A power multi-person cooperative operation authorization system and method based on a brain-computer interface

Abstract

The application relates to the technical field of brain-computer interfaces, in particular to a power multi-person collaborative operation authorization system and method based on a brain-computer interface. The method comprises the following steps: constructing a multi-person collaborative cognitive index system to obtain a multi-modal cognitive factor set, and constructing a value judgment matrix; a least square optimization model is established to obtain a subjective weight and perform consistency checking; objective information entropy weight is extracted in combination with historical data to obtain a comprehensive value weight and construct a fuzzy measure function; a comprehensive authorization priority evaluation value is calculated according to a real-time cognitive factor characteristic vector; disturbance analysis is carried out by using statistical distribution and Monte Carlo sampling to obtain an authorization return success rate and an authorization time delay confidence interval. The application can solve the problems of instruction transmission delay and cognitive ambiguity under strong electromagnetic interference, and through the combination of multi-person physiological cognition and edge computing power, the collaborative quality and authorization decision credibility in a high-risk power operation environment are improved.

Description

Technical Field

[0001] This invention relates to the field of brain-computer interface technology, specifically to a brain-computer interface-based multi-person collaborative operation authorization system and method for power systems. Background Technology

[0002] Power generation and maintenance are crucial supports for ensuring energy supply security and system stability. Their core tasks involve collaborative operations requiring multiple personnel, such as transformer hoisting, equipment phase verification, and switching operations. These operations are characterized by numerous steps, stringent timelines, and high safety risks, placing high demands on the cognitive consistency, coordination, and psychological stability of the participants. This provides a scientific data foundation and decision support for the safe operation and maintenance of power equipment, sustainable resource utilization, and risk early warning.

[0003] Current collaborative power operations primarily rely on verbal instructions, intercom devices, or hand signals in practice. Relevant information typically needs to be transmitted between team members via auditory, visual, or wireless links. Due to environmental factors such as strong electromagnetic interference, high noise levels on-site, and psychological stress on personnel, these communication links generally face challenges such as transmission delays, misunderstandings, and asynchronous responses. This is particularly true in live-line work and large equipment hoisting scenarios, where instructions exhibit deviations, out-of-order timing, information duplication, or distorted feedback during issuance, reception, understanding, and feedback. Secondly, existing technologies often employ process verification or single-person compliance mechanisms to ensure operational safety. However, these methods often fail to adequately consider the physiological and cognitive states of multiple personnel and edge computing capabilities. There is room for optimization in implementing real-time verification of cognitive consistency and ensuring collaborative reliability under complex operational constraints, which in turn affects the credibility and continuity of collaborative quality and its effectiveness in authorization decisions. In addition, most mainstream authorization schemes treat equipment status indiscriminately, lacking a graded assessment of the operator's cognitive status. This leads to delays in high-risk judgments and competition between routine operating procedures and limited resources, restricting the system's engineering applicability in extreme environments.

[0004] Compared to existing technologies, most current power collaborative operation methods are limited to the process compliance stage, and suffer from problems such as low efficiency of multi-person collaboration and insufficient reliability of status verification. There is an urgent need for a system solution that can combine brain-computer interfaces and status value assessment to conduct deeper hierarchical verification and authorization control collaboration, and ultimately achieve a balance between operational integrity, authorization efficiency and risk prevention in complex environments. Summary of the Invention

[0005] In view of this, the present invention proposes a brain-computer interface-based authorization method and system for multi-person collaborative operations in the power industry, which aims to solve the problems of insufficient reliability of collaborative quality and delayed authorization decision-making in high-risk operation environments caused by strong electromagnetic interference, on-site noise and psychological pressure, as well as the fact that the existing verification mechanism does not fully combine the physiological cognitive state of multiple people and edge computing power.

[0006] This invention proposes a brain-computer interface-based method for authorizing multi-person collaborative work in the power industry, comprising: Based on the physical spatial distribution and collaborative role responsibilities of power operation sites, a multi-person collaborative cognitive index system is constructed. A multimodal cognitive state factor set including EEG signals, near-infrared spectral signals and operation action characteristics is obtained. An interval value judgment matrix is ​​constructed based on the operation risk correlation attributes of each cognitive factor in the multimodal cognitive state factor set, and the interval value judgment matrix is ​​converted into a deterministic value matrix using the interval median. A least squares optimization model based on edge-side computational overhead constraints is established based on the deterministic value matrix to obtain the subjective value weights of each cognitive factor. The deterministic value matrix is ​​then dynamically validated based on the subjective value weights, and the deterministic value matrix is ​​iteratively adjusted based on the validation results. Historical operational environment disturbance data and baseline data on the physiological state of collaborating personnel are obtained. Objective information entropy weights are obtained based on historical operational environment disturbance data. Based on subjective value weights and objective information entropy weights according to a preset fusion coefficient, a comprehensive value weight of cognitive data is obtained. The comprehensive value weight is used as the fuzzy density, and the interaction coefficient between cognitive factors and operational risk environment is solved based on the fuzzy measure equation. A fuzzy measure function for the authorization priority of multi-person collaboration is constructed based on the interaction coefficient. Acquire real-time multimodal cognitive state factor data of the collaborative operation area, obtain cognitive factor feature vector at the time to be authorized based on the real-time multimodal cognitive state factor data, and obtain the comprehensive authorization priority evaluation value of the collaborative operation authorization package based on the fuzzy measure function and the cognitive factor feature vector. The statistical distribution between the comprehensive authorization priority assessment value and edge computing resources is obtained. Based on the statistical distribution and Monte Carlo sampling, the real-time multimodal cognitive state factor data and edge computing resources are subjected to multiple random perturbations and the hierarchical verification and authorization instruction transmission calculation are repeatedly performed to obtain the success rate probability distribution of collaborative operation authorization instruction return and the confidence interval of authorization delay.

[0007] Furthermore, based on the physical spatial distribution and collaborative role responsibilities at the power operation site, a multi-person collaborative cognitive index system is constructed to obtain a multimodal cognitive state factor set, including: Based on the process logic of power collaborative operations, a hierarchical model is performed to model the cognitive load, attentional orientation, and emotional stability involved in transformer hoisting, equipment phase verification, and switching operations. The system acquires EEG signal features, blood oxygen saturation changes, operation gesture sequences, and ambient noise sound pressure levels from the preset task configuration, and extracts the perceptual dimensions from the preset task configuration based on these features. Based on the psychological mechanism of power operation safety, a correlation analysis was conducted on the cognitive indicators of each dimension in the perception dimension, and the key cognitive indicators of each dimension in the perception dimension were extracted. Key cognitive indicators across various dimensions are subjected to feature reduction and standardization, and the set is determined as a multimodal cognitive state factor set.

[0008] Furthermore, when constructing an interval value judgment matrix based on the job risk correlation attributes of each cognitive factor in the multimodal cognitive state factor set, and converting this interval value judgment matrix into a deterministic value matrix using the interval median, the following steps are taken: Based on the interval number, the risk contribution of each cognitive factor in the multimodal cognitive state factor set is compared pairwise. The comparison results are determined as interval value elements, and an interval value judgment matrix is ​​constructed based on each interval value element. Based on the arithmetic mean relationship between the upper and lower bounds of the value elements in each interval, the median of the value elements in each interval is obtained. Replace each interval value element in the interval value judgment matrix with the interval midpoint of each interval value element, and determine the matrix after replacement as the deterministic value matrix.

[0009] Furthermore, based on the deterministic value matrix, a least-squares optimization model constrained by edge-side computational overhead is established to obtain the subjective value weights of each cognitive factor. When performing dynamic consistency verification of the deterministic value matrix based on these subjective value weights, the following steps are taken: Obtain the element deviation values ​​between each element and its corresponding weight ratio in the deterministic value matrix, and establish an objective function for the sum of squared errors based on the edge-side computational overhead based on the element deviation values. A least squares optimization model is established based on the objective function of sum of squared errors, non-negativity of weights, normalization of weights, and preset energy efficiency constraints of edge nodes. Based on the numerical iterative solution of the least squares optimization model, the subjective value weight vector that minimizes the judgment bias in the deterministic value matrix is ​​obtained. Based on the product relationship between the subjective value weight vector and the deterministic value matrix, the dynamic consistency index of the deterministic value matrix is ​​determined, and based on the quantitative relationship between the dynamic consistency index and the configured preset consistency threshold, it is determined whether the deterministic value matrix should be modified. When the dynamic consistency index is greater than or equal to the preset consistency threshold, the original parameters for maintaining the deterministic value matrix are determined. When the dynamic consistency index is less than the preset consistency threshold, the parameters of the deterministic value matrix are adjusted.

[0010] Furthermore, when the dynamic consistency index is less than the preset consistency threshold, the determination value matrix is ​​corrected, including: Obtain the ideal ratio matrix of the deterministic value matrix based on subjective value weights; Obtain the element-wise difference between each element of the deterministic value matrix and the corresponding ideal ratio in the ideal ratio matrix, map the element-wise difference to the corresponding feedback adjustment coefficient, and update the value of the element according to the feedback adjustment coefficient, wherein: The feedback adjustment coefficient is configured such that when the element difference is greater than the preset difference threshold, the feedback adjustment coefficient is less than 1; when the element difference is less than the preset difference threshold, the feedback adjustment coefficient is greater than 1; and when the element difference is equal to the preset difference threshold, the feedback adjustment coefficient is 1. The deterministic value matrix is ​​updated based on the updated elements, and the transpose elements at the corresponding positions in the deterministic value matrix are updated simultaneously. Based on the updated deterministic value matrix, a least squares optimization model is constructed for the second time, and subjective value weights are extracted again based on the constructed least squares optimization model. The updated deterministic value matrix is ​​then reviewed for consistency indicators based on the re-extracted subjective value weights until the dynamic consistency indicators reach the preset consistency threshold.

[0011] Furthermore, when acquiring historical operational environment disturbance data and baseline physiological state data of collaborating personnel, objective information entropy weights are obtained based on the historical operational environment disturbance data. Then, when obtaining the comprehensive value weight of cognitive data based on the subjective value weights and objective information entropy weights according to a preset fusion coefficient, the following is included: Statistical data on electromagnetic interference intensity, worker fatigue, and action execution deviation for each historical operation period were obtained, and the statistical data were normalized to construct a cognitive state standardization matrix. The probability distribution of each cognitive factor under different work environments is obtained based on the cognitive state standardization matrix, and the entropy value of each cognitive factor is obtained based on the information entropy formula. The objective information entropy weight of each cognitive factor is obtained based on the entropy value, and the subjective value weight and the objective information entropy weight are linearly weighted according to the preset weight fusion coefficient to obtain the initial weight vector. The weight fusion coefficient is configured to be dynamically set based on the danger level of the task and the remaining power of the edge node. The initial weight vector is processed using a normalization algorithm to obtain the comprehensive value weight of cognitive data.

[0012] Furthermore, based on the comprehensive value weight as the fuzzy density, and solving the interaction coefficient between the cognitive factor and the operational risk environment based on the fuzzy measure equation, when constructing the fuzzy measure function for multi-person collaborative authorization priority based on the interaction coefficient, it includes: The comprehensive value weight is used as the marginal fuzzy density of each cognitive factor to construct an initial fuzzy density vector; Based on the marginal fuzzy density of each cognitive factor, a fuzzy measure interaction equation is constructed regarding cognitive consistency, motor coordination, and psychological stress. By nonlinearly solving the fuzzy measure interaction equation, the interaction coefficients between each cognitive factor and the work risk environment are obtained. Based on the marginal fuzzy density and interaction coefficient, and according to the λ-fuzzy measure generation formula, the fuzzy measure value of any subset of cognitive factors is obtained, and the fuzzy measure function is constructed.

[0013] Furthermore, when acquiring real-time multimodal cognitive state factor data of the collaborative operation area, obtaining the cognitive factor feature vector at the time of authorization based on the real-time multimodal cognitive state factor data, and obtaining the comprehensive authorization priority evaluation value of the collaborative operation authorization package based on the fuzzy measure function and the cognitive factor feature vector, the process includes: Multi-level filtering, noise reduction, artifact removal, and missing data completion are performed on the real-time multimodal cognitive state factor data of the collaborative operation area to obtain a purified cognitive factor dataset. The purified cognitive factor dataset is normalized in terms of dimensions, and the normalized cognitive factors are combined according to the preset time sequence of the operation steps to form a real-time cognitive factor feature vector. Incremental contribution integral calculation is performed on the real-time cognitive factor feature vector based on the fuzzy measure function and the Choquet fuzzy integral calculation formula; The result of the integral calculation is determined as the comprehensive authorization priority evaluation value at the time of authorization.

[0014] Furthermore, when obtaining the statistical distribution between the overall authorized priority assessment value and edge computing resources, the following is included: Obtain the correlation sample pairs between the comprehensive authorization priority evaluation value and the job authorization success rate within the preset historical observation period, and align the execution time benchmark of the correlation sample pairs and remove outliers to construct a joint statistical matrix; Based on the kernel density estimation method, the joint probability density between the comprehensive authorization priority evaluation value and the edge available computing power in the joint statistical matrix is ​​estimated, and an authorization performance statistical model is established. Based on the authorized performance statistical model, the mutual information, covariance, and nonlinear correlation structural features between the comprehensive authorized priority evaluation value and computing resources are extracted. Based on the extracted features, a statistical distribution between the comprehensive authorization priority evaluation value and edge computing resources is constructed. Based on the constructed statistical distribution, a pseudo-random disturbance sequence that conforms to the dynamic environment characteristics of power operation is generated, and the pseudo-random disturbance sequence is superimposed on the real-time multimodal cognitive state factor data and edge computing resource parameters; A hierarchical verification operation is performed on the data after superimposed perturbation, which includes a first-level physiological feature legality verification and a second-level semantic consistency verification based on job logic. Authorization commands are mapped to different wireless transmission channels in descending order of comprehensive authorization priority evaluation value, and the feedback status of each simulated authorization is recorded. By repeatedly performing simulated authorization calculations a preset number of times, a sample set of authorization command return success rates is obtained, and the probability distribution function of the return success rate and the authorization delay confidence interval under a preset confidence level are calculated based on the sample set.

[0015] The beneficial effects of this invention are as follows: By utilizing the subjective value weights obtained from the interval value judgment matrix and the least squares optimization model, the collaborative experience and risk prediction logic of power operation experts can participate in the value definition of multi-person cognitive states in a structured and verifiable manner. Furthermore, dynamic consistency verification and feedback adjustment enhance the reliability of the weight construction under strong electromagnetic interference environments. Simultaneously, the objective information entropy weights extracted from historical environmental disturbances and physiological state data can accurately reflect the differences in information contribution of each cognitive factor under different operating conditions. This allows the weight allocation to be adaptively adjusted according to the real-time physiological fatigue of operators and environmental noise, constructing a comprehensive value weight system that considers both expert experience and objective operating environment characteristics. Secondly, by introducing fuzzy density and fuzzy measure functions to solve the interaction coefficients between cognitive factors and the operational risk environment, the limitations of traditional methods based on verbal instructions or single-process verification in expressing the coupling relationship of multi-person cognition are overcome. This effectively characterizes authorization needs under extreme conditions such as high psychological pressure and critical action execution, enabling authorization priority assessment to accurately reflect the urgency of collaborative operations under nonlinear environments and improving the decision-making credibility of the power collaborative operation system in complex environments. Finally, by constructing statistical distributions and performing Monte Carlo sampling to analyze the perturbations of real-time cognitive factors and computing resources, the probability distribution and confidence interval of the authorization success rate were obtained, thereby elevating operational assurance from passive compliance to proactive risk assessment. This mechanism enhances the robustness of the system operating at low-power edge devices, providing highly reliable physiological information support and decision-making basis for the safe operation and maintenance of power equipment.

[0016] On the other hand, this application also provides a brain-computer interface-based multi-person collaborative operation authorization system for power systems, including: The physiological perception module is used to construct a multi-person collaborative cognitive indicator system, obtain a set of multimodal cognitive state factors, generate a deterministic value matrix, and solve for subjective value weights. The collaborative evaluation module is connected to the physiological perception module. It is used to obtain objective information entropy weight based on historical data and combine it with subjective value weight to obtain the comprehensive value weight of cognitive data, thereby constructing a fuzzy measure function for the priority of multi-person collaborative authorization. The authorization control module is signal-connected to the collaborative evaluation module. It is used to obtain real-time cognitive factor feature vectors, calculate the comprehensive authorization priority evaluation value, and obtain the success rate probability distribution of authorization instruction feedback and the confidence interval of authorization delay based on statistical distribution and Monte Carlo sampling. The physiological sensing module includes a flexible EEG acquisition electrode array worn on the worker's head, a near-infrared blood oxygen monitoring unit set on the worker's wrist, and a motion capture camera set at the work site. The flexible EEG acquisition electrode array is connected to the signal amplification and conditioning circuit via shielded cables. The signal amplification and conditioning circuit is connected to the edge processing unit located in the operator's backpack terminal via the SPI (Serial Peripheral Interface) bus. The collaborative evaluation module is integrated into the embedded computing chip of the edge processing unit, which communicates bidirectionally with the authorized control terminal located in the substation control room via the power wireless private network.

[0017] It is understood that the brain-computer interface-based multi-person collaborative power operation authorization system and method in the above embodiments of this application have the same beneficial effects, and will not be described again. Attached Figure Description

[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart illustrating a brain-computer interface-based authorization method for multi-person collaborative work in the power industry, provided as an embodiment of the present invention; Figure 2 A hierarchical logical structure diagram of the multi-person collaborative cognitive index system provided in this embodiment of the invention; Figure 3 A flowchart for determining the subjective value weights of cognitive factors and dynamically resolving consistency, provided in an embodiment of the present invention; Figure 4 A schematic diagram of the authorization priority evaluation model architecture based on fuzzy measure and integral operation provided in an embodiment of the present invention; Figure 5 This is a functional block diagram of a brain-computer interface-based multi-person collaborative operation authorization system for power systems, provided as an embodiment of the present invention. Detailed Implementation

[0019] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] Goal Level: This represents the overall goal that the entire evaluation system ultimately aims to achieve. It is the core collaborative authorization decision result that all evaluation steps are directed towards.

[0021] Criteria Level: This level is used to structurally decompose the target level into the main evaluation dimensions that affect the power coordination authorization. It is an intermediate level that connects the overall authorization target with specific cognitive indicators.

[0022] Indicator Level: This layer is used to further refine the criteria layer, forming quantifiable and observable specific cognitive state factor indicators. It is the most basic and directly collectable EEG and physiological data layer in the assessment system.

[0023] Cognitive State Factor: refers to quantifiable physiological or behavioral characteristics that can affect the safety and consistency of collaborative power operations, and is used to reflect the potential cognitive load and psychological stability changes of operators.

[0024] Judgment Matrix: A matrix structure used to compare the relative importance of cognitive assessment elements in pairs. It reflects the weight relationship of different cognitive factors through the scoring results of experts on job risk.

[0025] Least Squares Optimization Model: This refers to a mathematical model that finds the optimal cognitive weight vector by minimizing the sum of squared deviations of the judgment matrix, and is used to obtain weight results that are most consistent with the risk judgment of power operations.

[0026] Objective Entropy Weight: This refers to the objective weight calculated based on the difference and information content of each cognitive indicator data in historical operation samples. It is used to reflect the magnitude of the indicator's contribution to the uncertainty of the collaborative system.

[0027] Fuzzy density: In fuzzy measurement theory, it refers to a parameter used to characterize the importance and interaction strength of a single cognitive factor or combination of factors in a specific work environment.

[0028] Fuzzy Measure Equation: A mathematical equation used to solve for the fuzzy measure values ​​of each cognitive factor and its combination based on the fuzzy density, in order to characterize the nonlinear coupling relationship between factors.

[0029] Cognitive Factor Data: refers to raw or processed numerical information that characterizes the current EEG, blood oxygen, and motor state of an operator, and is the input for calculating the overall authorization priority.

[0030] Confidence Interval: A numerical range derived from statistical inference, used to describe the range within which the authorized instruction return delay may fall at a given confidence level.

[0031] The mechanism of power operation safety psychology refers to the principle by which various cognitive factors affect the safety level of power operations through cognitive load, attention allocation, and motor coordination.

[0032] Interval Midpoint: The average of the two endpoints of a numerical interval, used to convert the result of an interval judgment into a deterministic value.

[0033] Dynamic Consistency Index: This index measures the degree of consistency in the changes of weights of a judgment matrix over multiple rounds of adjustments.

[0034] Relative Contribution: refers to the proportion of importance of a particular cognitive factor in the overall authorization assessment results.

[0035] Choquet Fuzzy Integral Formula: This formula calculates fuzzy integrals given a fuzzy measure. Below, on the cognitive factor vector The mathematical expression of the calculation formula for nonlinear aggregation is as follows: ; in, (1) (2) ... (n) The results are sorted from smallest to largest by cognitive factors; Indicates the sorted order of the first... A set consisting of one and subsequent factors; For set Fuzzy measure; (0) =0.

[0036] Sample Pair: In data analysis, a sample pair consists of two related authorization assessment values ​​and actual success rates.

[0037] Kernel Density Estimation: A nonparametric statistical method that estimates the true probability density of edge computing power and licensed performance by smoothing job samples using a kernel function.

[0038] Marginal distribution: refers to the probability distribution obtained when considering only one dimension of cognitive or computational power variable in a multidimensional random variable.

[0039] Joint distribution: refers to the probability distribution when two or more random variables take values ​​together in the same probability space.

[0040] Statistical Model: A model that uses mathematical structures to describe and infer the process of generating authorized data.

[0041] Operational Sequence Feature: refers to the logical flow attributes of power operations on the time axis, such as the sequence of operation steps and instruction intervals.

[0042] Fuzzy Integral Computation: refers to a method that uses fuzzy measures to perform nonlinear weighted aggregation of multiple cognitive factors.

[0043] Random perturbation refers to the introduction of random variations into the system input to simulate the uncertainty of environmental noise and electromagnetic interference.

[0044] Monte Carlo sampling: a method for estimating the probability distribution of authorization success rate through simulation with a large number of random samples and repeated calculations.

[0045] like Figures 1-4 As shown, this embodiment provides a brain-computer interface-based multi-person collaborative operation authorization method for power systems, including: Step S100: Based on the physical spatial distribution and collaborative role responsibilities of the power operation site, construct a multi-person collaborative cognitive index system, obtain a multimodal cognitive state factor set including EEG signals, near-infrared spectral signals and operation action characteristics, construct an interval value judgment matrix based on the operation risk correlation attributes of each cognitive factor in the multimodal cognitive state factor set, and convert the interval value judgment matrix into a deterministic value matrix using the interval median.

[0046] Specifically, when constructing a multi-person collaborative cognitive index system based on the physical spatial distribution and collaborative role responsibilities of power operation sites, the following steps are taken: hierarchical modeling of cognitive load, attentional focus, and emotional stability involved in transformer hoisting, equipment phase verification, and switching operations according to the procedural logic of power collaborative operations; acquisition of EEG signal characteristics, blood oxygen saturation changes, operation gesture sequences, and environmental noise sound pressure levels in the preset operation configuration, and extraction of the perception dimensions in the preset operation configuration based on EEG signal characteristics, blood oxygen saturation changes, operation gesture sequences, and environmental noise sound pressure levels; correlation analysis of each cognitive index in each dimension of the perception dimension based on the safety psychology mechanism of power operations, and extraction of key cognitive indicators in each dimension of the perception dimension; feature dimensionality reduction and standardization of the key cognitive indicators in each dimension, and determination of the set as a multimodal cognitive state factor set.

[0047] Specifically, when constructing the interval value judgment matrix based on the job risk correlation attributes of each cognitive factor in the multimodal cognitive state factor set, the process includes: comparing the job risk contribution of each cognitive factor in the multimodal cognitive state factor set pairwise based on the interval number, determining the comparison results as interval value elements, and constructing an interval value judgment matrix based on each interval value element; obtaining the interval median of each interval value element based on the arithmetic mean relationship between the upper and lower bounds of each interval value element; replacing each interval value element in the interval value judgment matrix with the interval median of each interval value element, and determining the matrix after replacement as the deterministic value matrix.

[0048] In this embodiment, the preset operation configuration is implemented in accordance with the "Electric Power Safety Work Regulations" and the risk level of the work site.

[0049] Understandably, by introducing a multi-person collaborative cognitive index system that matches the distribution of work space and role responsibilities, key safety psychological factors such as attention, cognitive load, and emotional stability, which are traditionally difficult to measure directly, are transformed into perceptible and calculable multimodal cognitive state factors. Furthermore, by using an interval value judgment matrix to perform interval-based modeling of the uncertainty of the contribution of different cognitive factors to operational risks, the assessment bias caused by fluctuations in the on-site environment and individual differences is effectively reflected. And by using interval median mapping, uncertain cognitive risks are transformed into a deterministic value matrix that can be used for decision-making and authorization control, thereby realizing the transformation from collaborative safety based on experience judgment to collaborative safety control driven by multimodal cognitive quantification.

[0050] Step S200: Establish a least squares optimization model based on edge-side computational overhead constraints based on the deterministic value matrix, obtain the subjective value weights of each cognitive factor, perform dynamic consistency verification on the deterministic value matrix based on the subjective value weights, and iteratively adjust the deterministic value matrix based on the verification results.

[0051] Specifically, when establishing a least squares optimization model based on edge-side computational overhead constraints using a deterministic value matrix, the process includes: obtaining the element deviation values ​​between each element in the deterministic value matrix and the corresponding weight ratios, and establishing an error sum of squares objective function based on the element deviation values; establishing a least squares optimization model based on the error sum of squares objective function, weight non-negativity constraints, weight normalization constraints, and preset edge node energy efficiency constraints; obtaining the subjective value weight vector that minimizes the judgment bias in the deterministic value matrix through numerical iterative solution of the least squares optimization model; determining the dynamic consistency index (DCI) of the deterministic value matrix based on the product relationship between the subjective value weight vector and the deterministic value matrix; and determining whether to modify the deterministic value matrix based on the quantitative relationship between the dynamic consistency index and the configured preset consistency threshold.

[0052] Specifically, when the Dynamic Consistency Index (DCI) is less than a preset consistency threshold, the deterministic value matrix is ​​corrected, including: obtaining the ideal ratio matrix of the deterministic value matrix based on the subjective value weights; obtaining the element difference between each element of the deterministic value matrix and the corresponding element's ideal ratio in the ideal ratio matrix, and mapping the corresponding feedback adjustment coefficient based on the element difference; updating the element's value based on the feedback adjustment coefficient, and simultaneously updating the transpose element at the corresponding position in the deterministic value matrix; and constructing a least squares optimization model based on the updated deterministic value matrix to extract the subjective value weights again.

[0053] Understandably, by introducing a least-squares optimization mechanism constrained by edge-side computing resources and energy efficiency on the basis of a deterministic value matrix, the solution process for cognitive factor weights is confined within a computational framework that can be deployed on-site and iterated in real time. On the one hand, this technology uses the least-squares model to quantitatively suppress subjective judgment bias, thereby obtaining cognitive factor weights that are more consistent with the actual contribution of operational risks. On the other hand, through dynamic consistency indicators and iterative correction mechanisms, it achieves continuous verification and adaptive adjustment of the inherent consistency of the judgment matrix, avoiding evaluation distortion caused by individual state fluctuations or environmental interference.

[0054] Step S300: Obtain historical work environment disturbance data and baseline data of physiological state of collaborating personnel; obtain objective information entropy weight based on historical work environment disturbance data; obtain comprehensive cognitive data value weight based on subjective value weight and objective information entropy weight according to preset fusion coefficient; use comprehensive value weight as fuzzy density, and solve the interaction coefficient between cognitive factors and work risk environment based on fuzzy measurement equation; construct fuzzy measurement function of multi-person collaborative authorization priority based on interaction coefficient.

[0055] Specifically, obtaining objective information entropy weights and acquiring comprehensive cognitive data value weights includes: acquiring statistical data on electromagnetic interference intensity, worker fatigue, and action execution deviations for each historical work period, and performing normalization processing on the statistical data to construct a cognitive state standardization matrix; obtaining the probability distribution of each cognitive factor under different work environments based on the cognitive state standardization matrix, and acquiring the entropy value of each cognitive factor based on the information entropy formula; acquiring the objective information entropy weights of each cognitive factor based on the entropy values, and linearly weighting the subjective value weights and objective information entropy weights according to a preset weight fusion coefficient to obtain an initial weight vector; and processing the initial weight vectors according to a normalization algorithm to obtain the comprehensive cognitive data value weights.

[0056] Specifically, when constructing the fuzzy measure function for multi-person collaborative authorization priority, the process includes: using the comprehensive value weight as the marginal fuzzy density of each cognitive factor to construct an initial fuzzy density vector; constructing fuzzy measure interaction equations for cognitive consistency, action coordination, and psychological stress based on the marginal fuzzy density of each cognitive factor; obtaining the interaction coefficients between each cognitive factor and the operational risk environment by nonlinearly solving the fuzzy measure interaction equations; and obtaining the fuzzy measure values ​​of any subset of cognitive factors based on the marginal fuzzy density and interaction coefficients according to the λ-fuzzy measure generation formula, thus constructing the fuzzy measure function.

[0057] Understandably, by introducing historical operational environment disturbance data and baseline data on personnel physiological states, environmental uncertainty and individual state fluctuations are objectively quantified in the form of information entropy weights and integrated with the aforementioned subjective value weights, thus forming a comprehensive value weight that simultaneously considers experiential cognition and data-driven characteristics. On this basis, fuzzy measurement theory is used to model the nonlinear interaction relationships between different cognitive factors and between cognitive factors and the operational risk environment, breaking through the limitations of traditional linear weighting in characterizing synergistic amplification or inhibition effects. Finally, by constructing a fuzzy measurement function for the priority of multi-person collaborative authorization, continuous assessment and hierarchical ranking of the overall risk level of collaborative operations can be achieved.

[0058] Step S400: Obtain real-time multimodal cognitive state factor data of the collaborative operation area; obtain the cognitive factor feature vector at the time to be authorized based on the real-time multimodal cognitive state factor data; and obtain the comprehensive authorization priority evaluation value of the collaborative operation authorization package based on the fuzzy measure function and the cognitive factor feature vector.

[0059] Specifically, when obtaining the comprehensive authorization priority evaluation value of the collaborative operation authorization package, the process includes: performing multi-level filtering and noise reduction, artifact removal, and missing data completion on the real-time multimodal cognitive state factor data of the collaborative operation area to obtain a purified cognitive factor dataset; performing dimensional normalization on the purified cognitive factor dataset, and combining the normalized cognitive factors according to the preset time sequence of the operation steps to form a real-time cognitive factor feature vector; performing incremental contribution integral calculation on the real-time cognitive factor feature vector based on the fuzzy measure function and the Choquet fuzzy integral calculation formula; and determining the integral calculation result as the comprehensive authorization priority evaluation value at the time of authorization.

[0060] Understandably, by purifying and temporally modeling the multimodal cognitive state factors collected in real time within the collaborative work area, the scattered and noisy on-site perception data is transformed into structured cognitive factor feature vectors. Furthermore, by utilizing Choquet fuzzy integrals based on fuzzy measure functions, the nonlinear collaborative contributions of each cognitive factor at different work stages are dynamically aggregated, thereby obtaining an authorization priority evaluation value that can comprehensively reflect the consistency of multi-person collaborative cognition, action coordination, and risk sensitivity, realizing the technological transformation from static rule-triggered authorization to real-time cognitive state-driven authorization.

[0061] Step S500: Obtain the statistical distribution between the comprehensive authorization priority evaluation value and edge computing resources, and based on the statistical distribution and Monte Carlo sampling, perform multiple random perturbations on the real-time multimodal cognitive state factor data and edge computing resources, and repeatedly perform hierarchical verification and authorization instruction transmission calculation to obtain the success rate probability distribution of collaborative operation authorization instruction return and the confidence interval of authorization delay.

[0062] Specifically, when obtaining the statistical distribution between the comprehensive authorization priority assessment value and edge computing resources, the process includes: obtaining the associated sample pairs of the comprehensive authorization priority assessment value and the job authorization success rate within a preset historical observation period, aligning the execution time benchmark of the associated sample pairs and removing outliers to construct a joint statistical matrix; estimating the joint probability density between the comprehensive authorization priority assessment value and the available edge computing power in the joint statistical matrix based on the kernel density estimation method to establish an authorization performance statistical model; and extracting the mutual information, covariance, and nonlinear correlation structural features between the comprehensive authorization priority assessment value and computing power resources based on the authorization performance statistical model to construct a statistical distribution.

[0063] Specifically, the random perturbation and hierarchical verification process includes: generating a pseudo-random perturbation sequence based on the constructed statistical distribution and superimposing it onto the real-time multimodal cognitive state factor data and edge computing resource parameters; performing hierarchical verification on the superimposed perturbation data, which includes a first-level physiological feature validity verification and a second-level semantic consistency verification based on job logic; mapping authorization instructions to different wireless transmission channels in descending order of comprehensive authorization priority evaluation value and recording the feedback status of each simulated authorization; and obtaining a sample set of authorization instruction return success rate by repeatedly executing simulated authorization calculations a preset number of times.

[0064] Understandably, by constructing a statistical correlation model between the comprehensive authorization priority assessment value and edge computing resources, and introducing a Monte Carlo random perturbation mechanism, the method performs probabilistic simulations of cognitive state fluctuations and computing power changes. This transforms the originally unpredictable authorization success rate and latency risks into assessable probability distributions and confidence intervals at the technical principle level. This method not only achieves a forward-looking quantitative assessment of the reliability of authorization command feedback, but also reflects the differentiated guarantee capabilities of different authorization priorities under conditions of limited communication and computing resources through hierarchical verification and multi-channel mapping mechanisms.

[0065] To enable those skilled in the art to fully understand and implement this invention, the specific implementation principle of this invention is further explained below in conjunction with a specific application scenario.

[0066] When establishing a multi-person collaborative cognitive indicator system and determining its comprehensive value weights, the hierarchical relationships of each level are clarified based on the framework of cognitive load, attention, and emotional stability criteria, combined with the psychological mechanisms of power operation safety. The goal layer measures the overall urgency of authorization for multi-person collaborative operations; the criterion layer connects the goal layer and the indicator layer, selecting the core dimensions that play a decisive role in operational risks; the indicator layer quantifies the physiological characteristics of each dimension of the criterion layer. After completing the system construction, the contribution of the indicators is quantified using an improved analytic hierarchy process.

[0067] In the cognitive assessment process, this invention uses interval numbers to construct a value judgment matrix to characterize the uncertainty of experts' judgments on the cognitive states of multiple individuals. The interval value judgment matrix can be expressed as: ; in, The lower bound of the interval represents the minimum risk contribution estimate; The upper bound of the interval represents the estimated value of the highest risk contribution.

[0068] Secondly, this invention uses the least squares method to optimize the final weight allocation, transforming the weight solution into a fitting programming problem: ; Constraints include weight normalization constraints. =1. Weight Positive Constraint ≥0 and edge node energy efficiency constraints ,in For the first The computational cost of each cognitive factor.

[0069] This invention designs a dynamic consistency verification mechanism, the formula of which is as follows: ; in, It is a static consistency indicator. The standard deviation is the weight. To adjust the coefficient. When If the value is less than 0.1, the test is passed; otherwise, the matrix parameters are corrected.

[0070] To overcome the limitations of subjective judgment, this invention employs the entropy weight method to correct subjective weights: ; in, This is the balance factor, usually taken as $0.6; Subjective value weight; $ represents the objective information entropy weight calculated based on historical operational electromagnetic interference and fatigue data.

[0071] In some implementations, to achieve nonlinear fusion of authorization priorities, this invention uses the comprehensive value weight as the fuzzy density to establish a fuzzy measurement system: ; The interaction between cognitive factors and the work environment is characterized by fuzzy measure equations: ; in, The interaction coefficient. When A value greater than 0 indicates a synergistic risk-enhancing effect between cognitive load and motor bias.

[0072] Based on the above measurement system, Choquet fuzzy integrals are used to achieve nonlinear fusion calculation. The standardized values ​​of real-time cognitive factors are arranged in ascending order. (1) (2) ... (n) The overall authorization priority R is represented as: ; Finally, Monte Carlo simulation was used to quantify the uncertainty. By repeatedly calculating the input cognitive data under 1000 random samples, the authorization success rate and its confidence interval were output: ; ; in, The number of simulations that pass the graded verification. This represents the confidence interval for the authorization delay.

[0073] In the above embodiments, by utilizing the subjective value weights obtained from the interval value judgment matrix and the least squares optimization model, the collaborative experience and risk prediction logic of power operation experts can participate in the value definition of multi-person cognitive states in a structured and verifiable manner. Dynamic consistency verification and feedback adjustment enhance the reliability of the weight construction in environments with strong electromagnetic interference. Simultaneously, the objective information entropy weights extracted from historical environmental disturbances and physiological state data can accurately reflect the differences in information contribution of each cognitive factor under different operating conditions. This allows the weight allocation to be adaptively adjusted according to the real-time physiological fatigue of operators and environmental noise, constructing a comprehensive value weight system that considers both expert experience and objective operating environment characteristics. Secondly, by introducing fuzzy density and fuzzy measure functions to solve the interaction coefficients between cognitive factors and the operational risk environment, the limitations of traditional methods based on verbal instructions or single-process verification in expressing the coupling relationship of multi-person cognition are overcome. This effectively characterizes authorization needs under extreme conditions such as high psychological pressure and critical action execution, enabling authorization priority assessment to accurately reflect the urgency of collaborative operations in nonlinear environments and improving the decision-making credibility of the power collaborative operation system in complex environments. Finally, by constructing statistical distributions and performing Monte Carlo sampling to analyze the perturbations of real-time cognitive factors and computing resources, the probability distribution and confidence interval of the authorization success rate were obtained, thereby elevating operational assurance from passive compliance to proactive risk assessment. This mechanism enhances the robustness of the system operating at low-power edge devices, providing highly reliable physiological information support and decision-making basis for the safe operation and maintenance of power equipment.

[0074] In another preferred embodiment based on the above embodiments, such as Figure 5 As shown, this embodiment provides a brain-computer interface-based multi-person collaborative operation authorization system for power systems, including: a physiological perception module, a collaborative evaluation module, and an authorization control module.

[0075] Specifically, the physiological sensing module includes a flexible EEG acquisition electrode array worn on the worker's head, a near-infrared blood oxygen monitoring unit located on the worker's wrist, and a motion capture camera located at the work site. The flexible EEG acquisition electrode array is connected to a signal amplification and conditioning circuit via a shielded cable, and the signal amplification and conditioning circuit is connected to an edge processing unit located in the worker's backpack terminal via an SPI bus.

[0076] Specifically, the collaborative evaluation module is integrated into the embedded computing chip of the edge processing unit to construct a multi-person collaborative cognitive index system, obtain a set of multimodal cognitive state factors, obtain objective information entropy weights based on historical data, combine subjective value weights to obtain comprehensive cognitive data value weights, and construct a fuzzy measurement function.

[0077] Specifically, the authorization control module and the collaborative evaluation module are signal-connected, and the edge processing unit communicates bidirectionally with the authorization control terminal located in the substation control room via the power wireless private network. The authorization control module acquires the real-time cognitive factor feature vector, calculates the comprehensive authorization priority evaluation value, and obtains the success rate probability distribution of authorization command feedback and the confidence interval of authorization delay based on Monte Carlo sampling.

[0078] To enable those skilled in the art to fully understand and implement this invention, the specific implementation principle of this invention will be further explained below in conjunction with the specific application scenario of substation transformer hoisting.

[0079] First, during the perception task at the power operation site, the commander and several operators wore flexible EEG acquisition electrode arrays and wrist-mounted near-infrared blood oxygen monitoring units. The flexible EEG acquisition electrode arrays captured EEG signals from the frontal and parietal lobes of the operators in real time, transmitting the weak electrophysiological signals through shielded cables to a signal amplification and conditioning circuit. The signal amplification and conditioning circuit performed bandpass filtering and power frequency notch filtering on the raw EEG signals, filtering out 50Hz environmental electromagnetic interference and acquiring purified EEG waveform data. Simultaneously, motion capture cameras positioned around the hoisting operation area continuously recorded the operators' hand gesture sequences. An edge processing unit synchronously received EEG, blood oxygen, and gesture data via an SPI bus and used feature extraction algorithms to calculate θ / …, reflecting cognitive load. Power spectral density ratio, rate of change of blood oxygen saturation, and trajectory deviation of the action execution.

[0080] Based on this, the edge processing unit calls the pre-stored interval scores from power experts to construct an interval value judgment matrix for each cognitive factor. The system extracts the arithmetic mean of the interval value elements as the interval median, converting the interval value judgment matrix into a deterministic value matrix. For this deterministic value matrix, the least squares optimization model uses numerical iteration to find the subjective value weight vector that minimizes the sum of squared judgment biases. Subsequently, the system calculates the dynamic consistency index (DCI) of the deterministic value matrix. If the DCI value exceeds the configured preset consistency threshold, the edge processing unit generates an ideal ratio matrix based on the subjective value weights and calculates the difference between the actual matrix elements and the ideal values, mapping out a feedback adjustment coefficient. The system uses this coefficient to numerically correct specific elements of the deterministic value matrix and synchronously updates its transpose elements until the DCI index meets the preset requirements, obtaining the final subjective value weights.

[0081] In some implementations, the collaborative evaluation module acquires electromagnetic interference intensity statistics and historical fatigue baseline data of workers in the substation's historical operating environment. The system normalizes this historical data, constructs a cognitive state standardization matrix, and calculates the objective information entropy weight of each cognitive factor using the information entropy formula. The system then fuses the subjective value weight and the objective information entropy weight using a preset linear weighting coefficient to generate a comprehensive cognitive data value weight. This comprehensive value weight serves as the marginal fuzzy density input to the fuzzy measure interaction equation. By nonlinearly solving this equation, the system obtains the interaction coefficient λ, reflecting the coupling effect of high cognitive load and high psychological pressure, and constructs a fuzzy measure function for multi-person collaborative authorization priority based on this coefficient.

[0082] Subsequently, the authorization control module receives the preprocessed cognitive factor feature vector in real time. The system uses the Choquet fuzzy integral calculation formula to sort the components of the real-time cognitive factor feature vector in ascending order of their values, and then performs incremental contribution integration using a fuzzy measure function to calculate the comprehensive authorization priority assessment value at the time of authorization. In this process, the Choquet integral quantifies the overall contribution of the consistency of cognitive states among different operators to the urgency of authorization through nonlinear aggregation.

[0083] Based on this, the authorization control module utilizes the kernel density estimation method to construct a joint probability density function of the comprehensive authorization priority assessment value and the edge available computing power based on historical authorization samples, forming a statistical model of authorization performance. The system initiates a Monte Carlo sampling procedure to generate a pseudo-random perturbation sequence that conforms to the characteristics of the current electromagnetic environment, and superimposes it onto the real-time multimodal cognitive state factor data. The system performs a two-level hierarchical verification on the superimposed perturbation data: the first level, physiological feature legitimacy verification, ensures the authenticity of personnel identity and state by verifying the matching degree between EEG artifacts and blood oxygen benchmarks; the second level, semantic consistency verification, ensures the correctness of operation semantics by comparing the consistency between the current action sequence and the hoisting procedure logic.

[0084] Finally, the authorization commands are mapped to different power wireless transmission channels in descending order of comprehensive authorization priority evaluation value. The system repeatedly performs simulated authorization calculations a preset number of times (e.g., 1000 times), recording the verification success status and feedback delay of each simulation. Ultimately, it statistically generates a probability distribution function for the success rate of authorization command return and calculates the authorization delay confidence interval at a 95% confidence level. This confidence interval data is transmitted back to the authorization control terminal in the substation control room as the reliability basis for the final authorization decision execution.

[0085] It is understood that the brain-computer interface-based multi-person collaborative operation authorization system and method for power systems described in the above embodiments have the same beneficial effects, and will not be repeated here.

[0086] Those skilled in the art will understand that this embodiment can be provided as a method, system, or computer program product. Therefore, this embodiment can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this embodiment can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0087] This embodiment is described with reference to flowchart illustrations and / or block diagrams of the method, apparatus (system), and computer program product according to this embodiment. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0088] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0089] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A brain-computer interface-based method for authorizing multi-person collaborative work in the power industry, characterized in that, include: Based on the physical spatial distribution and collaborative role responsibilities of power operation sites, a multi-person collaborative cognitive index system is constructed. A multimodal cognitive state factor set including EEG signals, near-infrared spectral signals and operation action characteristics is obtained. An interval value judgment matrix is ​​constructed based on the operation risk correlation attributes of each cognitive factor in the multimodal cognitive state factor set, and the interval value judgment matrix is ​​converted into a deterministic value matrix using the interval median. A least-squares optimization model based on the computational overhead constraint of the on-site backpack edge processing unit is established based on the deterministic value matrix. The subjective value weights of each cognitive factor are obtained, and the deterministic value matrix in the iterative update is dynamically verified based on the subjective value weights. The deterministic value matrix is ​​then iteratively adjusted based on the verification results. Historical operational environment disturbance data and baseline data on the physiological state of collaborating personnel are acquired. Objective information entropy weights are obtained based on the historical operational environment disturbance data. Based on the subjective value weights and objective information entropy weights, a fusion coefficient dynamically preset based on the operational risk level and the remaining power of the node is used to obtain the comprehensive value weight of cognitive data. The fusion coefficient ranges from [0,1] and is used to balance the proportion of experience weights and data weights. The comprehensive value weight is determined as a fuzzy density, and the interaction coefficient between cognitive factors and operational risk environment is solved based on the fuzzy measure equation. Based on the interaction coefficient, a fuzzy measure function is constructed to characterize the nonlinear coupling effect of cognitive factors and support the assessment of the priority of multi-person collaborative authorization. This function is used for the nonlinear aggregation calculation of the risk contribution of any subset of cognitive factors. Acquire real-time multimodal cognitive state factor data of the collaborative operation area, obtain cognitive factor feature vector at the time to be authorized based on the real-time multimodal cognitive state factor data, and obtain the comprehensive authorization priority evaluation value of the collaborative operation authorization package that carries the cognitive verification results and authorization decision of multiple people based on the fuzzy measure function and the cognitive factor feature vector. The statistical distribution between the comprehensive authorization priority assessment value and edge computing resources is obtained. Based on the statistical distribution and Monte Carlo sampling, the real-time multimodal cognitive state factor data and edge computing resources are subjected to multiple random perturbations and the hierarchical verification and authorization instruction transmission calculation are repeatedly performed to obtain the success rate probability distribution of collaborative operation authorization instruction return and the confidence interval of authorization delay.

2. The method for authorizing multi-person collaborative work in power systems based on a brain-computer interface as described in claim 1, characterized in that, When constructing a multi-person collaborative cognitive index system based on the physical spatial distribution and collaborative role responsibilities at power operation sites, it includes: Based on the process logic of power collaborative operations, a hierarchical model is performed to model the cognitive load, attentional orientation, and emotional stability involved in transformer hoisting, equipment phase verification, and switching operations. The system acquires EEG signal features, blood oxygen saturation changes, operation gesture sequences, and ambient noise sound pressure levels from the preset task configuration, and extracts the perceptual dimensions from the preset task configuration based on these features. Based on the psychological mechanism of power operation safety, a correlation analysis was conducted on the cognitive indicators of each dimension in the perception dimension, and the key cognitive indicators of each dimension in the perception dimension were extracted. Key cognitive indicators across various dimensions are subjected to feature reduction and standardization, and the set is determined as a multimodal cognitive state factor set.

3. The method for authorizing multi-person collaborative work in power systems based on a brain-computer interface as described in claim 2, characterized in that, When constructing an interval value judgment matrix based on the job risk correlation attributes of each cognitive factor in the multimodal cognitive state factor set, the following are included: Based on the interval number, the risk contribution of each cognitive factor in the multimodal cognitive state factor set is compared pairwise. The comparison results are determined as interval value elements, and an interval value judgment matrix is ​​constructed based on each interval value element. Based on the arithmetic mean relationship between the upper and lower bounds of the value elements in each interval, the median of the value elements in each interval is obtained. Replace each interval value element in the interval value judgment matrix with the interval midpoint of each interval value element, and determine the matrix after replacement as the deterministic value matrix.

4. The method for authorizing multi-person collaborative work in power systems based on a brain-computer interface as described in claim 1, characterized in that, When establishing a least-squares optimization model based on edge-side computational overhead constraints using the deterministic value matrix, the following are included: Obtain the element deviation values ​​between each element and its corresponding weight ratio in the deterministic value matrix, and establish an objective function for the sum of squared errors based on the edge-side computational overhead based on the element deviation values. Based on the objective function of sum of squared errors, non-negativity of weights, weight normalization constraints, and preset edge node energy efficiency constraints, a least squares optimization model is established. The edge node energy efficiency constraints are an extension and supplement to the edge-side computational overhead constraints: the edge-side computational overhead constraints limit the computational complexity and execution latency of the least squares optimization process, ensuring that the cognitive weight solution process can be completed in real-time under the computing power limitations of the edge nodes; the edge node energy efficiency constraints further restrict the computation process from an energy consumption perspective, ensuring the battery life of the backpack edge terminal during long-term operation by limiting the energy consumption upper limit of unit weight iteration. Together, these two constraints constitute a two-layer constraint system in an edge-side resource-constrained environment, collaboratively achieving a balance between high efficiency and low energy consumption in edge-side cognitive weight solution. Based on the numerical iterative solution of the least squares optimization model, the subjective value weight vector that minimizes the judgment bias in the deterministic value matrix is ​​obtained. Based on the product relationship between the subjective value weight vector and the deterministic value matrix, a dynamic consistency index for the deterministic value matrix is ​​determined. Then, based on the quantitative relationship between the dynamic consistency index and the configured preset consistency threshold, it is determined whether the deterministic value matrix should be modified.

5. The method for authorizing multi-person collaborative work in power systems based on a brain-computer interface as described in claim 4, characterized in that, When determining whether to modify the deterministic value matrix based on the quantitative relationship between the dynamic consistency index and the configured preset consistency threshold, the following should be included: Obtain the ideal ratio matrix of the deterministic value matrix based on subjective value weights; Obtain the element difference between each element of the deterministic value matrix and the corresponding ideal ratio in the ideal ratio matrix, and map the corresponding feedback adjustment coefficient based on the element difference; The element is numerically updated based on the feedback adjustment coefficient, and the transpose element at the corresponding position in the deterministic value matrix is ​​updated simultaneously. Based on the updated deterministic value matrix, a least squares optimization model is constructed for the second time, and the subjective value weights are extracted again.

6. The method for authorizing multi-person collaborative work in power systems based on a brain-computer interface as described in claim 1, characterized in that, When acquiring objective information entropy weights and obtaining the comprehensive value weights of cognitive data, the following are included: Statistical data on electromagnetic interference intensity, worker fatigue, and action execution deviation for each historical operation period were obtained, and the statistical data were normalized to construct a cognitive state standardization matrix. The probability distribution of each cognitive factor under different work environments is obtained based on the cognitive state standardization matrix, and the entropy value of each cognitive factor is obtained based on the information entropy formula. The objective information entropy weights of each cognitive factor are obtained based on the entropy value, and the subjective value weights and objective information entropy weights are linearly weighted according to the preset weight fusion coefficient to obtain the initial weight vector. The initial weight vector is processed using a normalization algorithm to obtain the comprehensive value weight of cognitive data.

7. The method for authorizing multi-person collaborative work in power systems based on a brain-computer interface as described in claim 6, characterized in that, When constructing a fuzzy measure function for multi-user collaborative authorization priority, the following is included: The comprehensive value weight is used as the marginal fuzzy density of each cognitive factor to construct an initial fuzzy density vector; Based on the marginal fuzzy density of each cognitive factor, a fuzzy measure interaction equation is constructed regarding cognitive consistency, motor coordination, and psychological stress. By nonlinearly solving the fuzzy measure interaction equation, the interaction coefficients between each cognitive factor and the work risk environment are obtained. Based on the marginal fuzzy density and interaction coefficient, and according to the λ-fuzzy measure generation formula, the fuzzy measure value of any subset of cognitive factors is obtained, and the fuzzy measure function is constructed.

8. The method for authorizing multi-person collaborative work in power systems based on a brain-computer interface as described in claim 7, characterized in that, When obtaining the statistical distribution between the overall authorization priority assessment value and edge computing resources, the following is included: Obtain the correlation sample pairs between the comprehensive authorization priority evaluation value and the job authorization success rate within the preset historical observation period, and align the execution time benchmark of the correlation sample pairs and remove outliers to construct a joint statistical matrix; Based on the kernel density estimation method, the joint probability density between the comprehensive authorization priority evaluation value and the edge available computing power in the joint statistical matrix is ​​estimated, and an authorization performance statistical model is established. Based on the authorized performance statistical model, the mutual information, covariance, and nonlinear correlation structural features between the comprehensive authorized priority evaluation value and computing resources are extracted to construct a statistical distribution.

9. The method for authorizing multi-person collaborative work in power systems based on a brain-computer interface as described in claim 8, characterized in that, When performing random perturbation and hierarchical verification, the following are included: A pseudo-random perturbation sequence is generated based on the constructed statistical distribution and superimposed onto the real-time multimodal cognitive state factor data and edge computing resource parameters; A hierarchical verification operation is performed on the data after superimposed perturbation. The hierarchical verification includes a first-level physiological feature legality verification and a second-level semantic consistency verification based on job logic. Authorization commands are mapped to different wireless transmission channels in descending order of comprehensive authorization priority evaluation value, and the feedback status of each simulated authorization is recorded. A sample set of authorization command return success rates is obtained by repeatedly performing simulated authorization calculations a preset number of times.

10. A brain-computer interface-based multi-person collaborative operation authorization system for power systems, employing a brain-computer interface-based multi-person collaborative operation authorization method for power systems as described in any one of claims 1-9, characterized in that, include: The physiological perception module is used to acquire a set of multimodal cognitive state factors, including electroencephalogram (EEG) signals, near-infrared spectral signals, and operational action features, and to generate a deterministic value matrix. The physiological sensing module includes a flexible EEG acquisition electrode array, a near-infrared blood oxygen monitoring unit, and a motion capture camera. The collaborative evaluation module, connected to the physiological perception module, is used to solve the subjective value weight vector based on the least squares optimization model and construct a fuzzy measurement function by combining the objective information entropy weight; the collaborative evaluation module is integrated into the edge processing unit within the backpack terminal; The authorization control module is signal-connected to the collaborative evaluation module and is used to calculate the comprehensive authorization priority evaluation value, and obtain the success rate probability distribution of authorization command return and the confidence interval of authorization delay based on Monte Carlo sampling. The authorized control module communicates bidirectionally with the authorized control terminal located in the control room via a dedicated power wireless network.

Images