A trust evaluation method and device applied to a power internet of things terminal device
By introducing interaction importance and recommendation importance indicators, setting a penalty mechanism, and adopting a differential integrated moving average autoregressive model, the accuracy and timeliness issues of trust assessment for power Internet of Things terminal devices are solved, the risk of switch attacks is reduced, and a more objective and real-time trust assessment is achieved.
Patent Information
- Application Number
- CN202210129929.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-02-11
- Publication Date
- 2026-01-02
- Estimated Expiration
- 2042-02-11
AI Technical Summary
Existing trust assessment methods for power IoT terminal devices are inaccurate, lack means to deal with internal attacks, and traditional trust assessment methods lack dynamism and timeliness, making them ineffective in dealing with switch attacks.
We introduce interaction importance and recommendation importance indicators, set up on/off attack penalties and trust decline penalties, calculate the direct trust value and recommendation trust value of target consumers through weighted fusion, and use a differential integrated moving average autoregressive model to predict the trust value.
It improves the accuracy and timeliness of trust assessment, reduces the risk of switch attacks, and ensures the objectivity and real-time nature of trust assessment results.
Smart Images
Figure CN116633794B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of Internet of Things, and in particular to a trust evaluation method and device applied to a power Internet of Things terminal device. BACKGROUND
[0002] There are a large number of terminal devices in the power Internet of Things, and the terminal has functions such as data acquisition, transmission, processing, storage, and the like, and the number is large and the types are complex. With the continuous development of the power Internet of Things, the security threats faced by the terminal device are increasing, and in some cases, malicious terminals perform well in unimportant interactions, thereby increasing their trust value, and then this trust value can be used in important interactions to perform malicious behavior, which is called on-off attack.
[0003] In recent years, trust management is considered to be one of the effective protection mechanisms to ensure network security, and is an effective way to resist internal attacks. Internal attacks are initiated by malicious terminals or denial-of-service terminals in the network, including on-off attacks, collusion attacks, etc. Trust management is based on the historical behavior of the terminal to evaluate the trust value of the terminal, so as to estimate the credibility of the terminal in performing a specific task. The existing trust degree of the terminal is calculated by using positive and negative feedback factors. This method is not accurate and does not consider the importance of interaction, and is significantly subjective. And lack of means to deal with internal attacks, so it is necessary to design a punishment mechanism based on the importance of interaction. Most researches improve the trust management based on beta distribution and binomial distribution to improve the accuracy of trust evaluation, but both of them use subjective method to allocate trust factors, and the allocation method of weight in trust evaluation is relatively single, which will lead to inaccurate trust evaluation. SUMMARY
[0004] Therefore, the technical problem to be solved by the present application is to overcome the defect that the trust evaluation is not accurate in the prior art, so as to provide a trust evaluation method and device applied to a power Internet of Things terminal device.
[0005] The first aspect of the present application provides a trust evaluation method applied to a power Internet of Things terminal device, comprising: calculating the direct trust value of a target consumer according to the interaction feedback of different consumers, the interaction importance index, the on / off attack punishment, and the trust decline punishment; obtaining a direct trust evaluation set of each recommender to each consumer, calculating the recommended interaction trust value of each recommender to the target consumer, and the recommendation importance index of the consumer is greater than the interaction importance index; calculating the recommended trust value of the target consumer according to the recommended interaction trust value of each recommender to the target consumer; and weighting and fusing the direct trust value and the recommended information value of the target consumer to obtain the comprehensive trust value of the target consumer.
[0006] Optionally, in the trust evaluation method applied to the power Internet of Things terminal device provided by the application, the direct trust value of the target consumer is calculated according to the interaction feedback of different consumers, the interaction importance index, the on / off attack penalty and the trust decline penalty, and the method comprises the following steps: according to the interaction feedback of different consumers, the initial positive feedback value and the initial negative feedback value are calculated; the updated positive feedback value and the updated negative feedback value are calculated according to the initial positive feedback value, the initial negative feedback value and the interaction importance index; and the direct trust value of the target consumer is calculated according to the updated positive feedback value, the updated negative feedback value, the on / off attack penalty and the trust decline penalty.
[0007] Optionally, in the trust evaluation method applied to the power Internet of Things terminal device provided by the application, the on / off attack penalty and the trust decline penalty are determined in combination with the comparison result of the initial positive feedback value and the interaction importance index.
[0008] Optionally, in the trust evaluation method applied to the power Internet of Things terminal device provided by the application, the recommended interaction trust value of each recommender to the target consumer is calculated, which comprises the following steps: the percentile of each direct trust evaluation set is calculated respectively; the subjective elimination parameter of each recommender is calculated according to the percentile; and the recommended interaction trust value of each recommender to the target consumer is calculated according to the direct trust evaluation set and the subjective elimination parameter of each recommender.
[0009] Optionally, in the trust evaluation method applied to the power Internet of Things terminal device provided by the application, the recommendation importance index is determined according to the historical interaction record between the terminal or the user, the last interaction record, the weight of the historical interaction record and the weight of the last interaction record.
[0010] Optionally, in the trust evaluation method applied to the power Internet of Things terminal device provided by the application, the attribute index of the subject and object behavior is determined according to the historical interaction record; the local variable weight vector is calculated according to the constant weight vector of each attribute index and the local state variable weight mode; and the weight of the direct trust value is calculated according to the local state variable weight mode and the local variable weight vector.
[0011] Optionally, in the trust evaluation method applied to the power Internet of Things terminal device provided by the application, a trust value time sequence is constructed according to the evaluation result of the comprehensive trust value of the target consumer in a preset time period; if the trust value time sequence is a stationary sequence, the autocorrelation coefficient and the partial autocorrelation coefficient of the trust value time sequence are calculated; an autoregressive integrated moving average model is established according to the autocorrelation coefficient and the partial autocorrelation coefficient, and the order and the end of the autoregressive integrated moving average model are determined according to the autocorrelation coefficient and the partial autocorrelation coefficient; and the trust value of the target consumer is predicted through the autoregressive integrated moving average model to obtain a trust value prediction result.
[0012] The second aspect of the present application provides a trust evaluation device applied to a power Internet of Things terminal device, comprising: a direct trust value calculation module, configured to calculate a direct trust value of a target consumer according to interaction feedback of different consumers, an interaction importance index, an on / off attack penalty and a trust decline penalty; a recommended interaction trust value calculation module, configured to obtain a direct trust evaluation set of each recommender to each consumer, and calculate a recommended interaction trust value of each recommender to the target consumer, wherein the recommendation importance index of the consumer is greater than the interaction importance index; a recommendation trust value calculation module, configured to calculate a recommendation trust value of the target consumer according to the recommended interaction trust value of each recommender to the target consumer; and a comprehensive trust value calculation module, configured to perform weighted fusion calculation on the direct trust value and the recommendation information value of the target consumer to obtain a comprehensive trust value of the target consumer.
[0013] The third aspect of the present application provides a computer device, comprising: at least one processor; and a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to perform the trust evaluation method applied to the power Internet of Things terminal device provided by the first aspect of the present application.
[0014] The fourth aspect of the present application provides a computer readable storage medium, which stores computer instructions, and the computer instructions are used to make a computer execute the trust evaluation method applied to the power Internet of Things terminal device provided by the first aspect of the present application.
[0015] The technical scheme of the present application has the following advantages:
[0016] The trust evaluation method and device applied to the power Internet of Things terminal device provided by the present application introduce the interaction importance index and the recommendation importance index when calculating the direct trust value and the recommendation trust value of the target consumer, solve the problem that the negative feedback and the positive feedback evaluation are not accurate in the traditional trust model, improve the accuracy of the interaction trust calculation, and set the penalty mechanism for the on / off attack and the trust decline penalty mechanism when calculating the direct trust value, adopt the penalty mechanism to reduce the danger of the on / off attack, retain the evaluation of the recent behavior to a certain extent through the feedback, eliminate the subjectivity of the recommendation trust, solve the problem that the traditional trust evaluation method lacks dynamic and is difficult to guarantee the timeliness, and make the trust evaluation result more objective and real-time. BRIEF DESCRIPTION OF DRAWINGS
[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0018] Figure 1 A flowchart illustrating a specific example of a trust assessment method applied to power Internet of Things (IoT) terminal devices in this invention.
[0019] Figure 2 This is a schematic diagram illustrating a specific example of a trust assessment device applied to a power Internet of Things (IoT) terminal device in an embodiment of the present invention.
[0020] Figure 3 This is a schematic diagram illustrating a specific example of a computer device in an embodiment of the present invention. Detailed Implementation
[0021] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0022] In the description of this invention, it should be noted that the technical features involved in the different embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.
[0023] This invention provides a trust assessment method for power Internet of Things (IoT) terminal devices, such as... Figure 1 As shown, it includes:
[0024] Step S11: Calculate the direct trust value of the target consumer based on the interaction feedback of different consumers, interaction importance index, on / off attack penalty, and trust decline penalty.
[0025] In an optional embodiment, both the on / off attack penalty and the trust reduction penalty are values ranging from 1 to... The set of 1s indicates that the character is "not in danger". It indicates "high risk".
[0026] Step S12: Obtain the set of direct trust assessments of each recommender for each consumer, calculate the trust value of each recommender for the target consumer, and the consumer's recommendation importance index is greater than the interaction importance index.
[0027] In an optional embodiment, the direct trust evaluation set of the recommender is determined by the direct trust value of the recommender to all consumers.
[0028] Step S13: calculating the recommendation trust value of the target consumer according to the recommendation interaction trust value of each recommender to the target consumer.
[0029] In an optional embodiment, the sum of the recommendation interaction trust values of all recommenders providing services for the target consumer to the target consumer is determined as the recommendation trust value of the target consumer.
[0030] Step S14: performing weighted fusion calculation on the direct trust value and the recommendation information value of the target consumer to obtain the comprehensive trust value of the target consumer.
[0031] The trust evaluation method applied to the power Internet of Things terminal device provided by the embodiment of the application introduces the interaction importance index and the recommendation importance index when calculating the direct trust value and the recommendation trust value of the target consumer, solves the problem that the negative feedback and the positive feedback evaluation are not accurate in the traditional trust model, improves the accuracy of the calculation of the interaction trust, and sets the punishment mechanism for coping with the on / off attack and the trust decline punishment mechanism when calculating the direct trust value, adopts the punishment mechanism to reduce the danger of the on / off attack, retains the evaluation of the recent behavior to a certain extent through feedback, solves the problem that the traditional Bayesian trust evaluation method lacks dynamic and is difficult to guarantee timeliness by eliminating the subjectivity of the recommendation trust, and makes the trust evaluation result more objective and real-time.
[0032] In an optional embodiment, the above step S11 specifically comprises:
[0033] First, the initial positive feedback value and the initial negative feedback value are calculated according to the interaction feedback of different consumers.
[0034] In an optional embodiment, the initial positive feedback value at the t time is calculated by the following formula:
[0035] wherein, represents the n th consumer, represents each input feedback value F.
[0036] In an optional embodiment, the initial negative feedback value at the t time is calculated by the following formula:
[0037] .
[0038] Then, the updated positive feedback value and the updated negative feedback value are calculated according to the initial positive feedback value, the initial negative feedback value and the interaction importance index.
[0039] In an optional embodiment, the updated positive feedback value is calculated by the following formula:
[0040] wherein, represents the initial positive feedback value of the consumer at time t, represents the interaction importance indicator, represents the number of input feedbacks.
[0041] In an optional embodiment, the updated negative feedback value is calculated by the following formula:
[0042] , represents the initial negative feedback value of the consumer at time t.
[0043] Finally, the direct trust value of the target consumer is calculated according to the updated positive feedback value, the updated negative feedback value, the on / off attack penalty, and the trust decrease penalty.
[0044] In an optional embodiment, the direct trust value of the target consumer (CR) at time t is calculated by the following formula:
[0045] ,
[0046] wherein, represents the initial positive feedback value of the i-th consumer, represents the updated positive feedback value of the target consumer, represents the initial negative feedback value of the i-th consumer, represents the updated negative feedback value of the target consumer, represents the threshold value of the on / off attack penalty, represents the trust decrease penalty value.
[0047] In an optional embodiment, the on / off attack penalty and the trust decrease penalty are determined in combination with the comparison result of the initial positive feedback value and the interaction importance indicator.
[0048] In an optional embodiment, the threshold value of the on / off attack penalty is set as if the minimum value of the high interaction satisfies and , then ; otherwise wherein, represents the limit of the high interaction, .
[0049] In an optional embodiment, the trust decrease penalty is set as if , then , otherwise wherein , represents the limit of the low interaction, and is an integer greater than 1.
[0050] In the above embodiment, the recommender whose recommendation importance indicator is greater than the interaction importance indicator is used to calculate the recommendation trust value of the target consumer. In an alternative embodiment, the recommendation importance indicator is determined according to the historical interaction records between the terminal or the user, the last interaction record, and the interaction weight function of the historical interaction records , the final interaction weight function of the last interaction record , wherein .
[0051] In an alternative embodiment, the recommendation importance indicator of the ith recommender is calculated by the following formula:
[0052] ,
[0053] wherein represents the number of interactions between the recommender and the specific consumer, represents the time of the last interaction between the recommender and the target consumer, , respectively represent their weight functions.
[0054] The step of calculating the recommendation interaction trust value of each recommender to the target consumer specifically comprises:
[0055] First, the percentile of each direct trust evaluation set is calculated respectively.
[0056] In an alternative embodiment, the recommender interacts with all consumers in the role of the recommender R , and the percentile of the direct trust evaluation set is , wherein represents the index of the first trust value in the IR set, represents the number of all consumers interacting with the recommender R.
[0057] Then, the subjective elimination parameter of each recommender is calculated according to the percentile respectively.
[0058] In an alternative embodiment, the subjective elimination parameter includes parameter A and parameter B, wherein , , is an integer, is a fraction and .
[0059] Finally, the recommendation interaction trust value of each recommender to the target consumer is calculated according to the direct trust evaluation set and the subjective elimination parameter of each recommender respectively.
[0060] In an optional embodiment, by the recommender Direct trust assessment set for interactions with all consumer roles Calculate recommenders As a service provider The user's trust value for recommended interactions is:
[0061] ,in , Then it is determined by the value of parameter A, satisfying .
[0062] In an optional embodiment, in step S13 above, the recommendation trust value of the target consumer is calculated using the following formula:
[0063] ,
[0064] Where n represents the number of recommenders. This represents the trust value of the i-th recommender in the recommendation interaction with the target consumer.
[0065] In existing technical solutions, the recommendation trust value of the target consumer at time t can be subjectively calculated. ,in This indicates that the importance of recommendation is greater than the importance of interaction (i.e., The direct trust value of recommender R to target consumer CR. This indicates the number of recommenders, and in this embodiment of the invention, trust is established through recommendation interaction. replace Eliminating the subjectivity of recommendation trust improves the accuracy of trust assessment results.
[0066] In an optional embodiment, when performing a weighted fusion calculation on the direct trust value and the recommended trust value in step S14 above, it is necessary to determine the weights of the direct trust value and the recommended trust value. In this embodiment of the invention, the weight of the direct trust value is calculated through the following steps:
[0067] First, determine the attribute indicators of the subject and object behaviors based on historical interaction records.
[0068] In an optional embodiment, before performing step S11, a trust assessment dataset for subject-object behavior is constructed from three perspectives—user, terminal, and application service—based on the historical interaction trust assessment records of the power mobile application's subjects and objects. Each dataset is then cleaned to remove incomplete or erroneous data noise, thus performing data preprocessing and deleting incomplete behavioral information. Subject-object behaviors include user authentication, user access, terminal verification, and application service functions, and each subject-object behavior contains multiple attributes.
[0069] Then, the local variable weight vector is calculated according to the constant weight vector of each attribute index and the local state variable weight mode.
[0070] Finally, the weight of the direct trust value is calculated according to the local state variable weight mode and the local variable weight vector.
[0071] After the attribute indexes of the subject and object behaviors are numerically valued, the index set is , the constant weight vector is , and the evaluation value of each index of the target consumer is .
[0072] The local state variable weight vector is determined as , so that the local variable weight vector is .
[0073] For the local state variable weight of the trust evaluation weight analysis, from the perspective of feasibility evaluation, the mode should be selected as follows:
[0074] , j = 1, 2,..., 8,
[0075] Where T, U, V, and C are parameters with values of [0, 1], which are determined by the trust evaluation system, T is the negative level, U is the passing level, V is the incentive level, and C is the adjustment level.
[0076] The weight of the direct trust value is: .
[0077] In an optional embodiment, after the weight of the direct trust value is determined , 1- is determined as the weight of the recommendation information value, and the comprehensive trust value of the target consumer is .
[0078] The embodiment of the application adopts the weight analysis combining the subject and object attribute evaluation and the local variable weight, sets standards for each attribute index in the direct and indirect trust evaluation by introducing the attributes of the interactive behaviors, determines the weights of the direct and indirect trust based on the attribute decision preference by adopting punishment and incentive on the index weight, avoids the subjectivity of the experience setting weight, and makes the target consumer tend to interact in the direction expected by the trust evaluation system.
[0079] In an optional embodiment, a trust threshold is set for the target consumer, if the comprehensive trust value of the target consumer is greater than or equal to the trust threshold, the target consumer is allowed to access, and if the comprehensive trust value of the target consumer is less than the trust threshold, the target consumer is not allowed to access.
[0080] In an optional embodiment, the trust evaluation method applied to the power Internet of Things terminal device provided by the embodiment of the application further comprises,
[0081] A trust value time series is constructed based on the comprehensive trust value assessment results of target consumers within a preset time period, and it is determined whether the trust time series is a stationary series.
[0082] In one alternative embodiment, after constructing the trust time series, the data is plotted to observe whether the arbitrary time series is a stationary series.
[0083] If the trust value time series is a stationary series, perform the following steps:
[0084] First, calculate the autocorrelation coefficient and partial autocorrelation coefficient of the trust value time series.
[0085] Then, a differential integrated moving average autoregressive model is established based on the autocorrelation coefficient and the partial autocorrelation coefficient. The hierarchy and termination of the differential integrated moving average autoregressive model are determined based on the autocorrelation coefficient and the partial autocorrelation coefficient.
[0086] In one optional embodiment, after establishing the differential integrated moving average autoregressive model and estimating the model parameters, parametric tests and residual white noise tests are performed. If the residual sequence is not a white noise sequence, the model is reconstructed until it passes the tests.
[0087] Finally, the trust value of the target consumers is predicted by the differential integrated moving average autoregressive model, and the trust value prediction results are obtained.
[0088] By using a differentially integrated moving average autoregressive model to predict the trust value of target consumers, the trust value is updated, thus achieving continuous trust assessment.
[0089] In an optional embodiment, if the trust value time series is not a stationary series, then a d-order difference operation is performed on the trust value time series to convert the trust value time series into a stationary time series.
[0090] This invention utilizes a differential integrated moving average autoregressive model for continuous trust assessment. Since all statistical characteristics of the data sequence of the final trust assessment result are independent of time, the use of a tested model for trust prediction solves the problem of trust decaying over time in trust assessment, making the trust assessment results more accurate.
[0091] This invention provides a trust assessment device for power Internet of Things (IoT) terminal equipment, such as... Figure 2 As shown, it includes:
[0092] The direct trust value calculation module 21 is used to calculate the direct trust value of the target consumer based on the interaction feedback of different consumers, interaction importance indicators, on / off attack penalties, and trust decline penalties. For details, please refer to the description of step S11 in the above embodiment, which will not be repeated here.
[0093] The recommendation interaction trust value calculation module 22 is configured to obtain a set of direct trust evaluations of each recommender on each consumer, and calculate a recommendation interaction trust value of each recommender on the target consumer, wherein the recommendation importance index of the consumer is greater than the interaction importance index. For details, refer to the description of step S12 in the above embodiment, which will not be repeated here.
[0094] The recommendation trust value calculation module 23 is configured to calculate the recommendation trust value of the target consumer according to the recommendation interaction trust value of each recommender on the target consumer. For details, refer to the description of step S13 in the above embodiment, which will not be repeated here.
[0095] The comprehensive trust value calculation module 24 is configured to perform weighted fusion calculation on the direct trust value and the recommendation information value of the target consumer to obtain a comprehensive trust value of the target consumer. For details, refer to the description of step S14 in the above embodiment, which will not be repeated here.
[0096] The computer device provided by the embodiment of the present application comprises: Figure 3 As shown in the figure, the computer device mainly comprises one or more processors 31 and a memory 32, Figure 3 For example, taking the processor 31 as an example.
[0097] The computer device can also comprise an input device 33 and an output device 34.
[0098] The processor 31, the memory 32, the input device 33 and the output device 34 can be connected through a bus or other means, Figure 3 For example, taking the connection through the bus as an example.
[0099] The processor 31 can be a central processing unit (CPU). The processor 31 can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic components, discrete hardware components, or a combination thereof. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The memory 32 can include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required by a function, and the data storage area can store data created according to the use of the trust evaluation device. In addition, the memory 32 can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory 32 can optionally include a memory disposed remotely relative to the processor 31, and these remote memories can be connected to the trust evaluation device through a network. The input device 33 can receive a user inputted calculation request (or other digital or character information), and generate a key signal input related to the trust evaluation device. The output device 34 can include a display device such as a display screen, to output a calculation result.
[0100] The embodiments of the present application provide a computer readable storage medium storing computer instructions, and the computer storage medium stores computer executable instructions, which can execute the trust evaluation method applied to the power Internet of Things terminal device in any method embodiment described above. The storage medium can be a disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD), etc. The storage medium can also include a combination of the above types of memories.
[0101] Obviously, the above embodiments are merely example for clearly illustrating but not limitation to the embodiments. Based on the above description, other different forms of changes or variations can be made by those skilled in the art. Here, all the embodiments need not and can not be enumerated. The obvious changes or variations derived from the above description are still within the protection scope of the present application.
Claims
1. A trust evaluation method applied to a power Internet of Things terminal device, characterized in that, The method comprises the following steps: calculating the direct trust value of the target consumer according to the interaction feedback of different consumers, the interaction importance index, the on / off attack penalty and the trust decline penalty; obtaining a direct trust evaluation set of each recommender to each consumer, and calculating the recommendation interaction trust value of each recommender to the target consumer, wherein the recommendation importance index of the consumer is greater than the interaction importance index; calculating the recommendation trust value of the target consumer according to the recommendation interaction trust value of each recommender to the target consumer; performing weighted fusion calculation on the direct trust value and the recommendation information value of the target consumer to obtain the comprehensive trust value of the target consumer; calculating the recommendation interaction trust value of each recommender to the target consumer comprises: calculating the percentile of each direct trust evaluation set respectively; calculating the subjective elimination parameter of each recommender according to the percentile respectively; calculating the recommendation interaction trust value of each recommender to the target consumer according to the direct trust evaluation set and the subjective elimination parameter of each recommender. 2.The trust evaluation method applied to the power internet of things terminal device of claim 1, wherein, The method comprises the following steps: calculating the initial positive feedback value and the initial negative feedback value according to the interaction feedback of different consumers; calculating the updated positive feedback value and the updated negative feedback value according to the initial positive feedback value, the initial negative feedback value and the interaction importance index; calculating the direct trust value of the target consumer according to the updated positive feedback value, the updated negative feedback value, the on / off attack penalty and the trust decline penalty.
3. The trust evaluation method applied to the terminal device of the power Internet of Things according to claim 2, characterized in that: the on / off attack penalty and the trust decline penalty are determined in combination with the comparison result of the initial positive feedback value and the interaction importance index.
4. The trust evaluation method applied to the terminal device of the power Internet of Things according to claim 1, characterized in that: the recommendation importance index is determined according to the historical interaction record between the terminal or the user, the last interaction record, the weight of the historical interaction record and the weight of the last interaction record.
5. The trust evaluation method applied to the terminal device of the power Internet of Things according to claim 1, characterized in that: the attribute index of the subject and object behavior is determined according to the historical interaction record; the local variable weight vector is calculated according to the constant weight vector of each attribute index and the local state variable weight mode; the weight of the direct trust value is calculated according to the local state variable weight mode and the local variable weight vector. 6.The trust evaluation method applied to the power internet of things terminal device of claim 1, wherein, The method further comprises the following steps: constructing a trust value time sequence according to the evaluation result of the comprehensive trust value of the target consumer within a preset time period; if the trust value time sequence is a stationary sequence, calculating the autocorrelation coefficient and the partial autocorrelation coefficient of the trust value time sequence; establishing an autoregressive integrated moving average model according to the autocorrelation coefficient and the partial autocorrelation coefficient, wherein the order and the end of the autoregressive integrated moving average model are determined according to the autocorrelation coefficient and the partial autocorrelation coefficient; predicting the trust value of the target consumer through the autoregressive integrated moving average model to obtain a trust value prediction result. 7.A trust evaluation device applied to a power Internet of Things terminal device, characterized in that, The method comprises the following steps: The direct trust value calculation module is configured to calculate a direct trust value of the target consumer according to the interaction feedback of different consumers, the interaction importance index, the on / off attack penalty, and the trust decrease penalty; The recommendation interaction trust value calculation module is configured to obtain a direct trust evaluation set of each recommender for each consumer, and calculate a recommendation interaction trust value of each recommender for the target consumer, the recommendation importance index of the consumer being greater than the interaction importance index; The recommendation trust value calculation module is configured to calculate a recommendation trust value of the target consumer according to the recommendation interaction trust value of each recommender for the target consumer; The comprehensive trust value calculation module is configured to perform weighted fusion calculation on the direct trust value and the recommendation information value of the target consumer to obtain a comprehensive trust value of the target consumer; The recommendation interaction trust value of each recommender for the target consumer is calculated by: calculating the percentile of each direct trust evaluation set respectively; calculating the subjective elimination parameter of each recommender according to the percentile respectively; calculating the recommendation interaction trust value of each recommender for the target consumer according to the direct trust evaluation set and the subjective elimination parameter of each recommender.
8. A computer device, comprising: It comprises: at least one processor; and a memory connected in communication with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to perform the trust evaluation method for the power Internet of Things terminal device as claimed in any one of claims 1-6.
9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions for causing the computer to execute the trust evaluation method for the power Internet of Things terminal device as claimed in any one of claims 1-6.