation. For example, deviation values ​​with a positive deviation direction [0.35, 0.2, 0.4] are assigned to the first set of deviation values, and deviation values ​​with a negative deviation direction [-0.15] are assigned to the second set of deviation values.

[0134] The sets of deviation values ​​in each consistent deviation direction are combined and calculated separately. For example, the first deviation value is summed and combined to obtain the corresponding positive deviation values ​​for the design stage: Simultaneously, the second set of deviation values ​​is calculated by combining the absolute values ​​or the average values ​​to obtain the corresponding negative deviation values ​​for the design stage: The obtained deviation values ​​indicate that there is a relatively concentrated deviation behavior in the positive deviation direction during this design phase, while there is only a single, small-amplitude deviation behavior in the negative deviation direction.

[0135] Based on the aforementioned directional deviation values, the deviation in this design stage can be determined as follows: the parameter evolution trajectory corresponding to this design stage exhibits significant deviation characteristics in the positive deviation direction and slight deviation characteristics in the negative deviation direction. These deviations are recorded as directional deviation values ​​and used as input information for subsequent identification of anomalous evolution patterns in the design stage combinations.

[0136] It is understandable that, through the above example, even if the deviation directions of different trajectory feature indicators differ within the same design phase, the deviation situation of that design phase can be structurally described by combining the deviation values, without the need to simplify or cancel out the information of different deviation directions during the deviation calculation phase.

[0137] S5. When the abnormal evolution pattern is identified, it is determined that the survey and design scheme represented by the corresponding design scheme data has potential risks, and the design stage and key design parameter information corresponding to the potential risks are output.

[0138] Specifically, when the evolution trajectory of a critical design parameter is determined to exhibit an abnormal evolution pattern, it is considered that the critical design parameter has exhibited abnormal evolutionary behavior during the design iteration process. Since critical design parameters are inherently related to engineering safety, this abnormal evolutionary behavior can be mapped to a potential risk associated with that critical design parameter in the corresponding design scheme. One abnormal evolution pattern can correspond to one potential risk; alternatively, multiple abnormal evolution patterns can be merged and mapped to the same potential risk category, such as the abnormal evolution of the same parameter on different objects, or the abnormal evolution of multiple parameters combined in the same design phase.

[0139] In this embodiment, the determination of potential risks is based on the existence of abnormal evolution patterns. That is, when no abnormal evolution pattern is identified, potential risk determination is not triggered; when at least one abnormal evolution pattern is identified, the survey and design scheme represented by the corresponding design scheme data is determined to have potential risks.

[0140] It should be noted that in this embodiment, the determination of potential risks does not depend on whether the parameters cross the safety boundary, nor does it require that the parameters have non-compliant values ​​at any design stage. Instead, the determination is based on the evolutionary behavior during the design process.

[0141] The granularity for determining potential risks can be one or more of the following:

[0142] Parameter-level risk: Identifying potential risks based on the abnormal evolution patterns of a specific key design parameter;

[0143] Object-level risk: Determine potential risks based on the abnormal evolution of key design parameters for a specific engineering object (such as a certain tower, a certain span, or a certain equipment unit);

[0144] Solution-level risk: When multiple key design parameters or multiple objects exhibit abnormal evolution patterns in the same design solution, the design solution as a whole is deemed to have potential risks.

[0145] When a design scheme is determined to have potential risks, it is necessary to output the design stage and key design parameters corresponding to the potential risks, so that subsequent personnel or systems can locate and analyze the source of the risks. Specifically, the output information includes at least one or more of the following:

[0146] 1) Key design parameter identification information: Key design parameters used to identify abnormal evolution modes, such as parameter name, parameter ID, professional category or corresponding engineering object identifier;

[0147] 2) Design phase information: Used to identify the design phase or combination of design phases involved in the abnormal evolution pattern, such as the preliminary design phase, the construction drawing design phase, or the specific phase number and time range;

[0148] 3) Abnormal evolution pattern association information: used to describe the type of abnormal evolution pattern corresponding to the potential risk, such as callback after multi-stage approach to the safety boundary, high modification density accompanied by approach behavior, etc.

[0149] 4) Deviation Summary Information: Used to describe the deviations in the design phases involved in the abnormal evolution pattern, such as the deviation value, the distribution of deviation directions, or the deviation level identifier.

[0150] In some embodiments, the above information may be organized into structured data records, such as stored in the form of potential risk records, each record corresponding to a potential risk instance and containing parameters, stages, and deviation information associated with the risk.

[0151] The output format of the potential risks is not limited to a specific representation and can be flexibly configured according to the system deployment and application scenario. It is understood that the potential risk results can be output to the risk list interface of the design management system to display potential risk items in the current design scheme to designers or reviewers; the potential risk results can also be output as a data interface to external systems, such as engineering review systems, quality management systems, or risk assessment systems, for use in subsequent processes. Furthermore, the output of potential risk results can be linked to the design stage status, for example, prompting or marking design parameters with potential risks in the previous stage before proceeding to the next design stage.

[0152] Please see Figure 2 This invention also provides an artificial intelligence-based power engineering survey and design risk identification system 100, the system comprising:

[0153] The design scheme data acquisition module 10 is used to acquire multiple rounds of design scheme data formed during the survey and design stage of the power project, and divide the design scheme data into multiple design stages according to the design iteration order, wherein each design stage corresponds to a set of design parameters.

[0154] The parameter evolution trajectory construction module 20 is used to select at least one key design parameter that has an impact on engineering safety from the design parameters of each design stage, and construct the parameter evolution trajectory of the key design parameter according to the order of the design stages.

[0155] The trajectory feature index extraction module 30 is used to extract trajectory feature indices that reflect the change characteristics of key design parameters during the design iteration process for the parameter evolution trajectory, including at least parameter change amplitude characteristics, parameter approximation characteristics relative to the safety boundary, parameter change oscillation characteristics, and parameter modification density characteristics.

[0156] The abnormal evolution pattern identification module 40 is used to analyze the trajectory feature index based on the preset normal design evolution pattern, and identify whether there is an abnormal evolution pattern in the parameter evolution trajectory, including at least one design parameter that is adjusted back to the compliance range after repeatedly approaching the safety boundary in multiple design stages.

[0157] The risk assessment and output module 50 is used to determine that the survey and design scheme represented by the corresponding design scheme data has potential risks when the abnormal evolution mode is identified, and to output the design stage and key design parameter information corresponding to the potential risks.

[0158] For an alternative embodiment, please refer to Figure 3 The parameter evolution trajectory construction module 20 includes:

[0159] The key parameter screening unit 21 is used to screen design parameters related to the stability of the engineering structure, electrical safety or operational reliability as key design parameters from the design parameters of each design stage according to the preset safety impact criteria.

[0160] The trajectory generation unit 22 is used to sequentially obtain the parameter values ​​of the key design parameters in each design stage according to the order of the design stages, and construct the parameter evolution trajectory of the key design parameters according to the order of the parameter values ​​changing with the design stages.

[0161] For an alternative embodiment, please refer to Figure 4 The trajectory generation unit 22 includes:

[0162] The parameter modification density statistics unit 221 is used to count the number of modifications of key design parameters in each design stage, form parameter modification density information for the corresponding design stage, and associate it with the values ​​of key design parameters for the corresponding design stage.

[0163] The multidimensional trajectory construction unit 222 is used to construct a parameter evolution trajectory containing a sequence of key design parameter values ​​and a sequence of parameter modification densities, based on the values ​​of key design parameters and their corresponding parameter modification density information, according to the design stage sequence.

[0164] For an alternative embodiment, please refer to Figure 5 The abnormal evolution pattern recognition module 40 includes:

[0165] The baseline range acquisition unit 41 is used to acquire the baseline range of trajectory characteristic indicators corresponding to the normal design evolution mode, including at least the range of parameter change amplitude distribution, the range of parameter approaching the safety boundary stage distribution, the range of parameter change oscillation distribution, and the range of parameter modification density distribution.

[0166] The deviation calculation unit 42 is used to compare the trajectory feature index corresponding to the parameter evolution trajectory with the baseline range of the trajectory feature index, and calculate the deviation of the parameter evolution trajectory at each design stage.

[0167] The anomaly determination unit 43 is used to identify whether there is a combination of design stages that meets the preset anomaly determination conditions in the parameter evolution trajectory based on the deviation, and to determine that there is an abnormal evolution mode in the parameter evolution trajectory when there is a combination of design stages that meets the anomaly determination conditions.

[0168] For an alternative embodiment, please refer to Figure 6 The deviation calculation unit 42 includes:

[0169] The deviation value determination subunit 421 is used to determine the deviation value of each trajectory feature index relative to the corresponding trajectory feature index reference range for the trajectory feature index within the same design stage, and to mark the corresponding deviation direction for the deviation value.

[0170] Deviation grouping subunit 422 is used to group deviation values ​​within the same design stage according to the deviation direction, forming at least one set of deviation values ​​with the same deviation direction.

[0171] The directional deviation calculation subunit 423 is used to combine and calculate the deviation value sets with the same deviation direction respectively to obtain the directional deviation value of the corresponding design stage, and determine the deviation of the parameter evolution trajectory in the corresponding design stage based on the directional deviation value.

[0172] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for risk identification in power engineering survey and design based on artificial intelligence, characterized in that, The method includes the following steps: S1, acquire the multi-round design scheme data formed during the survey and design stage of the power project, and divide the design scheme data into multiple design stages according to the design iteration order, wherein each design stage corresponds to a set of design parameters; S2, Select at least one key design parameter that has an impact on the safety of the project from the design parameters of each design stage, and construct the parameter evolution trajectory of the key design parameter according to the order of the design stages; S3. For the parameter evolution trajectory, extract trajectory feature indicators that reflect the change characteristics of key design parameters during the design iteration process, including at least parameter change amplitude characteristics, parameter approximation characteristics relative to the safety boundary, parameter change oscillation characteristics, and parameter modification density characteristics. S4. Analyze the trajectory feature indicators based on the preset normal design evolution mode to identify whether there is an abnormal evolution mode in the parameter evolution trajectory, including at least one design parameter that is adjusted back to the compliance range after repeatedly approaching the safety boundary in multiple design stages. S5, when the abnormal evolution pattern is identified, it is determined that the survey and design scheme represented by the corresponding design scheme data has potential risks, and the design stage and key design parameter information corresponding to the potential risks are output. From the design parameters of each design stage, select at least one key design parameter that has an impact on engineering safety, and construct the parameter evolution trajectory of the key design parameter according to the order of the design stages, including: Based on the preset safety impact criteria, design parameters related to the stability of the engineering structure, electrical safety, or operational reliability are selected from the design parameters of each design stage as key design parameters. According to the chronological order of the design stages, the parameter values ​​of the key design parameters in each design stage are obtained sequentially, and the parameter evolution trajectory of the key design parameters is constructed based on the sequential relationship of the parameter values ​​changing with the design stages. The parameter evolution trajectory of the key design parameters is constructed based on the sequential relationship of the parameter values ​​changing with the design stage, including: The number of times key design parameters were modified in each design stage was statistically analyzed to form parameter modification density information for the corresponding design stage, and then correlated with the values ​​of key design parameters for the corresponding design stage. Following the design phase sequence, based on the values ​​of key design parameters and their corresponding parameter modification density information, a parameter evolution trajectory containing the key design parameter value sequence and the parameter modification density sequence is constructed.

2. The method for identifying risks in power engineering survey and design based on artificial intelligence according to claim 1, characterized in that: The trajectory feature indicators are analyzed based on a preset normal design evolution pattern to identify whether there are abnormal evolution patterns in the parameter evolution trajectory, including: Obtain the baseline range of trajectory characteristic indicators corresponding to the normal design evolution mode, including at least the range of parameter change amplitude distribution, the range of parameter approaching the safety boundary stage distribution, the range of parameter change oscillation distribution, and the range of parameter modification density distribution. The trajectory feature index corresponding to the parameter evolution trajectory is compared with the baseline range of the trajectory feature index to calculate the deviation of the parameter evolution trajectory at each design stage; Based on the deviation, identify whether there is a combination of design stages in the parameter evolution trajectory that meets the preset anomaly judgment conditions. When there is a combination of design stages that meets the anomaly judgment conditions, determine that the parameter evolution trajectory has an abnormal evolution mode.

3. The method for identifying risks in power engineering survey and design based on artificial intelligence according to claim 2, characterized in that: The trajectory feature index corresponding to the parameter evolution trajectory is compared with the baseline range of the trajectory feature index to calculate the deviation of the parameter evolution trajectory at each design stage, including: For the trajectory feature indicators within the same design phase, the deviation value of each trajectory feature indicator relative to the corresponding trajectory feature indicator reference range is determined, and the deviation value is marked with the corresponding deviation direction; Based on the deviation direction, the deviation values ​​within the same design stage are grouped to form at least one set of deviation values ​​with the same deviation direction. The deviation values ​​of each deviation direction are combined and calculated to obtain the deviation values ​​of the corresponding design stage. Based on the deviation values ​​of the deviations, the deviation of the parameter evolution trajectory in the corresponding design stage is determined.

4. A power engineering survey and design risk identification system based on artificial intelligence, characterized in that, The system includes: The design scheme data acquisition module is used to acquire multiple rounds of design scheme data formed during the survey and design phase of power engineering, and divide the design scheme data into multiple design stages according to the design iteration order, wherein each design stage corresponds to a set of design parameters. The parameter evolution trajectory construction module is used to select at least one key design parameter that has an impact on engineering safety from the design parameters of each design stage, and construct the parameter evolution trajectory of the key design parameter according to the order of the design stages. The trajectory feature index extraction module is used to extract trajectory feature indices that reflect the changes in key design parameters during the design iteration process for the parameter evolution trajectory. These indices include at least parameter change amplitude features, parameter approximation features relative to the safety boundary, parameter change oscillation features, and parameter modification density features. An abnormal evolution pattern identification module is used to analyze the trajectory feature indicators based on a preset normal design evolution pattern, and identify whether there is an abnormal evolution pattern in the parameter evolution trajectory, including at least one design parameter that is adjusted back to the compliance range after repeatedly approaching the safety boundary in multiple design stages. The risk assessment and output module is used to determine that the survey and design scheme represented by the corresponding design scheme data has potential risks when the abnormal evolution pattern is identified, and to output the design stage and key design parameter information corresponding to the potential risks. The parameter evolution trajectory construction module includes: The key parameter screening unit is used to select design parameters related to the stability of the engineering structure, electrical safety, or operational reliability as key design parameters from the design parameters of each design stage according to the preset safety impact criteria. The trajectory generation unit is used to sequentially obtain the parameter values ​​of the key design parameters in each design stage according to the order of the design stages, and construct the parameter evolution trajectory of the key design parameters according to the order of the parameter values ​​changing with the design stages. The trajectory generation unit includes: The parameter modification density statistics unit is used to count the number of times key design parameters are modified in each design stage, form parameter modification density information for the corresponding design stage, and associate it with the values ​​of key design parameters for the corresponding design stage. The multidimensional trajectory construction unit is used to construct a parameter evolution trajectory containing a sequence of key design parameter values ​​and a sequence of parameter modification densities, based on the values ​​of key design parameters and their corresponding parameter modification density information, in accordance with the design phase sequence.

5. The artificial intelligence-based power engineering survey and design risk identification system according to claim 4, characterized in that: The abnormal evolution pattern recognition module includes: The baseline range acquisition unit is used to acquire the baseline range of trajectory characteristic indicators corresponding to the normal design evolution mode, including at least the distribution range of parameter change amplitude, the distribution range of parameter approaching the safety boundary in stages, the distribution range of parameter change oscillation, and the distribution range of parameter modification density. The deviation calculation unit is used to compare the trajectory feature index corresponding to the parameter evolution trajectory with the baseline range of the trajectory feature index, and calculate the deviation of the parameter evolution trajectory at each design stage. An anomaly determination unit is used to identify whether there is a combination of design stages in the parameter evolution trajectory that meets the preset anomaly determination conditions based on the deviation, and to determine that there is an abnormal evolution mode in the parameter evolution trajectory when there is a combination of design stages that meets the anomaly determination conditions.

6. The power engineering survey and design risk identification system based on artificial intelligence according to claim 5, characterized in that: The deviation calculation unit includes: The deviation value determination subunit is used to determine the deviation value of each trajectory feature index relative to the corresponding trajectory feature index reference range for the same design stage, and to mark the deviation value with the corresponding deviation direction. The deviation grouping subunit is used to group deviation values ​​within the same design stage according to the deviation direction, forming at least one set of deviation values ​​with the same deviation direction. The directional deviation calculation subunit is used to combine and calculate the deviation value sets with the same deviation direction for each deviation, to obtain the directional deviation value for the corresponding design stage, and to determine the deviation of the parameter evolution trajectory at the corresponding design stage based on the directional deviation value.

Images