Power system cloud platform man-machine interaction method, system, device and medium

Abstract

The invention relates to the technical field of cloud platform man-machine interaction, and discloses a power system cloud platform man-machine interaction method, system and device and a medium, and the method comprises the steps: building a first interaction demand evaluation model and a second improved interaction protocol decision model, and considering the personalized demands of a user and environmental factors; the first interaction demand evaluation model can accurately analyze the interaction demands of the target user in different scenes and environments, and the second improved interaction protocol decision model optimizes and adjusts the demands on the basis, so that the dynamic optimization of the interaction protocol is realized. In the interaction process, a preset upper-layer strategy function can select an appropriate rule from preset rules according to the equipment health state and the environmental factors, and an interaction protocol with the optimal efficiency-reliability ratio in the current state is obtained. And a specific interaction configuration instruction is sent to corresponding execution equipment or personnel through a lower execution function, so that interaction configuration operation is implemented, and smooth implementation of an interaction process is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cloud platform human-computer interaction, and particularly relates to a power system cloud platform human-computer interaction method, system, device and medium. BACKGROUND

[0002] In the actual operation of the power system cloud platform, the existing human-computer interaction technology has some significant problems. On the one hand, the traditional interaction mode often lacks accurate grasp of the personalized needs of users. Different power system operators have great differences in skill level, operation habits and specific scenarios, and the existing interaction system is difficult to provide customized interaction experience according to these differences, which may cause some users to feel inconvenient in operation and reduce work efficiency.

[0003] On the other hand, the existing interaction system does not fully consider environmental factors. The running environment of the power system is complex and changeable, and environmental conditions such as temperature, humidity and illumination will affect the vision, touch and other perceptions of the operator, and then affect the effect of human-computer interaction. However, the existing technology has not been able to well incorporate these environmental factors into the interaction design, so that the accuracy and reliability of the interaction are greatly reduced in some special environments.

[0004] In addition, in the optimization of the interaction protocol, the existing method is single and fixed, and lacks the ability of dynamic adjustment. With the continuous development and change of the power system, the interaction needs of users are also evolving, and the existing interaction protocol is difficult to quickly adapt to these changes and cannot provide the optimal interaction scheme in time. Moreover, the feedback and iterative updating mechanism of data in the interaction process is not perfect, which causes the system to be unable to timely optimize and improve itself according to the actual interaction effect, which is not conducive to improving the quality and performance of human-computer interaction in the long run.

[0005] In summary, it is of great practical significance and application value to develop a more intelligent, flexible and adaptive power system cloud platform human-computer interaction method that can adapt to complex environments and personalized needs. SUMMARY

[0006] In view of the above existing problems, the present application is proposed.

[0007] Therefore, the present application provides a power system cloud platform human-computer interaction method, system, device and medium, which can solve the problems of lack of accurate grasp of personalized needs of users, insufficient consideration of environmental factors, single and fixed optimization of interaction protocol, and imperfect data feedback and iterative updating mechanism in the prior art.

[0008] To solve the above technical problems, the present application provides the following technical solutions:

[0009] In a first aspect, the present application provides a power system cloud platform human-computer interaction method, comprising:

[0010] Obtaining user input data and environment data, and preprocessing the user input data and environment data;

[0011] Establishing a first interaction demand evaluation model according to the preprocessed user input data and environment data;

[0012] The first interaction demand evaluation model is used to analyze the interaction demand of the target user in different operation scenarios and environment conditions;

[0013] Based on the first interaction demand evaluation model, a second improved interaction protocol decision model is constructed;

[0014] The second improved interaction protocol decision model is used to optimize and adjust the interaction demand analyzed by the first interaction demand evaluation model;

[0015] The second improved interaction protocol decision model is called by a preset upper strategy function to generate an optimal interaction protocol, and a corresponding lower execution function is triggered to implement an interaction configuration operation.

[0016] As a preferred scheme of the power system cloud platform human-computer interaction method of the present application, wherein:

[0017] Obtaining interaction configuration operation execution result data;

[0018] The interaction configuration operation execution result data is fed back to the first interaction demand evaluation model and the second improved interaction protocol decision model for parameter iterative updating;

[0019] Until the iteration condition is met.

[0020] As a preferred scheme of the power system cloud platform human-computer interaction method of the present application, wherein:

[0021] A set of preset interaction demand indicators is included, and the set of interaction demand indicators includes a plurality of evaluation target user interaction demand indicators;

[0022] Based on the preprocessed user input data and environment data, interaction demand indicators are selected;

[0023] The selected interaction demand indicators are characterized by a scoring formula, and the interaction demand score of the target user is calculated according to the scoring formula;

[0024] According to the interaction demand score, the interaction demand of the user is classified to form the first interaction demand evaluation model.

[0025] As a preferred scheme of the power system cloud platform human-computer interaction method, wherein: the second improved interaction protocol decision model is constructed based on the first interaction demand evaluation model, and the second improved interaction protocol decision model comprises:

[0026] A preset interaction protocol optimization strategy set, the interaction protocol optimization strategy set comprises several optimization strategies for different interaction demand levels;

[0027] According to the user interaction demand level result obtained by the first interaction demand evaluation model, an appropriate optimization strategy is selected from the interaction protocol optimization strategy set;

[0028] The selected optimization strategy is parameterized and configured, and the second improved interaction protocol decision model capable of optimizing and adjusting the interaction demand is constructed by combining the interaction demand indicators and scores in the first interaction demand evaluation model.

[0029] This preferred scheme can accurately select an appropriate optimization strategy according to the specific interaction demand level of the user, so that the human-computer interaction protocol is more in line with the actual demand. By parameterizing and configuring the optimization strategy and combining the interaction demand indicators and scores, the second improved interaction protocol decision model has stronger pertinence and flexibility. This can not only effectively improve the efficiency of the power system cloud platform human-computer interaction, reduce unnecessary operation processes, but also significantly improve the user's interaction experience and reduce the problems and troubles that the user may encounter in the use process. At the same time, the model can optimize and adjust the interaction demand, which helps to respond to the changing user demand and environmental conditions in time, improves the stability and reliability of the power system cloud platform, and guarantees the safe and efficient operation of the power system. Moreover, in the long-term use process, with the continuous accumulation of interaction data, the model can be further optimized and improved, providing strong support for the intelligent development of the power system.

[0030] As a preferred scheme of the power system cloud platform human-computer interaction method, wherein: the second improved interaction protocol decision model is constructed based on the first interaction demand evaluation model, and the second improved interaction protocol decision model comprises:

[0031] Different interaction strategy rules under the combination of preset device health status and environmental factors;

[0032] The upper layer strategy function selects the corresponding rule from the preset interaction strategy rule according to the current obtained device health status and environmental factors;

[0033] The selected rule is transmitted to the second improved interaction protocol decision model as an input condition;

[0034] Comprehensive evaluation is performed on different interaction behaviors to obtain an optimal efficiency-reliability ratio interaction protocol under a current state.

[0035] As a preferred scheme of the power system cloud platform human-computer interaction method, the trigger corresponding lower layer execution function implements the interaction configuration operation, including:

[0036] When the optimal interaction protocol is determined, the upper layer strategy function triggers the corresponding lower layer execution function;

[0037] The lower layer execution function generates specific interaction configuration instructions according to the optimal interaction protocol;

[0038] The interaction configuration instructions include but are not limited to interface element setting, adjustment of interaction parameters, and optimization of display structure.

[0039] The interaction configuration instructions are sent to corresponding execution devices or personnel to implement specific interaction configuration operations.

[0040] As a preferred scheme of the power system cloud platform human-computer interaction method, the iteration condition includes:

[0041] When the feedback iteration number of the interaction configuration operation execution result data reaches the preset maximum iteration number, and the parameter change rate of the first interaction demand evaluation model and the second improved interaction protocol decision model is less than the preset threshold value;

[0042] Or the interaction performance index reflected by the interaction configuration operation execution result data exceeds the preset performance target value.

[0043] In a second aspect, the application provides a power system cloud platform human-computer interaction system, including:

[0044] A data acquisition and processing module is configured to acquire user input data and environment data, and pre-process the user input data and environment data;

[0045] A first model establishment module is configured to establish a first interaction demand evaluation model based on the pre-processed user input data and environment data;

[0046] The first interaction demand evaluation model is configured to analyze the interaction demand of a target user for human-computer interaction under different operation scenarios and environment conditions;

[0047] A second model establishment module is configured to construct a second improved interaction protocol decision model based on the first interaction demand evaluation model;

[0048] The second improved interaction protocol decision model is configured to optimize and adjust the interaction demand analyzed by the first interaction demand evaluation model;

[0049] The configuration operation module is configured to call the second improved interactive protocol decision model through a preset upper layer strategy function, generate an optimal interactive protocol, and trigger a corresponding lower layer execution function to implement an interactive configuration operation.

[0050] In a third aspect, the present application provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described above when executing the computer program.

[0051] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the method described above when executed by a processor.

[0052] Compared with the prior art, the present application has the following beneficial effects: the present application proposes a power system cloud platform human-computer interaction method, which fully considers user individualized needs and environmental factors by constructing a first interactive demand evaluation model and a second improved interactive protocol decision model. The first interactive demand evaluation model can accurately analyze the interactive needs of target users in different scenarios and environments, and the second improved interactive protocol decision model optimizes and adjusts these needs on this basis, thereby realizing dynamic optimization of the interactive protocol. In the interactive process, the preset upper layer strategy function selects appropriate rules from the preset rules according to the device health state and environmental factors, provides input conditions for the second improved interactive protocol decision model, and then obtains the interactive protocol with the optimal efficiency reliability ratio under the current state. And through the lower layer execution function, the specific interactive configuration instructions are sent to the corresponding execution device or personnel to implement the interactive configuration operation, ensuring the smooth progress of the interactive process.

[0053] Meanwhile, the present application also has a perfect data feedback and iterative updating mechanism. The interactive configuration operation execution result data is fed back to the two models for parameter iterative updating until the iterative condition is met, so that the system can continuously optimize and improve itself according to the actual interactive effect. BRIEF DESCRIPTION OF DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0055] Figure 1 A method flow chart of a power system cloud platform human-computer interaction method provided by an embodiment of the present application.

[0056] Figure 2An internal structure diagram of an electronic device of a power system cloud platform human-computer interaction method provided by an embodiment of the present application is shown. DETAILED DESCRIPTION

[0057] In order to make the above objectives, characteristics and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should fall within the scope of protection of the present application.

[0058] Embodiment 1, Reference Figure 1 For the first embodiment of the present application, the embodiment provides a power system cloud platform human-computer interaction method, comprising:

[0059] In the prior art, there are some problems, such as lack of accurate grasp of user's personalized needs, difficulty in providing customized interaction experience according to user's differences, leading to inconvenience of part of users and reduction of work efficiency; insufficient consideration of environmental factors, complex and changeable power operation environment will affect the perception of operating personnel, and the existing technology fails to effectively incorporate it into the interaction design, so that the interaction accuracy and reliability under special environment are greatly discounted; the interaction protocol optimization is single and fixed, lacks dynamic adjustment ability, cannot quickly adapt to the development of power system and the evolution of user's needs, and the data feedback and iterative updating mechanism is imperfect, which is not conducive to long-term improvement of human-computer interaction quality and performance.

[0060] The present application provides a method that can effectively solve the above-mentioned problems. Next, how to realize the power system cloud platform human-computer interaction method will be described in detail in combination with multiple embodiments.

[0061] Figure 1 A method flowchart of a power system cloud platform human-computer interaction method is shown, comprising:

[0062] S101, obtaining user input data and environment data, and preprocessing the user input data and environment data, wherein:

[0063] It should be noted that in order to realize the intelligentization and precision of the power system cloud platform human-computer interaction, it is essential to obtain comprehensive and accurate user input data and environment data.

[0064] In some specific embodiments, the user input data covers various instructions, operation records, feedback information and the like of the operating personnel in the operation process, which reflects the operation habits, skill level and current operation needs of the user.

[0065] In some specific embodiments, the environmental data includes temperature, humidity, light intensity, electromagnetic interference and other parameters of the power system operating environment, which directly affect the perception and operation of the operator.

[0066] It should be noted that after obtaining the user input data and the environmental data, considering the diversity and complexity of the data, in order to facilitate subsequent analysis and processing, the data needs to be preprocessed.

[0067] In some specific embodiments, the preprocessing process usually includes data cleaning, data normalization and data feature extraction. Data cleaning is to remove noise, outliers and duplicate data in the data to ensure the quality of the data. For example, there may be some misoperation records or invalid instructions in the user input data, which can be excluded by data cleaning.

[0068] In some specific embodiments, data normalization is to convert data of different ranges and scales to a unified interval to eliminate the dimensional differences between data and facilitate subsequent calculation and comparison. For example, the numerical range of temperature, humidity and light intensity is very different, and after normalization, they can be analyzed on the same scale. Data feature extraction is to extract representative and discriminative features from original data to reduce the dimension of data and improve the calculation efficiency and accuracy of the model.

[0069] In the human-computer interaction method of the present application, by preprocessing the user input data and the environmental data, more standardized, accurate and useful data can be obtained, providing a solid data foundation for subsequent establishment of the first interaction demand evaluation model.

[0070] Specifically, the system in the present application first collects two types of key data from the power system cloud platform, including:

[0071] User input data U: including user operation behavior logs such as button clicks, menu selections, command inputs, mouse tracks, operation time consumption, error times, and undo operation frequencies. In addition, it also includes user identity information such as role type (dispatcher, inspector, administrator), permission level, historical operation habit tag, etc.

[0072] Environmental data E: including device health status (such as transformer load rate, circuit breaker action times, alarm level), external physical environment (temperature, humidity, light intensity), network communication quality (delay, packet loss rate), system load (concurrent user number, CPU occupancy rate), etc. All raw data form a high-dimensional vector:

[0073] X raw = [U, E]

[0074] wherein represents the original input vector with dimension d.

[0075] Further, in order to ensure the accuracy and stability of subsequent modeling, the X raw needs to be cleaned and standardized.

[0076] Further, the data cleaning function f clean (·) is defined as follows:

[0077] X clean = f clean (X raw )

[0078] It should be noted that this function removes obvious outliers (in this invention, single clicks with operation time exceeding 1 hour are considered invalid), interpolates or fills missing fields (in this invention, mean, median or forward filling is used), filters duplicate records or redundant log entries, and outputs the cleaned intermediate data X clean .

[0079] Further, due to the huge difference in dimension of different features (such as temperature unit is Celsius, operation frequency is times / minute), normalization processing is needed. This invention uses the min-max scaling method:

[0080]

[0081] where x i is the value of the i-th feature in the original vector X clean ; x min,i and x max,i are the minimum and maximum values of the feature in the historical data, respectively; x′ i is the normalized value, ranging between [0, 1].

[0082] Further, the standardized input vector is obtained:

[0083] X preprocessed = [x′1, x′2, …, x′ d ] ∈ [0, 1] d

[0084] It should be noted that this vector will be used as the unified input format for subsequent modeling.

[0085] It should be noted that preprocessing is a basic step to ensure the generalization ability of the model, avoiding the over-amplification or neglect of certain features due to inconsistent dimensions.

[0086] S102, a first interaction demand assessment model is established according to the preprocessed user input data and environmental data, wherein:

[0087] It should be noted that after obtaining the preprocessed user input data and environment data, a first interaction demand evaluation model can be constructed using these data. The main purpose of this model is to in-depth analyze the specific needs of the target user for human-computer interaction in different operation scenarios and environmental conditions.

[0088] In some specific embodiments, the first interaction demand evaluation model can be constructed using machine learning algorithms such as decision trees, support vector machines, or neural networks, etc. These algorithms can learn the complex relationship between user input data, environment data, and interaction demand from a large amount of preprocessed data. Taking neural networks as an example, it has strong non-linear mapping ability and can process high-dimensional input data. The preprocessed user input data and environment data are taken as the nodes of the input layer, and after non-linear transformation of the hidden layer, the interaction demand of the target user in the current operation scenario and environmental condition is finally output.

[0089] In some specific embodiments, the first interaction demand evaluation model can also be constructed in combination with expert experience. Experts in the field of power systems have rich practical experience and professional knowledge, and they can reasonably predict and analyze the demand for human-computer interaction according to different operation scenarios and environmental conditions. Integrating expert experience into the model construction process can make the model more in line with the needs of practical applications, improve the accuracy and reliability of the model. For example, experts can give corresponding interaction demand suggestions according to different device health states and environmental factors. These suggestions can be used as prior knowledge for model training to help the model better learn and understand the user's interaction demand.

[0090] It should be noted that the first interaction demand evaluation model constructed by combining machine learning algorithms and expert experience can more comprehensively and accurately analyze the specific needs of the target user for human-computer interaction in different operation scenarios and environmental conditions. The output of this model will provide an important basis for the subsequent construction of the second improved interaction protocol decision model, thereby further optimizing the human-computer interaction protocol and improving the efficiency and reliability of the human-computer interaction of the power system cloud platform.

[0091] It should be noted that the present application does not limit the establishment method of the first interaction demand evaluation model, but regardless of the method used, it should ensure that the model can fully utilize the preprocessed data and accurately reflect the user's interaction demand in different scenarios and environments.

[0092] In the embodiments of the present application, the first interaction demand evaluation model is used to analyze the interaction demand of the target user for human-computer interaction in different operation scenarios and environmental conditions.

[0093] In the embodiments of the present application, the first interaction demand evaluation model is established according to the preprocessed user input data and the environment data, which comprises:

[0094] The preset interaction demand index set comprises a plurality of evaluation target user interaction demand indexes;

[0095] The interaction demand index selection is performed based on the preprocessed user input data and the environment data;

[0096] The selected interaction demand index is scored according to a scoring formula, and the interaction demand score of the target user is calculated according to the scoring formula;

[0097] The interaction demand of the user is graded according to the interaction demand score, and the first interaction demand evaluation model is formed.

[0098] Specifically, the first interaction demand evaluation model M1 is established, and the preset interaction demand index set is To quantify the interaction demand of the user, a group of measurable interaction demand indexes are defined in advance. These indexes should cover multiple dimensions such as operation efficiency, cognitive load, visual perception, fault tolerance, etc. Let the interaction demand index set be:

[0099]

[0100] Wherein, each I k represents the kth interaction demand dimension, for example, in the present application, I1 can be designed to represent operation complexity demand, reflecting whether the user faces multi-step, high-logic-coupled operation tasks; I2 represents information density preference, indicating whether the user prefers to view a large amount of data or a concise abstract; I3 represents response speed sensitivity, measuring the user's tolerance to system feedback delay; I4 represents visual comfort demand, evaluating whether the current lighting and screen brightness affect readability; I5 represents operation assistance dependency, judging whether the user frequently uses help prompts or guide functions.

[0101] Furthermore, not all indexes are important in every scenario. Therefore, the most relevant subset needs to be dynamically selected according to the current X preprocessed The correlation score s k of each index is calculated using a feature importance scoring method (such as Gini importance or mutual information based on random forest), and then the indexes with scores higher than the threshold τ are retained to form the selected index set:

[0102]

[0103] Wherein, τ is a preset significance threshold, usually taking 0.1 or being determined through cross-validation.

[0104] Furthermore, for each selected index Design a score function S k (·), which is mapped to a normalized score.

[0105] For example, for I3 (response speed sensitivity), we can define:

[0106]

[0107] where t response is the current system average response time; t min and t max are the minimum and maximum of historical response times; the lower the score, the more sensitive the user is to delay.

[0108] Similarly, other indicators also need to establish corresponding score functions.

[0109] Furthermore, the scores of all selected indicators are weighted and summed to obtain the total interaction demand score:

[0110]

[0111] where n is the number of indicators actually involved in scoring; w k is the weight coefficient of the kth indicator, satisfying It can be obtained by expert scoring method or machine learning method (such as linear regression fitting user satisfaction); S k (X preprocessed ) ∈ [0, 1] is the score of the kth indicator; D user ∈ [0, 1] or extended to [0, 5], representing the overall interaction demand intensity of the user.

[0112] Furthermore, according to the numerical interval of D user , the user demand is divided into several levels, which are used for subsequent strategy matching.

[0113] For example, three classifications can be used:

[0114]

[0115] where θ1 and θ2 are preset classification thresholds, for example θ1 = 1.5, θ2 = 3.0 (if D user ∈ [0, 5]); L represents the demand level of the current user.

[0116] Thus, the first interaction demand evaluation model M1 is defined as:

[0117]

[0118] That is, the input preprocessed data, output demand score and level.

[0119] It should be noted that the role of M1 is to "diagnose" the current interactive stress state of the user, providing a basis for the next step of optimization.

[0120] S103, constructing a second improved interactive protocol decision model based on the first interactive demand evaluation model, wherein:

[0121] It should be noted that when the output result of the first interactive demand evaluation model is obtained, the interactive demand score and demand level of the target user are obtained, and the second improved interactive protocol decision model can be constructed based on this. The main purpose of this model is to dynamically adjust and optimize the human-computer interaction protocol according to the user's interactive demand, so as to improve the efficiency and reliability of the human-computer interaction of the power system cloud platform.

[0122] In some specific embodiments, the second improved interactive protocol decision model can be constructed using reinforcement learning algorithm. Reinforcement learning is a machine learning method that interacts with the environment through an agent, constantly tries different actions, and learns the optimal strategy according to the reward signal feedback from the environment. In the human-computer interaction scenario of the present application, the agent can select appropriate interactive protocol parameters for adjustment according to the user's interactive demand score and demand level output by the first interactive demand evaluation model, such as interface layout, operation process, and display mode of prompt information. The reward signal feedback from the environment can be defined according to the user's operation efficiency, satisfaction and other indicators, for example, if the user's operation efficiency and satisfaction increase under the adjusted interactive protocol, a positive reward is given; otherwise, a negative reward is given. Through continuous interaction and learning with the environment, the agent can gradually find the optimal interactive protocol adjustment strategy, thereby realizing the optimization of human-computer interaction protocol.

[0123] In some specific embodiments, the second improved interactive protocol decision model can also be constructed in combination with a rule engine. Rule engine is a rule-based reasoning system that can match and process input information according to pre-set rules to obtain corresponding decision results. In the scenario of human-computer interaction of the power system cloud platform, the rule engine can pre-set a series of rules according to the professional knowledge and practical experience in the field of power system, which can describe the interactive protocol adjustment strategy that should be taken under different user interactive demand scores and demand levels. For example, when the user's interactive demand level is high demand, the rule engine can decide to increase the display frequency of operation prompt information and simplify the operation process according to the pre-set rules. Combining the rule engine with the reinforcement learning algorithm to construct the second improved interactive protocol decision model can fully utilize the advantages of both.

[0124] It should be noted that the construction manner of the second improved interaction protocol decision model is not strictly limited, as long as the output of the first interaction demand evaluation model, i.e., the interaction demand score and demand level of the target user, is inputted, and the optimized human-computer interaction protocol parameters are outputted.

[0125] In the embodiment of the application, the second improved interaction protocol decision model is used to optimize and adjust the interaction demand analyzed by the first interaction demand evaluation model.

[0126] In the embodiment of the application, the second improved interaction protocol decision model is constructed based on the first interaction demand evaluation model, and the construction comprises:

[0127] A preset interaction protocol optimization strategy set is provided, and the interaction protocol optimization strategy set comprises several optimization strategies for different interaction demand levels.

[0128] According to the user interaction demand level result obtained by the first interaction demand evaluation model, an adaptive optimization strategy is selected from the interaction protocol optimization strategy set.

[0129] The selected optimization strategy is parameterized and configured, and the second improved interaction protocol decision model capable of optimizing and adjusting the interaction demand is constructed in combination with the interaction demand indicators and scores in the first interaction demand evaluation model.

[0130] Specifically, the second improved interaction protocol decision model M2 is constructed, and a preset interaction protocol optimization strategy set For different demand levels, a set of executable interaction optimization strategies is preset. Let the strategy set be:

[0131]

[0132] Wherein, each P j is an optimization scheme, and the specific strategy set designed by the application is as follows: P1 represents a simplified mode, which is suitable for a low demand scenario, and the actions include hiding non-key buttons, reducing animation effects, and compressing information panels; P2 represents a standard enhancement mode, which is suitable for a medium demand, and the actions include increasing operation prompts, enabling dynamic highlighting, and moderately using sound effects; and P3 represents a high reliability mode, which is suitable for a high demand or an emergency situation, and the actions include forced double confirmation, starting voice broadcast, enlarging interface elements, and enhancing color contrast.

[0133] Further, according to the level L output by M1, the most matched strategy P * is selected from .

[0134] The strategy selection function is defined as:

[0135] P *= SelectStrategy(L)

[0136] For example, if L = L1, then P * = P1 if L = L2, then P * = P2 if L = L3, then P * = P3.

[0137] Further, to make the strategy more flexible, it is parameterized. Take P3 (high reliability mode) as an example:

[0138]

[0139] wherein θ confirm ∈{0,1} indicates whether to enable the double confirmation mechanism (1 indicates to enable) ; θ audio ∈{0,1} indicates whether to start voice feedback; θ scale ∈[1.0,2.5] indicates the magnification ratio of interface elements (1.0 is the original size) ; θ contrast ∈[1.0,3.0] indicates the color contrast enhancement coefficient (the higher the clearer). These parameters can be dynamically adjusted according to the specific value of D user .

[0140] For example, θ scale may be represented as follows:

[0141] When D user ∈[3.0,5.0]

[0142] The stronger the demand, the more the interface magnification.

[0143] Further, a second improved interactive protocol decision model M2 is constructed;

[0144] In the present application, M2 is a mapping function:

[0145]

[0146] wherein θ * is the optimal parameter vector for guiding the interaction configuration.

[0147] It should be noted that M2 not only selects the strategy, but also finely adjusts the parameters to realize the "personalization + contextualization" of the interaction optimization.

[0148] S104, the second improved interactive protocol decision model is called through the preset upper strategy function, the optimal interactive protocol is generated, and the corresponding lower execution function is triggered to implement the interaction configuration operation, wherein:

[0149] It is necessary to explain that when the second improved interactive protocol decision model is obtained, the preset upper strategy function can be used to call the model. The upper strategy function plays a role of overall planning and coordination, which will call the second improved interactive protocol decision model according to the overall operation of the power system cloud platform, the specific needs of the user and the resource limitations of the system and other factors. When calling, the upper strategy function will accurately pass the input information required by the second improved interactive protocol decision model to the second improved interactive protocol decision model, that is, the interactive demand score and demand level of the target user output by the first interactive demand evaluation model, and the preprocessed user input data and environmental data.

[0150] After receiving these input information, the second improved interactive protocol decision model will output the optimal interactive protocol according to its internal algorithm and logic. This optimal interactive protocol is obtained after accurate analysis of user interactive demand and optimization adjustment of interactive protocol, which fully considers the interactive demand of users in different scenarios and environments, aiming to maximize the efficiency and reliability of the human-computer interaction of the power system cloud platform.

[0151] Once the optimal interactive protocol is generated, the system will trigger the corresponding lower execution function to implement the interactive configuration operation. The lower execution function is responsible for applying the optimal interactive protocol to the human-computer interaction of the power system cloud platform. For example, it will adjust the interface layout according to the requirements of the interactive protocol, which may rearrange the position of buttons, change the display mode of menus; optimize the operation process, which may simplify some complex operation links; change the display mode of prompt information, which may adjust the font size, color and display position of prompt information. Through these specific operations, the optimized interactive protocol is truly implemented, realizing the efficient and reliable human-computer interaction of the power system cloud platform.

[0152] In the process of implementing the interactive configuration operation, the system will also monitor the feedback and operation of the user in real time. If it is found that the interactive effect does not meet the expectation, the system will restart the whole process of interactive demand evaluation and protocol optimization. For example, if the user's operation efficiency does not improve and the satisfaction does not increase under the new interactive protocol, the system will collect the user's input data and environmental data again, reevaluate the user's interactive demand score and demand level through the first interactive demand evaluation model, then update the second improved interactive protocol decision model based on this, generate a new optimal interactive protocol and implement the interactive configuration operation again, until the satisfactory interactive effect is achieved.

[0153] In addition, the system also records and analyzes the results of each interaction configuration operation to continuously accumulate experience and further improve the first interaction demand assessment model and the second improved interaction protocol decision model. Through long-term learning and optimization, the system can better adapt to the interaction needs of different users in different scenarios, continuously improving the performance and quality of the human-computer interaction of the power system cloud platform.

[0154] In the embodiments of the present application, the second improved interaction protocol decision model is called by the preset upper strategy function, which includes:

[0155] presetting interaction strategy rules under different combinations of device health status and environmental factors;

[0156] The upper strategy function selects the corresponding rule from the preset interaction strategy rules according to the current obtained device health status and environmental factors;

[0157] The selected rule is passed to the second improved interaction protocol decision model as an input condition;

[0158] Comprehensive evaluation is performed on different interaction behaviors to obtain the interaction protocol with the optimal efficiency reliability ratio under the current state.

[0159] In the embodiments of the present application, triggering the corresponding lower execution function to implement the interaction configuration operation includes:

[0160] After determining the optimal interaction protocol, the upper strategy function triggers the corresponding lower execution function;

[0161] The lower execution function generates specific interaction configuration instructions according to the optimal interaction protocol;

[0162] The interaction configuration instructions include but are not limited to interface element setting, adjusting interaction parameters, and optimizing display structure;

[0163] The interaction configuration instructions are sent to the corresponding execution device or personnel to implement specific interaction configuration operations.

[0164] Specifically, the upper strategy function calls the optimal protocol generation, and the interaction strategy rules under the preset combination of device health status and environmental factors are introduced to ensure the safety of the system, and the upper strategy function F upper is introduced, which determines whether to override the default strategy according to the device health status H and the environmental factor E.

[0165] Further, the rule set is defined. Each rule is in the form of:

[0166] If the device health status H is "fault warning" and the environmental light intensity E light is less than 50 lux, then the interaction mode P3 is forcibly enabled and the scaling factor θ scale= 2.0.

[0167] If the device health state H is "normal" and the network delay E network_delay is greater than 500 milliseconds, disable the animation effect to improve response speed.

[0168] Further, set the upper layer policy function selection rule as follows:

[0169] R * = F upper (H, E)

[0170] The output is the current rule R * that may modify the output of M2 or directly specify parameters.

[0171] Further, obtain the interactive protocol with optimal efficiency and reliability ratio, comprehensively consider efficiency and reliability, and define a comprehensive evaluation function:

[0172]

[0173] wherein π represents a candidate interactive protocol; η eff (π) ∈ [0, 1] is the efficiency gain, defined as the operation step reduction ratio or the response time shortening rate; is the reliability improvement degree, defined as the operation error probability reduction ratio; λ > 0 is the trade-off coefficient, usually set to a value greater than 1 (such as λ = 2), embodying the "reliability first" principle; the greater J(π) is, the more optimal the protocol is under the current conditions. Select the protocol π* with the maximum J(π) as the final decision result.

[0174] Further, trigger the lower layer execution function to implement the interactive configuration, when π* is determined, the upper layer policy function triggers the lower layer execution function Flower.

[0175] Further, Flower converts the abstract protocol π* into executable commands, which can be designed as follows in the present application: set the interface font size to 18pt, change the button background color to red (#FF0000), enable voice prompt: "Please confirm the circuit breaker operation", and add a secondary confirmation pop-up window.

[0176] It should be noted that these instructions can be sent to the front-end application or terminal device in JSON or XML format.

[0177] Further, the instructions can be pushed to the client through WebSocket or RESTAPI to update the user interface and interactive logic in real time. This step realizes the landing of "decision to execution", ensuring that the intelligent optimization can really act on the user experience.

[0178] In the embodiment of the present application, the interactive configuration operation execution result data is obtained;

[0179] The interactive configuration operation execution result data is fed back to the first interactive demand evaluation model and the second improved interactive protocol decision model for parameter iterative updating;

[0180] until the iteration condition is met.

[0181] In summary, the present application proposes a power system cloud platform human-computer interaction method, which fully considers the user's individualized demand and environmental factors by constructing a first interactive demand evaluation model and a second improved interactive protocol decision model. The first interactive demand evaluation model can accurately analyze the interactive demand of the target user in different scenarios and environments, and the second improved interactive protocol decision model optimizes and adjusts these demands on this basis, thereby realizing the dynamic optimization of the interactive protocol. In the interaction process, the preset upper strategy function selects the appropriate rule from the preset rule according to the device health state and environmental factors, provides the input condition for the second improved interactive protocol decision model, and then obtains the interactive protocol with the optimal efficiency reliability ratio under the current state. And through the lower execution function, the specific interactive configuration instruction is sent to the corresponding execution device or personnel, and the interactive configuration operation is implemented to ensure the smooth progress of the interaction process.

[0182] Meanwhile, the present application also has a perfect data feedback and iterative updating mechanism. The interactive configuration operation execution result data is fed back to the two models for parameter iterative updating until the iteration condition is met, so that the system can continuously optimize and improve itself according to the actual interaction effect.

[0183] In one preferred embodiment, the detailed steps of feedback and model iterative updating can be designed as follows:

[0184] The interactive configuration execution result data Y is obtained result , the following feedback data is collected: actual operation completion time t actual ; error occurrence number e; user's active score r∈[1, 5]; whether to quit halfway q∈{0, 1}, combined into a result vector:

[0185] Y result = [t actual , e, r, q]

[0186] Further, the parameters of M1 and M2 are updated using Y result .

[0187] For example, if the user's score is low but the system predicts high satisfaction, the weight w k in M1 is adjusted:

[0188]

[0189] in, It is a loss function, such as mean squared error. η is the model's predicted score; η is the learning rate; t represents the number of iterations.

[0190] Furthermore, it checks whether the iteration termination condition is met. The iteration will continue until one of the following conditions is met:

[0191] Condition 1: The maximum number of iterations is reached and the parameters tend to converge;

[0192] t≥T max and

[0193] Among them, T max θ is the preset maximum number of iterations (e.g., 100 times); θ is the model parameter vector; ∈ is the convergence threshold (e.g., 0.001), indicating that the parameter changes are minimal and tend to be stable.

[0194] Condition 2: The interaction performance exceeds the preset target;

[0195] Performance (Y) result )≥P target

[0196] Where Performance(·) is the performance metric function, P target This indicates the performance value of the corresponding indicator.

[0197] Once condition two is met, meaning the interaction performance exceeds the preset target, it indicates that the model has achieved a relatively ideal effect in the current iteration. At this point, iteration can be stopped, and the current model parameters can be used as the final parameters for subsequent human-computer interactions. In practical applications, the performance indicator function Performance(·) can be defined according to specific business needs and evaluation criteria, such as the improvement in user satisfaction or the reduction in interaction response time. For different performance indicators, the preset target value P... target This will vary and needs to be set appropriately based on the actual situation. When condition one is met, i.e., the maximum number of iterations is reached and the parameters tend to converge, it indicates that the model has basically stabilized after multiple iterations, and further iterations may not bring significant performance improvements. At this point, iteration is also stopped, and the current model parameters are used as the final parameters.

[0198] It should be noted that once any condition is met, the iteration stops and the current optimal model is retained.

[0199] Example 3, referring to Figure 2 This embodiment also provides a human-computer interaction system for a power system cloud platform, including:

[0200] a data acquisition and processing module configured to acquire user input data and environment data, and to pre-process the user input data and the environment data;

[0201] a first model establishing module configured to establish a first interaction demand evaluation model according to the pre-processed user input data and the environment data;

[0202] the first interaction demand evaluation model is configured to analyze interaction demands of a target user for human-computer interaction in different operation scenarios and environment conditions;

[0203] a second model establishing module configured to construct a second improved interaction protocol decision model based on the first interaction demand evaluation model;

[0204] the second improved interaction protocol decision model is configured to optimize and adjust the interaction demands analyzed by the first interaction demand evaluation model;

[0205] a configuration operation module configured to call the second improved interaction protocol decision model through a preset upper-layer strategy function, to generate an optimal interaction protocol, and to trigger a corresponding lower-layer execution function to implement an interaction configuration operation.

[0206] The above-mentioned various unit modules can be embedded in or independent of a processor in an electronic device in a hardware form, or can be stored in a memory in the electronic device in a software form, so as to be called and executed by the processor to perform the operations corresponding to the above-mentioned various modules.

[0207] The embodiment also provides an electronic device, which can be a terminal. An internal structure diagram of the electronic device can be as shown in Figure 2 The electronic device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the electronic device is configured to provide calculation and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The communication interface of the electronic device is configured to perform wired or wireless communication with an external terminal. The wireless communication can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program is executed by the processor to implement a power system cloud platform human-computer interaction method. The display screen of the electronic device can be a liquid crystal display screen or an electronic ink display screen. The input device of the electronic device can be a touch layer overlaid on the display screen, or can be a key, trackball or touchpad arranged on the shell of the electronic device, or can be an external keyboard, touchpad or mouse, etc.

[0208] The embodiment also provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the following steps:

[0209] Obtain user input data and environment data, and pre-process the user input data and the environment data;

[0210] Establish a first interaction demand evaluation model according to the pre-processed user input data and the environment data;

[0211] The first interaction demand evaluation model is used to analyze the interaction demand of a target user for human-computer interaction in different operation scenarios and environment conditions;

[0212] A second improved interaction protocol decision model is constructed based on the first interaction demand evaluation model;

[0213] The second improved interaction protocol decision model is used to optimize and adjust the interaction demand analyzed by the first interaction demand evaluation model;

[0214] The second improved interaction protocol decision model is called by a preset upper strategy function to generate an optimal interaction protocol, and a corresponding lower execution function is triggered to implement an interaction configuration operation.

[0215] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application, and although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and all modifications and replacements should be covered in the scope of the claims of the present application.

[0216] Although the preferred embodiments of the present application have been described, those skilled in the art can make further changes and modifications to the embodiments once they understand the basic inventive concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0217] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A power system cloud platform human-computer interaction method, characterized in that, The method comprises the following steps: acquiring user input data and environment data, and preprocessing the user input data and the environment data; establishing a first interaction demand evaluation model according to the preprocessed user input data and the environment data; the first interaction demand evaluation model is used to analyze the interaction demand of a target user in different operation scenarios and environment conditions; a second improved interaction protocol decision model is constructed based on the first interaction demand evaluation model; the second improved interaction protocol decision model is used to optimize and adjust the interaction demand analyzed by the first interaction demand evaluation model; an optimal interaction protocol is generated by calling the second improved interaction protocol decision model through a preset upper layer strategy function, and a corresponding lower layer execution function is triggered to implement an interaction configuration operation.

2. The power system cloud platform human-computer interaction method of claim 1, wherein, Further comprising: acquiring interaction configuration operation execution result data; feeding back the interaction configuration operation execution result data to the first interaction demand evaluation model and the second improved interaction protocol decision model for parameter iterative updating; until the iteration condition is met.

3. The power system cloud platform human-computer interaction method of claim 2, wherein, The first interaction demand evaluation model is established according to the preprocessed user input data and the environment data, which comprises the following steps: a preset interaction demand index set is established, the interaction demand index set comprises a plurality of evaluation target user interaction demand indexes; interaction demand index selection is performed based on the preprocessed user input data and the environment data; the selected interaction demand indexes are represented by a scoring formula, and the interaction demand score of the target user is calculated according to the scoring formula; the interaction demand of the user is graded according to the interaction demand score, and the first interaction demand evaluation model is formed.

4. The power system cloud platform human-computer interaction method of claim 3, wherein, The second improved interaction protocol decision model is constructed based on the first interaction demand evaluation model, which comprises the following steps: a preset interaction protocol optimization strategy set is established, the interaction protocol optimization strategy set comprises a plurality of optimization strategies for different interaction demand grades; an adaptive optimization strategy is selected from the interaction protocol optimization strategy set according to the user interaction demand grading result obtained by the first interaction demand evaluation model; the selected optimization strategy is parameterized configured, and the second improved interaction protocol decision model capable of optimizing and adjusting the interaction demand is constructed by combining the interaction demand indexes and scores in the first interaction demand evaluation model.

5. The power system cloud platform human-computer interaction method of claim 4, wherein, The second improved interaction protocol decision model is called by the preset upper layer strategy function, which comprises the following steps: preset interaction strategy rules under different combinations of device health status and environmental factors are selected; the upper layer strategy function selects the corresponding rule from the preset interaction strategy rules according to the current acquired device health status and environmental factors; the selected rule is passed to the second improved interaction protocol decision model as an input condition; the optimal efficiency reliability ratio of the interaction protocol under the current state is obtained by comprehensively evaluating different interaction behaviors.

6. The power system cloud platform human-machine interaction method of claim 5, wherein, The corresponding lower layer execution function is triggered to implement the interaction configuration operation, which comprises the following steps: when the optimal interaction protocol is determined, the upper layer strategy function triggers the corresponding lower layer execution function; the lower layer execution function generates specific interaction configuration instructions according to the optimal interaction protocol; the interaction configuration instructions include but are not limited to interface element setting, adjusting interaction parameters, and optimizing display structure. The interaction configuration instruction is sent to the corresponding execution device or personnel to implement a specific interaction configuration operation.

7. The power system cloud platform human-machine interaction method of claim 6, wherein, The iteration condition comprises: When the feedback iteration number of the interaction configuration operation execution result data reaches a preset maximum iteration number, and the parameter variation rate of the first interaction demand evaluation model and the second improved interaction protocol decision model is less than a preset threshold value; Or the interaction performance index reflected by the interaction configuration operation execution result data exceeds a preset performance target value.

8. A power system cloud platform human-machine interaction system, applying the method of any one of claims 1-7, characterized in that, Comprise: A data acquisition and processing module configured to acquire user input data and environment data, and to pre-process the user input data and the environment data; A first model establishing module configured to establish a first interaction demand evaluation model based on the pre-processed user input data and the environment data; The first interaction demand evaluation model is configured to analyze the interaction demand of a target user for human-computer interaction in different operation scenarios and environment conditions; A second model establishing module configured to construct a second improved interaction protocol decision model based on the first interaction demand evaluation model; The second improved interaction protocol decision model is configured to optimize and adjust the interaction demand analyzed by the first interaction demand evaluation model; An operation configuration module configured to call the second improved interaction protocol decision model through a preset upper-layer strategy function, to generate an optimal interaction protocol, and to trigger a corresponding lower-layer execution function to implement an interaction configuration operation. 9.An electronic device comprising a memory and a processor, the memory storing a computer program, wherein, The processor executes the computer program to implement the steps of the power system cloud platform human-computer interaction method of any one of claims 1-7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the power system cloud platform human-computer interaction method of any one of claims 1-7.

Images