Multi-source fusion autonomous controllable evaluation method and device based on exogenous risk perception

Abstract

The invention relates to a multi-source fusion autonomous controllable evaluation method and device based on exogenous risk perception. The method comprises the following steps: acquiring external source data related to equipment to be evaluated, historical external source evaluation data of the equipment to be evaluated, and an internal source self-control evaluation result of an internal source self-control index corresponding to the equipment to be evaluated; obtaining exogenous risk assessment results of a plurality of preset exogenous risk indexes according to the exogenous data; according to the historical exogenous assessment data and the exogenous risk assessment result, adjusting a multi-source index weight, and obtaining a target assessment weight; the target evaluation weight comprises the weight of the internal source self-control index and the weight of the external source risk index; and according to the internal source self-control assessment result, the external source risk assessment result and the target assessment weight, determining a fusion assessment result of the to-be-assessed device. By adopting the method, the autonomous controllability of the power equipment can be comprehensively evaluated.

Description

Technical Field

[0001] The present application relates to the field of power grid technology, and in particular to a multi-source fusion autonomous and controllable evaluation method and device based on exogenous risk perception. Background Art

[0002] With the development of the power industry and the emphasis on the independent controllability of key technologies, evaluation methods for the independent controllability of power equipment have gradually emerged.

[0003] However, the autonomous controllability evaluation methods in related technologies usually focus on the quantitative analysis of the equipment's own technology, which makes it difficult to provide timely warnings of potential risks and urgently needs to be improved and enhanced. Summary of the Invention

[0004] Based on this, it is necessary to provide a multi-source fusion autonomous controllability evaluation method, device, computer equipment, computer-readable storage medium and computer program product based on exogenous risk perception that can comprehensively evaluate the autonomous controllability of power equipment in response to the above technical problems.

[0005] In a first aspect, the present application provides a multi-source fusion autonomous and controllable evaluation method based on exogenous risk perception, the method comprising:

[0006] Acquire exogenous data related to the device to be evaluated, historical exogenous evaluation data of the device to be evaluated, and endogenous self-control evaluation results of endogenous self-control indicators corresponding to the device to be evaluated;

[0007] Obtaining exogenous risk assessment results of a plurality of preset exogenous risk indicators based on the exogenous data;

[0008] Adjust the weights of the multi-source indicators based on the historical exogenous assessment data and the exogenous risk assessment results, and obtain a target assessment weight; the target assessment weight includes the weight of the endogenous self-controlled indicator and the weight of the exogenous risk indicator;

[0009] Determine a fusion evaluation result of the device to be evaluated based on the endogenous self-controlled evaluation result, the exogenous risk evaluation result, and the target evaluation weight.

[0010] In one embodiment, obtaining external data related to the device to be evaluated includes:

[0011] Obtaining a knowledge graph of the device to be evaluated; wherein the knowledge graph includes a plurality of entity nodes and association relationships between the entity nodes, the plurality of entity nodes including at least some of the device to be evaluated, components of the device to be evaluated, the manufacturer of the device to be evaluated, suppliers of the components, and supervisors of the manufacturer and the supplier;

[0012] Determine, based on the knowledge graph, the manufacturer of the device to be evaluated, the suppliers of each component in the device to be evaluated, and the supervisors corresponding to the manufacturer and the suppliers;

[0013] Obtain risk status data of the device to be evaluated, the manufacturer, the supplier, and the regulator within a preset time period;

[0014] The risk status data of the device to be evaluated, the manufacturer, the supplier, and the regulator are integrated to obtain external source data of the device to be evaluated.

[0015] In one embodiment, the plurality of exogenous risk indicators include at least part of a credibility indicator, a reliability indicator, and a controllability indicator, and the exogenous risk assessment result includes at least part of a credibility index, a reliability index, and a controllability index;

[0016] The step of obtaining, based on the exogenous data, exogenous risk assessment results of a plurality of preset exogenous risk indicators includes:

[0017] Classifying the external data according to the trained language processing model to obtain credibility data, reliability data, and controllability data of the device to be evaluated;

[0018] Determining a data source of the credibility data, and determining a credibility index of the device to be evaluated based on a preset sentiment analysis model, the credibility data, and the data source;

[0019] Determine a reliability index of the device to be evaluated based on the trained graph neural network model, the reliability data, and the knowledge graph of the device to be evaluated;

[0020] The controllability index of the device to be evaluated is determined according to a preset rule engine and the controllability data.

[0021] In one embodiment, the credibility data includes a plurality of trustworthy texts; the data sources include at least public sources and media sources;

[0022] Determining the credibility index of the device to be evaluated based on a preset sentiment analysis model, the credibility data, and the data source includes:

[0023] Obtaining the sentiment score and text topic of each of the trustworthy texts according to the sentiment analysis model;

[0024] The credibility index of the device to be evaluated is calculated based on the preset first source weight of the public source, the preset second source weight of the media source, the preset topic weight corresponding to each text topic, and the sentiment score, text topic and data source of each trust text.

[0025] In one embodiment, the historical exogenous evaluation data includes the historical evaluation results of each of the exogenous risk indicators of the device to be evaluated, the external historical weight of each of the exogenous risk indicators, and the historical internal weight of each of the endogenous self-controlled indicators;

[0026] The step of adjusting the weights of multiple indicators based on the historical exogenous assessment data and the exogenous risk assessment results and obtaining target assessment weights includes:

[0027] Obtaining a target risk indicator from a plurality of exogenous risk indicators; wherein the target risk indicator is an exogenous risk indicator for which the difference between a historical assessment result and a corresponding exogenous risk assessment result is greater than a preset threshold;

[0028] In the absence of a target risk indicator, the external historical weight of each exogenous risk indicator is used as the weight of the corresponding exogenous risk indicator, and the historical internal weight of each endogenous self-controlled indicator is used as the weight of the corresponding endogenous self-controlled indicator;

[0029] In the case where some exogenous risk indicators are target risk indicators, the historical external weights of the target risk indicators are increased, and the historical external weights of the external risk indicators other than the target risk indicators among the multiple exogenous risk indicators are reduced, and the adjusted historical external weights are used as the weights of the corresponding exogenous risk indicators, and the historical internal weights of each endogenous self-controlled indicator are used as the weights of the corresponding endogenous self-controlled indicator;

[0030] When all the exogenous risk indicators are target risk indicators, the historical external weights of the exogenous risk indicators are increased, and the historical internal weights of the endogenous self-controlled indicators are reduced, and the adjusted historical external weights are used as the weights of the corresponding exogenous risk indicators, and the adjusted historical internal weights are used as the weights of the corresponding endogenous self-controlled indicators.

[0031] In one embodiment, determining the fusion evaluation result of the device to be evaluated based on the endogenous self-controlled evaluation result, the exogenous risk evaluation result, and the target evaluation weight includes:

[0032] Determine the autonomous controllability evaluation score of the device to be evaluated based on the endogenous autonomous control evaluation result, the exogenous risk evaluation result, and the target evaluation weight;

[0033] When the autonomous controllable evaluation score is greater than or equal to a preset first score threshold, determining that the fusion evaluation result of the device to be evaluated is autonomous controllable;

[0034] When the autonomous controllable evaluation score is less than the first score threshold and greater than or equal to a preset second score threshold, determining that the fusion evaluation result of the device to be evaluated is basically controllable;

[0035] When the autonomous controllable evaluation score is less than the second score threshold, it is determined that the fusion evaluation result of the device to be evaluated is restricted.

[0036] In a second aspect, the present application provides a multi-source fusion autonomous and controllable evaluation device based on exogenous risk perception, the device comprising:

[0037] An information acquisition module is used to acquire exogenous data related to the device to be evaluated, historical exogenous evaluation data of the device to be evaluated, and endogenous self-control evaluation results of endogenous self-control indicators corresponding to the device to be evaluated;

[0038] A risk assessment module, configured to obtain exogenous risk assessment results of a plurality of preset exogenous risk indicators based on the exogenous data;

[0039] A weight determination module is used to adjust the weights of multiple source indicators based on the historical exogenous evaluation data and the exogenous risk assessment results, and obtain a target evaluation weight; the target evaluation weight includes the weight of the endogenous self-controlled indicator and the weight of the exogenous risk indicator;

[0040] A result determination module is used to determine the fusion evaluation result of the device to be evaluated according to the endogenous self-controlled evaluation result, the exogenous risk evaluation result and the target evaluation weight.

[0041] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described in any one of the above embodiments when executing the computer program.

[0042] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method described in any one of the above embodiments when the computer program is executed by a processor.

[0043] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which implements the steps of the method described in any one of the above embodiments when executed by a processor.

[0044] The above-mentioned multi-source fusion autonomous and controllable evaluation method, device, computer equipment, computer-readable storage medium and computer program product based on exogenous risk perception include obtaining exogenous data related to the equipment to be evaluated, historical exogenous evaluation data of the equipment to be evaluated, and endogenous self-controlled evaluation results of the endogenous self-controlled indicators corresponding to the equipment to be evaluated. Then, based on the exogenous data, the exogenous risk evaluation results of multiple exogenous risk indicators are obtained. Then, based on the historical exogenous evaluation data and exogenous risk evaluation results of the equipment to be evaluated, the change between the current exogenous risk evaluation results and the historical exogenous risk evaluation results is determined. It is judged in turn whether the weights of the endogenous self-controlled indicators and the weights of the exogenous risk indicators need to be adjusted. Finally, based on the endogenous self-controlled evaluation results, the exogenous risk evaluation results and the target evaluation weights, the fusion evaluation results of the equipment to be evaluated are determined. The multi-source fusion autonomous controllability evaluation method based on exogenous risk perception in this application, on the basis of the traditional reliance on quantitative analysis of the equipment's own technology, that is, analyzing the autonomous controllability of power equipment only based on the endogenous autonomous control evaluation results of the equipment to be evaluated, incorporates the analysis of exogenous data related to the equipment to be evaluated, so that the fusion evaluation results of the equipment to be evaluated can reflect the impact of the external environment on the autonomous controllability of the power equipment, and can timely warn of potential risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments of the present application or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying any creative work.

[0046] Figure 1 1. A flowchart of a multi-source fusion autonomous and controllable evaluation method based on exogenous risk perception in one embodiment;

[0047] Figure 2 Schematic diagram of the process of step S101 in one embodiment;

[0048] Figure 3 Schematic diagram of the process of step S102 in one embodiment;

[0049] Figure 4 Schematic diagram of the process of step S302 in one embodiment;

[0050] Figure 5 Schematic diagram of the process of step S103 in one embodiment;

[0051] Figure 6 Schematic diagram of the process of step S104 in one embodiment;

[0052] Figure 7 This is a structural block diagram of a multi-source fusion autonomous and controllable evaluation device based on exogenous risk perception in one embodiment;

[0053] Figure 8 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0055] As mentioned in the background section, with the development of the power industry and the emphasis on independent controllability of key technologies, evaluation methods for the independent controllability of power equipment have gradually emerged. Current independent controllability evaluations typically focus on quantitative analysis of the equipment's own technology and supply chain factors. For example, existing technical solutions often assess the degree of independent controllability of power equipment through a pre-defined indicator system, including considerations such as the localization rate of core components, whether key technologies are mastered by domestic manufacturers, the degree of supply chain dependence, and the security and trustworthiness of software and hardware. These methods, generally based on expert experience and historical data, score or weight the equipment's technical indicators (such as performance parameters and consistency standards) and supply chain structure to produce a comprehensive score or grade for the equipment's independent controllability. Furthermore, some evaluation systems also focus on whether the equipment's operation and maintenance are fully domestically performed, whether it has the ability to replace imported components, and the equipment's performance indicators in terms of information security and manageability. These existing technologies have played a role in evaluating the independent controllability of power equipment, providing decision-making support for improving the security and independent controllability of national power infrastructure.

[0056] While the aforementioned autonomous controllability evaluation methods provide a means of analyzing internal device and supply chain factors, they still have significant shortcomings. First, they primarily focus on static indicators and an internal perspective, failing to fully consider the dynamic nature of the external environment. For example, within the macro-environment in which a device operates, public opinion and external risk factors (such as policy changes, industry regulation, new security threats, and supply chain disruptions) can significantly impact the device's actual controllability. However, existing technologies lack mechanisms for capturing and analyzing this external, unstructured information. Second, traditional methods often rely on expert scoring or fixed models, resulting in a slow response to emerging risk signals and insufficient real-time awareness. When negative comments about a device or manufacturer appear on social media or international events lead to increased supply chain risks, existing evaluation systems struggle to update results promptly, potentially underestimating potential risks. Furthermore, most existing technologies use preset weights to linearly aggregate indicators and lack a dynamic weight adjustment mechanism, preventing the evaluation model from adaptively adjusting to changes in risk levels. This limits the accuracy and timeliness of the evaluation results. Finally, the evaluation results are often presented in a single format, typically with a single score or grade, failing to provide multi-dimensional information such as risk warnings and cause analysis tailored to different audiences. These shortcomings indicate that existing autonomous controllability evaluation technologies cannot fully reflect the impact of the external environment on the autonomous controllability of power equipment, and are unable to provide timely warnings of potential risks, and are in urgent need of improvement and enhancement.

[0057] In an exemplary embodiment, see Figure 1 The present application provides a multi-source fusion autonomous and controllable evaluation method based on exogenous risk perception, which is described by taking the application of the method to a control terminal as an example. The method includes steps S101 to S104.

[0058] S101: Acquire exogenous data related to the device to be evaluated, historical exogenous evaluation data of the device to be evaluated, and an endogenous self-control evaluation result of an endogenous self-control indicator corresponding to the device to be evaluated.

[0059] In this embodiment, the control terminal can pre-establish connections with public social media platforms, news media platforms, official platforms of companies related to the power equipment to be evaluated, and official platforms of regulators related to the equipment to be evaluated, so that the control terminal can obtain external data related to the equipment to be evaluated from these platforms. In one example, external data may include comments or articles about the equipment to be evaluated posted on public social media platforms, media reports and evaluations of the equipment to be evaluated and its supply chain, information on the status and market dynamics of upstream and downstream companies in the equipment's supply chain, and information on regulatory environment changes released by relevant regulators.

[0060] S102: Obtaining exogenous risk assessment results of a plurality of preset exogenous risk indicators based on exogenous data.

[0061] In one example, multiple exogenous risk indicators include at least a credibility indicator, a reliability indicator, and a controllability indicator, and the exogenous risk assessment results include at least a credibility index, a reliability index, and a controllability index. The credibility index indicates the public's trust in the device being assessed, the reliability index indicates the reliability of the device's supply chain, and the controllability index indicates the controllability of the device under the current regulatory environment. These indicators dynamically reflect the potential impact of the current external environment on the device being assessed.

[0062] After obtaining the exogenous data of the device to be evaluated, some preset algorithms can be used to analyze the exogenous data to obtain the credibility index of the credibility indicator of the device to be evaluated, the reliability index of the reliability indicator of the device to be evaluated, and the controllability index of the controllability indicator of the device to be evaluated.

[0063] S103: Based on historical exogenous assessment data and exogenous risk assessment results, adjust the weights of multiple source indicators and obtain target assessment weights; the target assessment weights include the weights of endogenous self-controlled indicators and the weights of exogenous risk indicators.

[0064] The historical exogenous assessment data refers to the data from the previous assessment cycle of the device being assessed. This application uses a dynamic weighting mechanism. In one example, the exogenous risk assessment results of the current assessment cycle can be compared with the exogenous risk assessment results of the previous cycle. Each indicator is weighted based on its relevance and real-time risk level. For example, when the risk of the reliability indicator increases, the weight of the reliability indicator is increased.

[0065] S104: Determine the integrated assessment result of the device to be assessed based on the internal self-controlled assessment result, the external risk assessment result, and the target assessment weight.

[0066] Finally, the autonomous and controllable assessment score of the device to be evaluated in this assessment cycle can be calculated based on the endogenous self-control assessment results, the exogenous risk assessment results and the target assessment weight. Based on the autonomous and controllable assessment score of the device to be evaluated in this assessment cycle, the integrated assessment result of the device to be evaluated in this assessment cycle can be obtained.

[0067] The multi-source fusion autonomous controllable evaluation method based on exogenous risk perception includes obtaining exogenous data related to the device to be evaluated, historical exogenous evaluation data of the device to be evaluated, and endogenous self-controllable evaluation results of the endogenous self-controllable indicators corresponding to the device to be evaluated. Then, based on the exogenous data, the exogenous risk evaluation results of multiple exogenous risk indicators are obtained. Then, based on the historical exogenous evaluation data and exogenous risk evaluation results of the device to be evaluated, the change between the exogenous risk evaluation results and the historical exogenous risk evaluation results is determined. It is judged in turn whether it is necessary to adjust the weights of the endogenous self-controllable indicators and the weights of the exogenous risk indicators. Finally, the fusion evaluation results of the device to be evaluated are determined based on the endogenous self-controllable evaluation results, the exogenous risk evaluation results and the target evaluation weights. The multi-source fusion autonomous controllable evaluation method based on exogenous risk perception of the present application, on the basis of the traditional quantitative analysis that only relies on the technology of the device itself, that is, the autonomous controllability of the power equipment is analyzed only based on the endogenous self-controllable evaluation results of the device to be evaluated, incorporates the analysis of exogenous data related to the device to be evaluated, so that the fusion evaluation results of the device to be evaluated can reflect the impact of the external environment on the autonomous controllable capability of the power equipment, and can timely warn of potential risks.

[0068] In an exemplary embodiment, please participate Figure 2 , step S101, obtaining external data related to the device to be evaluated, including steps S201 to S204.

[0069] S201: Obtain a knowledge graph of the device to be evaluated; wherein the knowledge graph includes multiple entity nodes and the association relationship between each entity node, and the multiple entity nodes include at least part of the device to be evaluated, parts of the device to be evaluated, the manufacturer of the device to be evaluated, the supplier of parts, and the supervisors of the manufacturer and supplier.

[0070] It can be understood that the device to be evaluated is assembled from multiple parts, and the device to be evaluated has at least one manufacturer. The manufacturer of the device to be evaluated purchases the parts of the device to be evaluated from the supplier to produce the device to be evaluated. At the same time, the manufacturer of the device to be evaluated and the manufacturer of the parts also have a supervisor. In one example, the supervisor can formulate production standards for the manufacturer and the manufacturer, so that the manufacturer of the device to be evaluated and the manufacturer of the parts produce the device to be evaluated and the parts of the device to be evaluated that meet the requirements under the prescribed production standards. Based on the above relationship between the device to be evaluated, the parts of the device to be evaluated, the manufacturer of the device to be evaluated, the supplier of the parts, the manufacturer and the supervisor of the supplier, a knowledge graph about the device to be evaluated can be formulated. In the application, when the relationship between the above entity nodes changes, the control terminal will also update the knowledge graph. For example, when the supplier of a part in the device to be evaluated changes, the control terminal will update the supplier of the part in the knowledge graph.

[0071] S202: Determine the manufacturer of the device to be evaluated, the suppliers of each component in the device to be evaluated, and the supervisors corresponding to the manufacturer and the suppliers based on the knowledge graph.

[0072] After acquiring the knowledge graph of the device to be evaluated, the control terminal can establish connections with all manufacturers, suppliers, and regulators related to the device to be evaluated. Furthermore, the control terminal can establish connections with public social platforms and news media platforms related to the power equipment to be evaluated (i.e., the device to be evaluated), the official platforms of all manufacturers and suppliers related to the device to be evaluated, and the official platforms of all regulators related to the device to be evaluated.

[0073] S203: Obtain risk status data of the equipment to be evaluated, the manufacturer, the supplier, and the regulator within a preset time period.

[0074] The control terminal can then periodically collect risk status data on the device to be assessed, as well as the manufacturer, supplier, and regulator from public social platforms, news media platforms, and official platforms of manufacturers, suppliers, and regulators. In one example, the preset time period and cycle can be seven days. The control terminal collects risk status data on the device to be assessed, the manufacturer, supplier, and regulator in real time every day, and compiles the risk status data for the seven-day period every seven days to conduct an assessment.

[0075] S204: Integrate the risk status data of the equipment to be evaluated, the manufacturer, the supplier, and the regulator to obtain external source data of the equipment to be evaluated.

[0076] After obtaining risk status data for the equipment to be assessed, the manufacturer, the supplier, and the regulator, the control terminal can perform preliminary processing and aggregation of each risk status data to generate external data. In one example, the risk status data may be in formats such as text, images, and tables. The control terminal can normalize each risk status data, converting it into text format. The normalized risk status data can then be cleaned to filter out irrelevant noise.

[0077] In an exemplary embodiment, see Figure 3 , step S102, obtaining the exogenous risk assessment results of the preset multiple exogenous risk indicators according to the exogenous data, including steps S301 to S304.

[0078] S301: Classify external data according to the trained language processing model to obtain credibility data, reliability data, and controllability data of the device to be evaluated.

[0079] For example, the language processing model can be a natural language processing (NLP) model, which can use natural language processing technology to perform word segmentation, entity recognition, and topic analysis on external data, classify the external data, and obtain credibility data related to the public's trust in the device to be evaluated, credibility data related to the reliability of the supply chain of the device to be evaluated, and controllability data related to the controllability of the device to be evaluated under the current regulatory environment. For example, risk events, evaluations, or hot topics involving the device to be evaluated and its manufacturer can be identified from social media discussion posts, news articles, and the latest production standards released by regulators. The data can then be classified according to credibility, reliability, and controllability to obtain credibility data, reliability data, and controllability data for the device to be evaluated.

[0080] S302: Determine the data source of the credibility data, and determine the credibility index of the device to be evaluated based on a preset sentiment analysis model, the credibility data, and the data source.

[0081] In this embodiment, a sentiment analysis algorithm can be used to evaluate the positive and negative tendencies of the credibility data and extract the public's confidence index in the safety and reliability of the device.

[0082] S303: Determine the reliability index of the device to be evaluated based on the trained graph neural network model, reliability data, and the knowledge graph of the device to be evaluated.

[0083] In this application, reliability data can be used to update the knowledge graph of the equipment being evaluated. Risk events related to various manufacturers and regulators in the reliability data can be linked to the respective manufacturers and regulators in the knowledge graph. A graph neural network model can then be used to conduct deep learning on the risk associations in the knowledge graph, uncovering complex association patterns. For example, the impact of a supplier's risk events on the overall autonomous controllability of power equipment can be explored. By deeply learning the risk associations of each entity node in the graph neural network, the risk contribution value of each entity node can be quantified, thereby determining the reliability index of the equipment being evaluated.

[0084] S304: Determine the controllability index of the device to be evaluated based on a preset rule engine and controllability data.

[0085] Among them, a dedicated rule engine can be combined to parse and extract structured information such as policy announcements in the controllability data, such as whether new regulatory requirements have raised technical standards, whether geopolitical events threaten the stability of the supply chain, etc., and finally obtain the controllability index of the equipment to be evaluated.

[0086] In an exemplary embodiment, the credibility data includes multiple trustworthy texts; the data sources include at least public sources and media sources. Figure 4 , step S302, determining the credibility index of the device to be evaluated according to the preset sentiment analysis model, credibility data and data source, including step S401 and step S402.

[0087] S401: Obtain the sentiment score and text topic of each trustworthy text according to the sentiment analysis model.

[0088] In this embodiment, a sentiment analysis model can be used to score each obtained trust text. For example, the sentiment score can range from -2 to +2, with a positive sentiment score for trust text expressing a positive sentiment, a negative sentiment score for trust text expressing a negative sentiment, and a score of 0 for trust text expressing a neutral sentiment. Further scoring is performed based on the intensity of the sentiment. Furthermore, the sentiment analysis model can be used to determine the textual themes of the trust texts. For example, the textual themes can include at least device safety and device reliability.

[0089] S402: Calculate the credibility index of the device to be evaluated based on the preset first source weight of the public source, the preset second source weight of the media source, the preset topic weight corresponding to each text topic, and the sentiment score, text topic and data source of each trust text.

[0090] In one example, different source weights can be set for trust texts from different sources. For example, the second source weight corresponding to a media source can be higher than the first source weight corresponding to a public source. For trust texts with different text themes, different topic weights can be set as needed. For example, the topic weight corresponding to device safety can be greater than or equal to the topic weight corresponding to device reliability. The control terminal can perform weighted calculations based on each trust text, ultimately obtaining a credibility index that reflects the public's level of trust in the device safety and reliability of the device to be evaluated.

[0091] In an exemplary embodiment, the historical exogenous evaluation data includes historical evaluation results of each exogenous risk indicator of the device to be evaluated, the external historical weight of each exogenous risk indicator, and the historical internal weight of each endogenous self-controlled indicator.

[0092] See also Figure 5 , step S103, adjust the multi-source indicator weights according to the historical exogenous evaluation data and the exogenous risk assessment results, and obtain the target evaluation weights, including steps S501 to S504.

[0093] S501: Obtain a target risk indicator from a plurality of exogenous risk indicators; wherein the target risk indicator is an exogenous risk indicator for which the difference between a historical assessment result and a corresponding exogenous risk assessment result is greater than a preset threshold.

[0094] The historical assessment results include the credibility index, reliability index, and controllability index of the device to be assessed in the previous assessment cycle. After obtaining the credibility index, reliability index, and controllability index of the device to be assessed in the current assessment cycle, the control terminal can perform a difference calculation between the credibility index, reliability index, and controllability index of the device to be assessed in the current assessment cycle and the credibility index, reliability index, and controllability index of the device to be assessed in the previous assessment cycle. If the score of an exogenous risk assessment result of the device to be assessed in the current assessment cycle is less than the score of the corresponding historical assessment result, and the difference is greater than a preset threshold, it indicates that the risk of the exogenous risk indicator is relatively high. For example, if the reliability index of the device to be assessed in the current assessment cycle is 75, the reliability index of the device to be assessed in the previous assessment cycle is 95, and the preset threshold is 15, then the reliability index of the device to be assessed is marked as a target risk indicator.

[0095] S502: In the absence of a target risk indicator, the external historical weight of each exogenous risk indicator is used as the weight of the corresponding exogenous risk indicator, and the historical internal weight of each endogenous self-controlled indicator is used as the weight of the corresponding endogenous self-controlled indicator.

[0096] If, after comparison with the credibility index, reliability index and controllability index of the previous assessment cycle, there is no target risk indicator in this assessment cycle, then when calculating the fusion assessment results of the previous assessment cycle, the external historical weights of each exogenous risk indicator and the historical internal weights of each endogenous self-controlled indicator shall be used as the weights of the corresponding exogenous risk indicators and the weights of the endogenous self-controlled indicators respectively.

[0097] In another example, when determining the target assessment weight, if there is no target risk indicator in this assessment cycle, and the index of each exogenous risk indicator remains stable for multiple consecutive cycles, the historical external weight of each exogenous risk indicator can be reduced, and the historical internal weight of each endogenous self-controlled indicator can be increased. The adjusted historical external weights can be used as the weights of the corresponding exogenous risk indicators, and the adjusted historical internal weights can be used as the weights of the corresponding endogenous self-controlled indicators.

[0098] S503: When some exogenous risk indicators are target risk indicators, the historical external weights of the target risk indicators are increased, and the historical external weights of the external risk indicators other than the target risk indicators among the multiple exogenous risk indicators are reduced, and the adjusted historical external weights are used as the weights of the corresponding exogenous risk indicators, and the historical internal weights of each endogenous self-controlled indicator are used as the weights of the corresponding endogenous self-controlled indicators.

[0099] In one example, multiple internal self-controlled indicators include the proportion of independently developed core technologies for the equipment to be evaluated, the degree of localization of key components, the availability of alternative components, the level of information security protection, and operational and maintenance autonomy. Multiple exogenous risk indicators include credibility, reliability, and controllability. In the previous historical evaluation cycle, the weight ratio between these internal self-controlled indicators and these exogenous risk indicators was 0.5:0.5. Within these internal self-controlled indicators, the historical internal weight of the proportion of independently developed core technologies for the equipment to be evaluated was 0.3, the historical internal weight of the degree of localization of key components was 0.2, the historical internal weight of the availability of alternative components was 0.1, the historical internal weight of the information security protection level was 0.2, and the historical internal weight of operational and maintenance autonomy was 0.2. Within these exogenous risk indicators, the historical external weights of the credibility indicator were 0.3, the historical external weights of the reliability indicator were 0.4, and the historical external weights of the controllability indicator were 0.3. During this assessment cycle, if the reliability indicator of the equipment being assessed is determined to be the target risk indicator, while the credibility and controllability indicators are not, the historical external weight of the reliability indicator can be increased (for example, from 0.4 to 0.5), determining the reliability indicator's weight for this assessment cycle to be 0.5. Simultaneously, the historical external weights of the credibility and controllability indicators can be reduced (for example, by adjusting the historical external weights of the credibility and controllability indicators to 0.25, respectively, determining the credibility and controllability indicators' weights for this assessment cycle to be 0.25, respectively). Furthermore, the weight ratio between multiple endogenous self-controlled indicators and multiple exogenous risk indicators remains at 0.5:0.5, and the weights between multiple endogenous self-controlled indicators remain unchanged.

[0100] S504: When all exogenous risk indicators are target risk indicators, increase the historical external weight of each exogenous risk indicator and reduce the historical internal weight of each endogenous self-controlled indicator, and use the adjusted historical external weights as the weights of the corresponding exogenous risk indicators, and use the adjusted historical internal weights as the weights of the corresponding endogenous self-controlled indicators.

[0101] Among them, if it is determined that the reliability index, credibility index and controllability index of the equipment to be evaluated are all target risk indicators, the weights between multiple exogenous risk indicators can be maintained unchanged, and the weights of each endogenous self-controlled indicator can be reduced in proportion.

[0102] For example, the multiple endogenous self-controlled indicators include the proportion of independent R&D of the core technology of the equipment to be evaluated, the degree of localization of key components, the availability of alternative components, the level of information security protection, and the autonomy of operation and maintenance. The multiple exogenous risk indicators include credibility indicators, reliability indicators, and controllability indicators. In the previous historical evaluation cycle, the weight ratio between the multiple endogenous self-controlled indicators and the multiple exogenous risk indicators was 0.5:0.5. Within the multiple endogenous self-controlled indicators, the historical internal weight of the proportion of independent R&D of the core technology of the equipment to be evaluated was 0.3, the historical internal weight of the degree of localization of key components was 0.2, the historical internal weight of the availability of alternative components was 0.1, the historical internal weight of the information security protection level was 0.2, and the historical internal weight of the autonomy of operation and maintenance was 0.2. Within the multiple exogenous risk indicators, the historical external weight of the credibility indicator was 0.3, the historical external weight of the reliability indicator was 0.4, and the historical external weight of the controllability indicator was 0.3. In this evaluation cycle, if it is determined that the reliability index, credibility index and controllability index of the equipment to be evaluated are all target risk indicators, they can be maintained within multiple exogenous risk indicators, with the weight of the credibility index remaining at 0.3, the weight of the reliability index remaining at 0.4, and the weight of the controllability index remaining at 0.3. Within multiple endogenous self-controlled indicators, the weights between the endogenous self-controlled indicators remain unchanged, and the weight ratio between multiple endogenous self-controlled indicators and multiple exogenous risk indicators is adjusted from 0.5:0.5 to 0.4:0.6.

[0103] In the application, benchmark weights can also be set for each endogenous self-controlled indicator and each exogenous risk indicator. The adjusted weights of each endogenous self-controlled indicator and each exogenous risk indicator cannot be lower than the corresponding benchmark weights to avoid excessively reducing the impact of a certain indicator on the fusion assessment results.

[0104] In an exemplary embodiment, see Figure 6 , step S104, determining the fusion evaluation result of the device to be evaluated according to the internal self-controlled evaluation result, the exogenous risk evaluation result and the target evaluation weight, including steps S601 to S604.

[0105] S601: Determine the autonomous controllability evaluation score of the device to be evaluated based on the internal autonomous control evaluation results, the external risk assessment results, and the target evaluation weight.

[0106] In the application, after determining the target assessment weight, the autonomous and controllable assessment score of the device to be assessed in this assessment cycle can be calculated based on the endogenous self-control assessment results, the exogenous risk assessment results and the target assessment weight.

[0107] S602: When the autonomous controllable evaluation score is greater than or equal to a preset first score threshold, determine that the fusion evaluation result of the device to be evaluated is autonomous controllable.

[0108] If the autonomous controllability evaluation score is greater than or equal to the preset first score threshold, it means that the degree of autonomous controllability of the device to be evaluated is high, and the fusion evaluation result of the device to be evaluated is autonomous controllable. For example, if the autonomous controllability evaluation score is greater than or equal to 80 points, it can be determined that the fusion evaluation result of the device to be evaluated is autonomous controllable.

[0109] S603: When the autonomous controllable evaluation score is less than the first score threshold and greater than or equal to the preset second score threshold, determine that the fusion evaluation result of the device to be evaluated is basically controllable.

[0110] If the autonomous controllability evaluation score is less than the first score threshold and greater than or equal to the preset second score threshold, it means that the degree of autonomous controllability of the device to be evaluated is moderate, and the fusion evaluation result of the device to be evaluated is basically controllable. For example, if the autonomous controllability evaluation score is less than 80 points and greater than or equal to 60 points, it can be determined that the fusion evaluation result of the device to be evaluated is basically controllable.

[0111] S604: When the autonomous controllable evaluation score is less than the second score threshold, determine that the fusion evaluation result of the device to be evaluated is restricted.

[0112] If the autonomous controllability evaluation score is less than the second score threshold, it means that the degree of autonomous controllability of the device to be evaluated is low, and the fusion evaluation result of the device to be evaluated is restricted. For example, if the autonomous controllability evaluation score is less than 60 points, it can be determined that the fusion evaluation result of the device to be evaluated is restricted.

[0113] In another example, the fusion assessment results can also include risk warnings and recommended solutions for the risk warnings, that is, textual descriptions and warnings for the main risk factors found. The risk warnings will list several factors in the exogenous risk indicators that have the greatest impact on autonomous controllability, such as "reliability index: the risk of interruption of a key component supplier is high", "credibility index: the public's trust in equipment reliability has declined recently", etc.

[0114] In this application, the integrated assessment results can be presented in reports or dashboards, making them easily accessible to users at different levels. For example, managers can view a simple level (autonomous controllable, basically controllable, and restricted) and the autonomous controllability evaluation score. Technical personnel can further view specific indicator data and weight distributions. Risk supervisors can also obtain risk warnings and recommended solutions. By presenting multi-dimensional results, this invention not only provides assessment conclusions but also transparently presents the risk evidence supporting these conclusions, making the assessment results more interpretable and instructive.

[0115] In a detailed embodiment, when it is necessary to evaluate the degree of autonomy and controllability of the device to be evaluated, the control terminal first obtains exogenous data about the device to be evaluated within a preset time period (for example, seven days), such as comments or articles about the device to be evaluated on public social platforms, media reports and evaluations of the device to be evaluated and its supply chain, the status of upstream and downstream enterprises in the supply chain of the device to be evaluated and market dynamics information, and information on changes in the regulatory environment released by relevant regulators, etc., as well as historical exogenous evaluation data of the device to be evaluated and endogenous self-control evaluation results of the endogenous self-control indicators corresponding to the device to be evaluated.

[0116] Afterwards, the control terminal can perform data normalization, data cleaning and data classification on the external data to obtain the credibility data, reliability data and controllability data of the equipment to be evaluated. Among them, the credibility index is used to indicate the public's trust in the equipment to be evaluated, the reliability index is used to indicate the reliability of the supply chain of the equipment to be evaluated, and the controllability index is used to indicate the controllability of the equipment to be evaluated under the current regulatory environment. These indicators dynamically reflect the potential impact of the current external environment on the equipment to be evaluated.

[0117] Next, a sentiment analysis algorithm is used to calculate the sentiment score and text topic for each trustworthy text in the credibility data. For example, the credibility data includes 340 trustworthy texts, and of the 340 related comments, 90% are negative, 7% are positive, and 3% are neutral, resulting in a calculated credibility index of only 20 points (out of a maximum score of 100). The reliability data is then used to update the knowledge graph of the device to be evaluated. Risk events related to each manufacturer and regulator in the reliability data are linked to each manufacturer and regulator in the knowledge graph. A graph neural network model is then used to perform deep learning on the risk associations in the knowledge graph, uncovering complex correlation patterns. By deeply learning the risk associations of each entity node in the graph neural network, the risk contribution of each entity node can be quantified, thereby determining the reliability index of the device to be evaluated. Furthermore, a dedicated rule engine is used to parse and extract structured information such as policy announcements from the controllability data, such as whether new regulatory requirements have raised technical standards or whether geopolitical events threaten supply chain stability. Ultimately, the controllability index of the device to be evaluated is derived.

[0118] Afterwards, target risk indicators are obtained from multiple exogenous risk indicators based on historical exogenous assessment data and exogenous risk assessment results, and it is determined whether the weights of each exogenous risk indicator and each target risk indicator need to be adjusted to ensure that major risks can be perceived and handled in advance.

[0119] Finally, the autonomous controllability assessment score of the equipment to be evaluated in this assessment cycle can be calculated based on the endogenous self-control assessment results, the exogenous risk assessment results and the target assessment weights. Based on the autonomous controllability assessment score of the equipment to be evaluated in this assessment cycle, and based on the autonomous controllability assessment score, the autonomous controllability assessment level (autonomous controllable, basically controllable, restricted), risk warnings and recommended solutions can be given.

[0120] Compared with existing autonomous controllability evaluation technologies, the solution of the present invention has the following outstanding advantages:

[0121] First, external risk awareness is enhanced. This invention incorporates external information such as social media reviews, news reports, regulatory changes, and supply chain dynamics into the evaluation system, enabling the perception and quantification of external risks. This multi-source information fusion overcomes the shortcomings of existing technologies that focus solely on internal static indicators, resulting in more comprehensive assessment results.

[0122] Second, real-time dynamic assessment. By introducing a dynamic weight adjustment mechanism and continuous data updates, this solution can respond to changes in the external environment in a timely manner. Compared with the traditional model's regular manual update method, the system can automatically adjust the weight parameters of the assessment model according to risk changes, realizing real-time autonomous controllability level assessment and risk warning, greatly improving the timeliness of the assessment.

[0123] Third, the application of artificial intelligence technologies. This invention fully utilizes AI algorithms such as natural language processing (NLP), knowledge graphs, sentiment analysis, and graph neural networks to mine effective information from massive amounts of unstructured data. The introduction of these technologies improves the accuracy and depth of risk information extraction, enabling the discovery of complex correlations and potential risk signals, surpassing the analytical capabilities of manual or traditional statistical methods.

[0124] Fourth, the comprehensive model improves accuracy. By integrating external risk indicators with internal technical indicators, the resulting dynamic evaluation model more accurately reflects the actual state of a device's autonomous controllability. In particular, when external risks become significant, the model adjusts accordingly, bringing the comprehensive score closer to reality. Experimental and simulation analysis demonstrates that, compared to models that consider only traditional indicators, the assessment results of our proposed model are more sensitive and reliable to risk fluctuations, reducing the potential for misjudgment.

[0125] Fifth, multi-dimensional results and decision support. This invention outputs not only a single score but also includes a grading and detailed risk warning information, forming a multi-dimensional assessment report. This provides targeted references for different decision-makers: decision-makers can formulate macro-strategies based on the comprehensive grading, while technical personnel can take corrective measures based on specific risk warnings. The existence of an early warning mechanism enables managers to take preemptive action to prevent risks before they occur, significantly enhancing the proactiveness of autonomous and controllable risk management.

[0126] In summary, the present invention is significantly superior to the existing technology in terms of the breadth of information acquisition, model intelligence, response speed, and practicality of results, and has significant innovation and practical value.

[0127] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0128] Based on the same inventive concept, the embodiment of the present application also provides a multi-source fusion autonomous and controllable evaluation device based on exogenous risk perception for implementing the multi-source fusion autonomous and controllable evaluation method based on exogenous risk perception. The implementation solution provided by the device is similar to the implementation solution described in the above method. Therefore, the specific limitations in the embodiments of one or more multi-source fusion autonomous and controllable evaluation devices based on exogenous risk perception provided below can be referred to the limitations of the multi-source fusion autonomous and controllable evaluation method based on exogenous risk perception above, and will not be repeated here.

[0129] In an exemplary embodiment, Figure 7 As shown, the present application provides a multi-source fusion autonomous and controllable evaluation device based on exogenous risk perception, which includes an information acquisition module 701, a risk assessment module 702, a weight determination module 703 and a result determination module 704.

[0130] The information acquisition module 701 is used to acquire exogenous data related to the device to be evaluated, historical exogenous evaluation data of the device to be evaluated, and endogenous self-control evaluation results of endogenous self-control indicators corresponding to the device to be evaluated.

[0131] The risk assessment module 702 is used to obtain exogenous risk assessment results of a plurality of preset exogenous risk indicators based on exogenous data.

[0132] The weight determination module 703 is used to adjust the weights of multiple source indicators based on historical exogenous evaluation data and exogenous risk assessment results, and obtain target evaluation weights; the target evaluation weights include the weights of endogenous self-controlled indicators and the weights of exogenous risk indicators.

[0133] The result determination module 704 is used to determine the fusion evaluation result of the device to be evaluated based on the internal self-controlled evaluation result, the external risk evaluation result and the target evaluation weight.

[0134] In an exemplary embodiment, the information acquisition module includes a graph acquisition submodule, a relationship determination submodule, a data acquisition submodule, and a data integration submodule.

[0135] The graph acquisition submodule is used to obtain a knowledge graph of the device to be evaluated; wherein the knowledge graph includes multiple entity nodes and the association relationships between the entity nodes, and the multiple entity nodes include at least part of the device to be evaluated, the parts of the device to be evaluated, the manufacturer of the device to be evaluated, the supplier of the parts, and the regulators of the manufacturer and supplier.

[0136] The relationship determination submodule is used to determine the manufacturer of the equipment to be evaluated, the suppliers of each component in the equipment to be evaluated, and the supervisors corresponding to the manufacturer and supplier based on the knowledge graph.

[0137] The data acquisition submodule is used to obtain risk status data of the equipment to be evaluated, the manufacturer, the supplier and the regulator within a preset time period.

[0138] The data integration submodule is used to integrate the risk status data of the equipment to be evaluated, the manufacturer, the supplier and the regulator to obtain the external data of the equipment to be evaluated.

[0139] In an exemplary embodiment, the plurality of exogenous risk indicators include at least part of a credibility indicator, a reliability indicator, and a controllability indicator, and the exogenous risk assessment result includes at least part of a credibility index, a reliability index, and a controllability index.

[0140] The risk assessment module includes a data classification submodule, a credibility index determination submodule, a reliability index determination submodule and a controllability index determination submodule.

[0141] The data classification submodule is used to classify external data according to the trained language processing model to obtain the credibility data, reliability data and controllability data of the equipment to be evaluated.

[0142] The credibility index determination submodule is used to determine the data source of the credibility data and determine the credibility index of the device to be evaluated based on the preset sentiment analysis model, credibility data and data source.

[0143] The reliability index determination submodule is used to determine the reliability index of the equipment to be evaluated based on the trained graph neural network model, reliability data and the knowledge graph of the equipment to be evaluated.

[0144] The controllability index determination submodule is used to determine the controllability index of the device to be evaluated based on a preset rule engine and controllability data.

[0145] In an exemplary embodiment, the credibility data includes multiple pieces of trustworthy text; and the data sources include at least public sources and media sources.

[0146] The credibility index determination submodule includes a text analysis unit and a calculation unit.

[0147] The text analysis unit is used to obtain the sentiment score and text topic of each trust text according to the sentiment analysis model.

[0148] The calculation unit is used to calculate the credibility index of the device to be evaluated based on the preset first source weight of the public source, the preset second source weight of the media source, the preset topic weight corresponding to each text topic, and the sentiment score, text topic and data source of each trust text.

[0149] In an exemplary embodiment, the historical exogenous evaluation data includes historical evaluation results of each exogenous risk indicator of the device to be evaluated, the external historical weight of each exogenous risk indicator, and the historical internal weight of each endogenous self-controlled indicator.

[0150] The weight determination module includes a target determination submodule, a first value assignment submodule, a second value assignment submodule and a third value assignment submodule.

[0151] The target determination submodule is used to obtain a target risk indicator from multiple exogenous risk indicators; wherein the target risk indicator is an exogenous risk indicator whose difference between a historical assessment result and a corresponding exogenous risk assessment result is greater than a preset threshold.

[0152] The first assignment submodule is used to use the external historical weight of each exogenous risk indicator as the weight of the corresponding exogenous risk indicator, and to use the historical internal weight of each endogenous self-controlled indicator as the weight of the corresponding endogenous self-controlled indicator when there is no target risk indicator.

[0153] The second assignment sub-module is used to increase the historical external weight of the target risk indicator and reduce the historical external weight of the external risk indicators other than the target risk indicator among multiple exogenous risk indicators when some exogenous risk indicators are target risk indicators, and use the adjusted historical external weights as the weights of the corresponding exogenous risk indicators, and use the historical internal weights of each endogenous self-controlled indicator as the weight of the corresponding endogenous self-controlled indicator.

[0154] The third assignment sub-module is used to increase the historical external weight of each exogenous risk indicator and reduce the historical internal weight of each endogenous self-controlled indicator when all exogenous risk indicators are target risk indicators, and use the adjusted historical external weights as the weights of the corresponding exogenous risk indicators, and use the adjusted historical internal weights as the weights of the corresponding endogenous self-controlled indicators.

[0155] In an exemplary embodiment, the result determination module includes a score determination submodule, a first evaluation submodule, a second evaluation submodule, and a third evaluation submodule.

[0156] The score determination submodule is used to determine the autonomous controllability evaluation score of the device to be evaluated based on the endogenous autonomous control evaluation results, the exogenous risk assessment results and the target assessment weight.

[0157] The first evaluation submodule is used to determine that the fusion evaluation result of the device to be evaluated is autonomous and controllable when the autonomous and controllable evaluation score is greater than or equal to a preset first score threshold.

[0158] The second evaluation submodule is used to determine that the fusion evaluation result of the device to be evaluated is basically controllable when the autonomous controllable evaluation score is less than the first score threshold and greater than or equal to a preset second score threshold.

[0159] The third evaluation submodule is configured to determine that the fusion evaluation result of the device to be evaluated is restricted when the autonomous controllable evaluation score is less than a second score threshold.

[0160] Each module in the aforementioned multi-source fusion autonomous and controllable assessment device based on exogenous risk perception can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a computer device's memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0161] In an exemplary embodiment, the present application provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described in any one of the above embodiments when executing the computer program.

[0162] The internal structure diagram of the computer device can be as follows Figure 8As shown. The computer device includes a processor, memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, memory, and input / output interface are connected via a system bus, while the communication interface, display unit, and input device are connected to the system bus via the input / output interface. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operating system and computer programs stored in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals via wired or wireless means. The wireless means can be implemented via Wi-Fi, mobile cellular networks, near-field communication (NFC), or other technologies. When executed by the processor, the computer program implements a multi-source fusion autonomous and controllable assessment method based on exogenous risk perception. The display unit of the computer device is used to produce visual images and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse.

[0163] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0164] In an exemplary embodiment, the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method in any of the above embodiments are implemented.

[0165] In an exemplary embodiment, the present application provides a computer program product, including a computer program, which implements the steps of the method in any of the above embodiments when the computer program is executed by a processor.

[0166] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0167] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), quantum computing-based data processing logic devices, artificial intelligence (AI) processors, and the like.

[0168] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0169] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A multi-source fusion autonomous and controllable evaluation method based on exogenous risk perception, characterized by: The method comprises: Acquire exogenous data related to the device to be evaluated, historical exogenous evaluation data of the device to be evaluated, and endogenous self-control evaluation results of endogenous self-control indicators corresponding to the device to be evaluated; Obtaining exogenous risk assessment results of a plurality of preset exogenous risk indicators based on the exogenous data; Adjust the weights of the multi-source indicators based on the historical exogenous assessment data and the exogenous risk assessment results, and obtain a target assessment weight; the target assessment weight includes the weight of the endogenous self-controlled indicator and the weight of the exogenous risk indicator; Determine a fusion evaluation result of the device to be evaluated based on the endogenous self-controlled evaluation result, the exogenous risk evaluation result, and the target evaluation weight.

2. The method according to claim 1, characterized in that The obtaining of external data related to the device to be evaluated includes: Obtaining a knowledge graph of the device to be evaluated; wherein the knowledge graph includes a plurality of entity nodes and association relationships between the entity nodes, the plurality of entity nodes including at least some of the device to be evaluated, components of the device to be evaluated, the manufacturer of the device to be evaluated, suppliers of the components, and supervisors of the manufacturer and the supplier; Determine, based on the knowledge graph, the manufacturer of the device to be evaluated, the suppliers of each component in the device to be evaluated, and the supervisors corresponding to the manufacturer and the suppliers; Obtain risk status data of the device to be evaluated, the manufacturer, the supplier, and the regulator within a preset time period; The risk status data of the device to be evaluated, the manufacturer, the supplier, and the regulator are integrated to obtain external source data of the device to be evaluated.

3. The method according to claim 1, characterized in that The multiple exogenous risk indicators include at least part of a credibility indicator, a reliability indicator, and a controllability indicator, and the exogenous risk assessment result includes at least part of a credibility index, a reliability index, and a controllability index; The step of obtaining exogenous risk assessment results of a plurality of preset exogenous risk indicators based on the exogenous data includes: Classifying the external data according to the trained language processing model to obtain credibility data, reliability data, and controllability data of the device to be evaluated; Determining a data source of the credibility data, and determining a credibility index of the device to be evaluated based on a preset sentiment analysis model, the credibility data, and the data source; Determine a reliability index of the device to be evaluated based on the trained graph neural network model, the reliability data, and the knowledge graph of the device to be evaluated; The controllability index of the device to be evaluated is determined according to a preset rule engine and the controllability data.

4. The method according to claim 3, characterized in that The credibility data includes multiple credibility texts; the data sources include at least public sources and media sources; Determining the credibility index of the device to be evaluated based on a preset sentiment analysis model, the credibility data, and the data source includes: Obtaining the sentiment score and text topic of each of the trustworthy texts according to the sentiment analysis model; The credibility index of the device to be evaluated is calculated based on the preset first source weight of the public source, the preset second source weight of the media source, the preset topic weight corresponding to each text topic, and the sentiment score, text topic and data source of each trust text.

5. The method according to any one of claims 1 to 4, characterized in that The historical exogenous evaluation data includes the historical evaluation results of each of the exogenous risk indicators of the equipment to be evaluated, the external historical weight of each of the exogenous risk indicators, and the historical internal weight of each of the endogenous self-controlled indicators; The step of adjusting the weights of multiple indicators based on the historical exogenous assessment data and the exogenous risk assessment results and obtaining target assessment weights includes: Obtaining a target risk indicator from a plurality of exogenous risk indicators; wherein the target risk indicator is an exogenous risk indicator for which the difference between a historical assessment result and a corresponding exogenous risk assessment result is greater than a preset threshold; In the absence of a target risk indicator, the external historical weight of each exogenous risk indicator is used as the weight of the corresponding exogenous risk indicator, and the historical internal weight of each endogenous self-controlled indicator is used as the weight of the corresponding endogenous self-controlled indicator; In the case where some exogenous risk indicators are target risk indicators, the historical external weights of the target risk indicators are increased, and the historical external weights of the external risk indicators other than the target risk indicators among the multiple exogenous risk indicators are reduced, and the adjusted historical external weights are used as the weights of the corresponding exogenous risk indicators, and the historical internal weights of each endogenous self-controlled indicator are used as the weights of the corresponding endogenous self-controlled indicator; When all the exogenous risk indicators are target risk indicators, the historical external weights of the exogenous risk indicators are increased, and the historical internal weights of the endogenous self-controlled indicators are reduced, and the adjusted historical external weights are used as the weights of the corresponding exogenous risk indicators, and the adjusted historical internal weights are used as the weights of the corresponding endogenous self-controlled indicators.

6. The method according to claim 1, characterized in that The determining, based on the endogenous self-controlled evaluation result, the exogenous risk evaluation result, and the target evaluation weight, of the fusion evaluation result of the device to be evaluated includes: Determine the autonomous controllability evaluation score of the device to be evaluated based on the endogenous autonomous control evaluation result, the exogenous risk evaluation result, and the target evaluation weight; When the autonomous controllable evaluation score is greater than or equal to a preset first score threshold, determining that the fusion evaluation result of the device to be evaluated is autonomous controllable; When the autonomous controllable evaluation score is less than the first score threshold and greater than or equal to a preset second score threshold, determining that the fusion evaluation result of the device to be evaluated is basically controllable; When the autonomous controllable evaluation score is less than the second score threshold, it is determined that the fusion evaluation result of the device to be evaluated is restricted.

7. A multi-source fusion autonomous and controllable evaluation device based on exogenous risk perception, characterized in that: The device comprises: An information acquisition module is used to acquire exogenous data related to the device to be evaluated, historical exogenous evaluation data of the device to be evaluated, and endogenous self-control evaluation results of endogenous self-control indicators corresponding to the device to be evaluated; A risk assessment module, configured to obtain exogenous risk assessment results of a plurality of preset exogenous risk indicators based on the exogenous data; A weight determination module is used to adjust the weights of multiple source indicators based on the historical exogenous evaluation data and the exogenous risk assessment results, and obtain a target evaluation weight; the target evaluation weight includes the weight of the endogenous self-controlled indicator and the weight of the exogenous risk indicator; A result determination module is used to determine the fusion evaluation result of the device to be evaluated according to the endogenous self-controlled evaluation result, the exogenous risk evaluation result and the target evaluation weight.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images