Power grid dispatcher cognitive load dynamic adjusting system and method based on multi-mode biological feedback

Abstract

The invention discloses a power grid dispatcher cognitive load dynamic adjustment system and method based on multi-modal biofeedback, and the method comprises the steps: carrying out the real-time quantification of the cognitive load of a power grid dispatcher through fusing the multi-modal biofeedback signals, such as electroencephalogram, eye movement and galvanic skin; and a human-computer interface information presentation mode and a task allocation strategy are dynamically adjusted according to the load level. The system specifically comprises a biological signal acquisition module, a cognitive load calculation engine module, a dynamic adjustment execution module and a closed-loop feedback adjustment module. According to the method, the problems of information overload, poor individual adaptability, high misoperation risk and the like in a traditional power grid dispatching system can be effectively solved, the efficiency and safety of dispatchers in complex fault handling are improved, and the time sensitivity requirement of power control is met.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric power dispatching control, and in particular relates to a system and method for dynamically adjusting the cognitive load of power grid dispatchers based on multimodal biofeedback. Background Art

[0002] The current power grid dispatching system primarily relies on fixed interfaces to display massive amounts of real-time data. Dispatchers must make rapid decisions under high-pressure conditions, which can easily lead to cognitive fatigue and operational errors due to information overload. Traditional solutions, such as alarm filtering and emergency plan push, are static rules that cannot dynamically adapt to individual cognitive states. Existing biofeedback technologies (such as brainwave monitoring) are mostly used in the medical or sports training fields and have not yet been deeply integrated with power operations. At the same time, conventional human-computer interaction systems lack the ability to perceive the dispatcher's physiological state in real time, making it difficult to adjust information presentation methods or task allocation strategies in a timely manner. In addition, existing cognitive load assessment methods often rely on subjective questionnaires or single-modal signals (such as heart rate), which lack accuracy. Therefore, there is an urgent need for a dynamic adjustment system that integrates multimodal biofeedback with power grid business characteristics to scientifically quantify cognitive load and implement adaptive decision-making assistance to improve dispatching safety and efficiency. Summary of the Invention

[0003] The present invention proposes a system and method for dynamically adjusting the cognitive load of power grid dispatchers based on multimodal biofeedback to solve the problems existing in the above-mentioned prior art.

[0004] To achieve the above objectives, the present invention provides a system for dynamically adjusting cognitive load of power grid dispatchers based on multimodal biofeedback, comprising:

[0005] Bio-signal acquisition module, used to collect the dispatcher's EEG signals, eye movement signals and skin conduction signals in real time;

[0006] a cognitive load calculation engine module, configured to calculate a cognitive load index based on the EEG signal, the eye movement signal, and the electrocutaneous signal using a multimodal feature fusion model;

[0007] A dynamic adjustment execution module, configured to dynamically adjust the information presentation mode and task proxy mechanism of the power grid dispatching interface according to the cognitive load index;

[0008] Closed-loop feedback regulation module is used to synchronize power grid events with biological signals and optimize regulation strategies in real time.

[0009] Optionally, the biological signal acquisition module includes:

[0010] A prefrontal EEG sensor set is used to collect EEG energy in the delta, theta, and beta frequency bands through a dry electrode array;

[0011] An eye tracking sensor group is used to calculate the eye movement entropy value by combining pupil diameter changes and gaze point trajectory;

[0012] The galvanic skin response sensor set is used to measure the rate of change of skin conductance using wristband electrodes.

[0013] Optionally, the cognitive load calculation engine module includes:

[0014] A multimodal feature fusion model module is used to calculate the cognitive load index based on the EEG energy ratio, eye movement entropy, and galvanic skin response gradient;

[0015] The load level dynamic division module divides the cognitive load level according to the cognitive load index.

[0016] Optionally, the expression of the multimodal feature fusion model module is:

[0017]

[0018] Where, CLI t represents the cognitive load index, Indicates the brainwave energy ratio between the theta band and the beta band, represents the rate of change of skin conductance, α, λ, γ are weight coefficients, H eye represents the eye movement entropy value.

[0019] Optionally, the dynamic adjustment execution module includes:

[0020] an information presentation optimization unit for dynamically simplifying the grid topology map and aggregating telemetry data based on cognitive load levels;

[0021] The task agent unit is used to recommend treatment plans or trigger the voice command system based on the cognitive load level and the urgency of the power grid event.

[0022] Optionally, the adjustment method of the information presentation hope includes:

[0023] Keep N-1 key connections and reduce visual elements;

[0024] Clustering associated parameters into fan-shaped blocks reduces information search time.

[0025] Optionally, the task agent unit includes:

[0026] Plan recommendation engine, used to output treatment plans based on the current fault type and cognitive load level;

[0027] Voice command system, used for fuzzy matching and secondary confirmation of key operations.

[0028] Optionally, the closed-loop feedback regulation module includes:

[0029] a data synchronization unit to align the time scales of biosignals and power grid events;

[0030] A load evaluation unit, used to update the cognitive load value every 200ms;

[0031] a strategy matching unit for selecting a regulation strategy based on the cognitive load level and the urgency of the power grid event;

[0032] Effect feedback unit, used to optimize weight coefficients through Q-learning.

[0033] The present invention also proposes a method for dynamically adjusting cognitive load of power grid dispatchers based on multimodal biofeedback, comprising the following steps:

[0034] Collect dispatchers' EEG signals, eye movement signals and skin conduction signals in real time;

[0035] Calculating a cognitive load index using a multimodal feature fusion model based on the EEG signal, the eye movement signal, and the electrocutaneous signal;

[0036] Dynamically adjust the information presentation mode and task agent mechanism of the power grid dispatching interface according to the cognitive load index;

[0037] Synchronize power grid events and biological signals to optimize regulation strategies in real time.

[0038] Optionally, the dynamically adjusting the information presentation mode and task proxy mechanism of the power grid dispatching interface according to the cognitive load index includes:

[0039] Cognitive load grading based on cognitive load index;

[0040] When the cognitive load level is level I, the human-computer interface is kept normally displayed and the task proxy mechanism is not triggered;

[0041] When the cognitive load level is level II, the topology diagram of the human-computer interface is dynamically simplified and the telemetry data is radially aggregated to trigger the emergency plan recommendation engine;

[0042] When the cognitive load level is level III, the human-machine interface is further simplified, the voice command system module is triggered, and whether to enter the fully automatic hosting mode is decided based on the urgency of the power grid event.

[0043] Compared with the prior art, the present invention has the following advantages and technical effects:

[0044] The present invention's system and method for dynamically regulating cognitive load for power grid dispatchers based on multimodal biofeedback accurately quantifies dispatchers' cognitive load by collecting multimodal biofeedback signals, including the EEG theta / beta wave energy ratio, eye movement entropy, and galvanic skin response gradient, in real time. This addresses the lag and inaccuracy of traditional subjective questionnaire assessments, often associated with single-modal signals. The system dynamically adjusts the human-machine interface information presentation based on cognitive load levels, such as by dynamically simplifying topological diagrams and radially aggregating telemetry data. This reduces visual element and information search time, effectively alleviating the information overload and lack of personalized adaptation inherent in fixed interfaces. Furthermore, the task proxy mechanism module, through a plan recommendation engine and voice command system, reduces decision-making cognitive load and mitigates the risk of operational errors. The closed-loop feedback regulation module achieves time-scale alignment between biosignals and power grid events and a coupled decision tree for power grid events, ensuring that the system completes biofeedback responses and regulatory actions within milliseconds, meeting the time-sensitive requirements of power control and improving the safety and efficiency of power grid dispatch. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings:

[0046] Figure 1 Schematic diagram of the system structure of an embodiment of the present invention;

[0047] Figure 2 Schematic diagram of policy matching according to an embodiment of the present invention. DETAILED DESCRIPTION

[0048] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0049] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0050] Example 1

[0051] like Figure 1 As shown, this embodiment provides a system for dynamically adjusting cognitive load of power grid dispatchers based on multimodal biofeedback, including:

[0052] Bio-signal acquisition module, used to collect the dispatcher's EEG signals, eye movement signals and skin conduction signals in real time;

[0053] The cognitive load calculation engine module is used to calculate the cognitive load index based on EEG signals, eye movement signals, and electrocutaneous signals through a multimodal feature fusion model;

[0054] Dynamic adjustment execution module, used to dynamically adjust the information presentation mode and task proxy mechanism of the power grid dispatch interface according to the cognitive load index;

[0055] Closed-loop feedback regulation module is used to synchronize power grid events with biological signals and optimize regulation strategies in real time.

[0056] Furthermore, the bio-signal acquisition module is divided into a frontal lobe EEG sensor group, an eye tracking sensor group, and a skin galvanic response sensor group. The frontal lobe EEG sensor group uses a dry electrode array to collect energy in the δ (1-4 Hz), θ (4-8 Hz), and β (12-30 Hz) frequency bands in real time. The wave energy ratio quantifies the working memory load (neuroscience principle: increased theta waves reflect cognitive resource tension). The eye tracking sensor group uses a 120Hz sampling infrared camera to calculate the entropy value H by combining the pupil diameter change and the gaze point trajectory. eye , can identify the degree of distraction (cognitive principle: entropy value > 0.6 indicates decreased visual search efficiency). The galvanic skin response sensor group uses wristband electrodes to measure the rate of change of skin conductance It can help verify the increase in stress levels (physiological principle: sympathetic nerve excitement leads to sweat gland secretion).

[0057] Furthermore, the cognitive load calculation engine module is divided into a multimodal feature fusion model module and a load level dynamic division module. The multimodal feature fusion model module is established using an LSTM neural network:

[0058]

[0059] Where: (CLI t : cognitive load index, α / λ / γ are weight coefficients, H eye is the eye movement entropy value)

[0060] The weight coefficient is calibrated through 500 sets of scheduling scenario experiments, which can solve the problem of single signal being susceptible to interference and improve the accuracy of load assessment.

[0061] The classification standards of the load level dynamic classification module are as follows:

[0062]

[0063] m is the statistical inflection point of the sudden change in the probability of erroneous operation, and n is the critical point of cognitive failure.

[0064] Theta / beta wave energy ratio threshold (core indicator):

[0065] 0.8 critical point: EEG research at the University of Wisconsin confirmed that when the theta / beta wave energy ratio is ≥0.8, working memory efficiency decreases by 32%.

[0066] 30% mutation threshold: A sudden increase in theta wave energy >30% is a precursor to cognitive collapse.

[0067] Eye movement entropy calibration:

[0068] 0.5 cutoff value: MIT eye movement experiments show that when the entropy value is ≥ 0.5, the confusion of the visual search path increases significantly.

[0069] The dynamic adjustment execution module is divided into the information presentation module and the task agent mechanism module.

[0070] Information presentation module: 1. Dynamic simplification algorithm for topology maps: Retains N-1 key connections (meaning disconnecting irrelevant connection lines without affecting the normal operation of the main system), which can reduce visual elements by more than 50% (compared to traditional full topology maps); 2. Radial aggregation of telemetry data: Clusters related parameters (such as line current and temperature) into sector-shaped blocks, which can reduce information search time by 40%.

[0071] The task proxy mechanism module reduces the cognitive load of decision-making: 1. Plan Recommendation Engine: This module inputs the current fault type (SCADA event) and CLI level, and outputs the top three solutions. (The solutions come from the library module, which in turn draws on a library of historical cases.) 2. Voice Command System: This module supports fuzzy matching (e.g., "Adjust the load on main transformer #3" can be identified as "Adjust the load factor of 110kV main transformer #3") and supports a mechanism to prevent errors: key operations require secondary confirmation (e.g., a voice response of "Confirm load shedding").

[0072] The closed-loop feedback regulation module includes:

[0073] a data synchronization unit to align the time scales of biosignals and power grid events;

[0074] A load evaluation unit, used to update the cognitive load value every 200ms;

[0075] The strategy matching unit is used to select the regulation strategy according to the cognitive load level and the urgency of the power grid event. The regulation strategy is as follows: Figure 2 As shown;

[0076] The effect feedback unit is used to recalculate the CLI after adjustment and optimize the weight coefficient through Q-learning.

[0077] This embodiment also provides a method for adjusting a power grid dispatcher's cognitive load dynamic adjustment system based on multimodal biofeedback, comprising the following steps:

[0078] Collect dispatchers' EEG signals, eye movement signals and skin conduction signals in real time;

[0079] Calculating a cognitive load index using a multimodal feature fusion model based on the EEG signal, the eye movement signal, and the electrocutaneous signal;

[0080] Dynamically adjust the information presentation mode and task agent mechanism of the power grid dispatching interface according to the cognitive load index;

[0081] Synchronize power grid events and biological signals to optimize regulation strategies in real time.

[0082] Dynamically adjusting the information presentation mode and task agent mechanism of the power grid dispatching interface according to the cognitive load index includes:

[0083] Cognitive load grading based on cognitive load index;

[0084] When the cognitive load level is level I, the human-computer interface is kept normally displayed and the task proxy mechanism is not triggered;

[0085] When the cognitive load level is level II, the topology diagram of the human-computer interface is dynamically simplified and the telemetry data is radially aggregated to trigger the emergency plan recommendation engine;

[0086] When the cognitive load level is level III, the human-machine interface is further simplified, the voice command system module is triggered, and whether to enter the fully automatic hosting mode is decided based on the urgency of the power grid event.

[0087] The above are merely preferred embodiments of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A dynamic adjustment system for cognitive load of power grid dispatchers based on multimodal biofeedback, characterized by: include: Bio-signal acquisition module, used to collect the dispatcher's EEG signals, eye movement signals and skin conduction signals in real time; a cognitive load calculation engine module, configured to calculate a cognitive load index based on the EEG signal, the eye movement signal, and the electrocutaneous signal using a multimodal feature fusion model; A dynamic adjustment execution module, configured to dynamically adjust the information presentation mode and task proxy mechanism of the power grid dispatching interface according to the cognitive load index; Closed-loop feedback regulation module is used to synchronize power grid events with biological signals and optimize regulation strategies in real time.

2. The system according to claim 1, wherein: The biological signal acquisition module includes: A prefrontal EEG sensor set is used to collect EEG energy in the delta, theta, and beta frequency bands through a dry electrode array; An eye tracking sensor group is used to calculate the eye movement entropy value by combining pupil diameter changes and gaze point trajectory; The galvanic skin response sensor set is used to measure the rate of change of skin conductance using wristband electrodes.

3. The system according to claim 1, wherein: The cognitive load calculation engine module includes: A multimodal feature fusion model module is used to calculate the cognitive load index based on the EEG energy ratio, eye movement entropy, and galvanic skin response gradient; The load level dynamic division module divides the cognitive load level according to the cognitive load index.

4. The system according to claim 3, characterized in that The expression of the multimodal feature fusion model module is: Where, CLI t represents the cognitive load index, Indicates the brainwave energy ratio between the theta band and the beta band, represents the rate of change of skin conductance, α, λ, γ are weight coefficients, H eye represents the eye movement entropy value.

5. The system according to claim 1, wherein: The dynamic adjustment execution module includes: an information presentation optimization unit for dynamically simplifying the grid topology map and aggregating telemetry data based on cognitive load levels; The task agent unit is used to recommend treatment plans or trigger the voice command system based on the cognitive load level and the urgency of the power grid event.

6. The system according to claim 5, characterized in that The adjustment methods of the information presentation hope include: Keep N-1 key connections and reduce visual elements; Clustering associated parameters into fan-shaped blocks reduces information search time.

7. The system according to claim 5, characterized in that The task agent unit includes: Plan recommendation engine, used to output treatment plans based on the current fault type and cognitive load level; Voice command system, used for fuzzy matching and secondary confirmation of key operations.

8. The system according to claim 1, wherein: The closed-loop feedback regulation module includes: a data synchronization unit to align the time scales of biosignals and power grid events; A load evaluation unit, used to update the cognitive load value every 200ms; a strategy matching unit for selecting a regulation strategy based on the cognitive load level and the urgency of the power grid event; Effect feedback unit, used to optimize weight coefficients through Q-learning.

9. A method for dynamically adjusting cognitive load of power grid dispatchers based on multimodal biofeedback, characterized in that: The following steps are involved: Collect dispatchers' EEG signals, eye movement signals and skin conduction signals in real time; Calculating a cognitive load index using a multimodal feature fusion model based on the EEG signal, the eye movement signal, and the electrocutaneous signal; Dynamically adjust the information presentation mode and task agent mechanism of the power grid dispatching interface according to the cognitive load index; Synchronize power grid events and biological signals to optimize regulation strategies in real time.

10. The method according to claim 9, characterized in that The method of dynamically adjusting the information presentation mode and task proxy mechanism of the power grid dispatching interface according to the cognitive load index includes: Cognitive load grading based on cognitive load index; When the cognitive load level is level I, the human-computer interface is kept normally displayed and the task proxy mechanism is not triggered; When the cognitive load level is level II, the topology diagram of the human-computer interface is dynamically simplified and the telemetry data is radially aggregated to trigger the emergency plan recommendation engine; When the cognitive load level is level III, the human-machine interface is further simplified, the voice command system module is triggered, and whether to enter the fully automatic hosting mode is decided based on the urgency of the power grid event.

Images