Adaptive Sleep Method and Energy Saving System for Electromechanical Equipment Driven by User Activity

Abstract

This application relates to the technical field of electromechanical equipment, and provides an adaptive sleep method and an energy-saving system for electromechanical equipment driven by user activity. By integrating a perception acquisition module, an edge processing module, and a sleep decision-making module, an organically coordinated energy-saving system is constructed to achieve the energy-saving management of electromechanical equipment. Information is interacted in real time and the modules work together, enabling accurate perception of the operating state, user behavior, and environmental changes of electromechanical equipment, so as to determine whether to trigger the projection interaction mode and sleep mode of electromechanical equipment, in order to improve the intelligence level, energy utilization efficiency, and user experience of electromechanical equipment, and achieve energy saving of electromechanical equipment.

Description

Technical Field

[0001] This application relates to the technical field of electromechanical equipment, and particularly to an adaptive sleep method and energy-saving system for electromechanical equipment driven by user activity. Background Art

[0002] In the actual application of electromechanical equipment, most of the existing sleep strategies of electromechanical equipment do not fully consider the factor of user activity. Usually, a preset fixed sleep time or a simple operation interval determination method is adopted, which cannot accurately match the actual usage status of users. Moreover, the current evaluation methods for user activity are single and inaccurate, mainly relying on simple device operation counts, such as the number of clicks and key presses, etc., ignoring the importance of multi-dimensional information such as environmental data and network data for judging user activity.

[0003] For example, the patent application with the authorization announcement number CN214278671U discloses an energy-saving control system for electromechanical equipment installation, including an electromechanical equipment installation end and a device shell. The electromechanical equipment installation end conducts the installation process of electromechanical equipment, and at the same time, signals will be transmitted and collected during installation. This technical solution includes a lighting system and a signal converter. Among them, the signals of the lighting system are directly transmitted to the wireless signal transmitter for transmission, and the signal feedback module of the system can transmit the energy-saving signal in a timely manner.

[0004] The above technical solutions have the problems raised in this background art: they do not fully consider the factor of user activity and cannot accurately match the actual usage status of users; and the current evaluation methods for user activity are single and inaccurate. Summary of the Invention

[0005] In view of the deficiencies of the prior art, this application provides an adaptive sleep method and energy-saving system for electromechanical equipment driven by user activity.

[0006] In a first aspect, this application provides an adaptive energy-saving system for electromechanical equipment driven by user activity. The system includes: a perception acquisition module, an edge processing module, and a sleep decision module;

[0007] The perception acquisition module is used to acquire perception data and complete the automatic networking of sensor nodes and the transmission of perception data through a self-organization mechanism. When an abnormality occurs in the sensor nodes, it triggers network topology reconstruction;

[0008] The edge processing module is used to extract the perception features of the perception data to dynamically evaluate user activity, and comprehensively judge whether to trigger the projection interaction mode according to the activity interval, activity trend, and the duration of the user in the low-activity state;

[0009] The sleep decision module is used to continuously monitor the duration of the electromechanical device in the projection interaction mode when the electromechanical device is in the projection interaction mode, determine whether to switch to the sleep mode according to the duration and the task priority, and when it is necessary to switch to the sleep mode, comprehensively weigh the power consumption and performance of the electromechanical device as well as the user's historical preferences, formulate and optimize an intelligent sleep strategy through a genetic algorithm, determine the sleep depth and sleep time of the electromechanical device, and display the operation simulation state of the electromechanical device under different intelligent sleep strategies through holographic projection, including the power consumption change and performance fluctuation of the electromechanical device.

[0010] As an alternative implementation, the perception data includes device operation data, environmental data, and network data, and the logic of the self-organization mechanism includes:

[0011] Generate a node identifier for each sensor node based on the geographical location of the sensor node, and perform neighbor discovery according to the sensor node to generate a neighbor table for each sensor node;

[0012] Based on the link quality index and energy reserve in the neighbor table, construct an initial network topology, and elect a parent node according to the node degree, signal strength, and energy reserve to form a hierarchical network topology;

[0013] According to the hierarchical network topology structure, construct a routing table, and automatically select a single-hop or multi-hop transmission mode according to the data priority;

[0014] Each sensor node determines a data transmission path according to the routing table and starts a hybrid retransmission mechanism;

[0015] During the data transmission process, each sensor node periodically performs network topology scanning, real-time monitors the status information of each sensor node, and determines whether each sensor node has an abnormality to send an abnormality signal to trigger network topology reconstruction.

[0016] As an alternative implementation, the logic of the network topology reconstruction includes:

[0017] Receive the abnormality signal, transmit the abnormality data of the abnormal sensor node to the sink node through the network and integrate it to form an abnormality data set;

[0018] Based on the abnormality data set, determine the abnormality type of the sensor node, and the abnormality type includes local node failure, local network failure, and global network failure;

[0019] Trigger a multi-level reconstruction mechanism according to the abnormality type of the sensor node, and verify the reconstruction effect to optimize the network topology reconstruction.

[0020] As an alternative implementation, the evaluation logic of the user activity includes:

[0021] Extract the perceptual features of the perceptual data, where the perceptual features include device operation frequency, light change rate, and network request frequency;

[0022] Calculate the activity score based on the perceptual features to divide the activity intervals;

[0023] Calculate the change rate of the activity score to judge the activity trend.

[0024] As an optional implementation manner, the triggering logic of the projection interaction mode includes:

[0025] Judge whether the user activity is in the low activity interval and judge the user's activity trend to obtain whether the user is in the low activity state;

[0026] Record the continuous duration of the user in the low activity state to obtain whether to trigger the projection interaction mode;

[0027] If the projection interaction mode is triggered, obtain the user's historical preferences based on the device operation data to generate personalized projection content.

[0028] As an optional implementation manner, the conversion judgment logic of the sleep mode includes:

[0029] Obtain the running tasks of the electromechanical device and evaluate the task priorities;

[0030] Monitor the continuous duration of the electromechanical device in the projection interaction mode;

[0031] Judge whether it is necessary to convert to the sleep mode according to the running task priorities and the continuous duration of the electromechanical device in the projection interaction mode. If it is necessary to convert to the sleep mode, generate a sleep signal.

[0032] As an optional implementation manner, the intelligent sleep strategy includes:

[0033] Receive the sleep signal, encode the sleep depth, sleep time, and wake-up conditions to form an individual, and randomly generate multiple individuals to form an initial population. Each individual in the initial population represents an intelligent sleep strategy;

[0034] Determine the fitness function according to the power consumption and performance of the electromechanical device and the user's historical preferences, and calculate the fitness value of each individual;

[0035] Update the initial population according to the fitness value of each individual to generate a sleep population;

[0036] Select the individual with the highest fitness value from the sleep population and decode it into an intelligent sleep strategy, including sleep depth, sleep time, and wake-up conditions.

[0037] As an optional implementation manner, the generation logic of the sleep population includes:

[0038] Determine the selection probability of each individual according to its fitness value and sort them in descending order to select parental individuals;

[0039] Perform crossover operations on the selected parental individuals to generate offspring individuals;

[0040] Perform mutation operations on the offspring individuals, and merge the mutated offspring individuals with the parental individuals to obtain a new population;

[0041] Repeat calculating the fitness value of each individual in the new population to update the new population until the iteration times are met, and obtain a dormant population.

[0042] As an optional implementation manner, the logic of the holographic projection display includes:

[0043] Simulate and record the power consumption changes and performance fluctuations of electromechanical devices under different intelligent dormancy strategies, and generate operation simulation data of electromechanical devices;

[0044] Construct a holographic projection model of the electromechanical device through 3D modeling, and map the operation simulation data of the electromechanical device into the holographic projection model of the electromechanical device to display the operation simulation state of the electromechanical device under different intelligent dormancy strategies;

[0045] Add an interaction function to the holographic projection model of the electromechanical device, and monitor user interaction operations to update the holographic projection display.

[0046] In a second aspect, the present application provides a method for adaptively dormant electromechanical devices driven by user activity. The method includes: obtaining perception data, and completing the automatic networking of sensor nodes and the transmission of perception data through a self-organization mechanism;

[0047] When an abnormality occurs in the sensor node, trigger network topology reconstruction;

[0048] Extract the perception features of the perception data to dynamically evaluate user activity;

[0049] Comprehensively judge whether to trigger the projection interaction mode according to the activity interval, activity trend, and the duration of the user in the low-activity state;

[0050] When the electromechanical device is in the projection interaction mode, continuously monitor the duration of the electromechanical device in the projection interaction mode;

[0051] Judge whether it is necessary to switch the dormancy mode according to the duration and task priority;

[0052] When it is necessary to switch the dormancy mode, comprehensively consider the power consumption and performance of the electromechanical device and the user's historical preferences, formulate and optimize an intelligent dormancy strategy through a genetic algorithm, and determine the dormancy depth and dormancy time of the electromechanical device;

[0053] The operating simulation states of the electromechanical device under different intelligent sleep strategies are displayed through holographic projection, including the power consumption changes and performance fluctuations of the electromechanical device;

[0054] Generate a wake-up signal for the electromechanical device and perform safety protection for the electromechanical device, and cooperate with distributed power supply and energy recovery to manage the power supply of the electromechanical device.

[0055] Compared with the prior art, the beneficial effects of this application are: An energy-saving system with organic coordination is constructed by integrating a perception acquisition module, an edge processing module, and a sleep decision module, realizing the energy-saving management of electromechanical devices. Information is interacted and coordinated in real time among modules, and it can accurately perceive the operating state of electromechanical devices, user behaviors, and environmental changes, so as to judge whether to trigger the projection interaction mode and sleep mode of the electromechanical device, in order to improve the intelligent level, energy utilization efficiency, and user experience of the electromechanical device, and achieve energy saving of the electromechanical device.

[0056] The perception acquisition module constructs a stable and reliable sensor network through node identification and self-organization mechanism of geographical locations, not only realizing the efficient acquisition and transmission of perception data, but also being able to quickly trigger network topology reconstruction when sensor nodes are abnormal, ensuring the continuity and accuracy of data transmission, and providing comprehensive and real-time data support for the entire energy-saving system.

[0057] The edge processing module can accurately extract the features of perception data, dynamically evaluate user activity, and accurately judge whether to trigger the projection interaction mode according to the activity interval, activity trend, and duration, and generate personalized projection content according to the user's historical preferences, improving the interaction experience between the user and the electromechanical device, and enabling the user to obtain more convenient and personalized services when operating the electromechanical device.

[0058] When the electromechanical device is in the projection interaction mode, the sleep decision module continuously monitors user activity, combines task priorities to judge whether to switch to the sleep mode, and formulates and optimizes intelligent sleep strategies by comprehensively weighing the power consumption, performance, and user historical preferences of the electromechanical device through genetic algorithms, determining the optimal sleep depth and sleep time, thus effectively realizing the intelligent sleep of the electromechanical device, reducing energy consumption, and extending the service life of the electromechanical device. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of this application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of this application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. Among them:

[0060] Figure 1 It is the system flow chart of the user activity-driven electromechanical equipment adaptive energy-saving system provided by the embodiments of the present application;

[0061] Figure 2 It is the self-organization mechanism flow chart of the user activity-driven electromechanical equipment adaptive energy-saving system provided by the embodiments of the present application;

[0062] Figure 3 It is the sleep mode conversion judgment logic diagram of the user activity-driven electromechanical equipment adaptive energy-saving system provided by the embodiments of the present application;

[0063] Figure 4 It is the intelligent sleep strategy flow chart of the user activity-driven electromechanical equipment adaptive energy-saving system provided by the embodiments of the present application;

[0064] Figure 5 It is the method flow chart of the user activity-driven electromechanical equipment adaptive sleep method provided by the embodiments of the present application. Specific embodiments

[0065] To make the objectives, technical solutions, and advantages of the embodiments of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings of the specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments.

[0066] Embodiment 1

[0067] As Figure 1 shown, it is the system flow chart of the user activity-driven electromechanical equipment adaptive energy-saving system provided by the embodiments of the present application. The user activity-driven electromechanical equipment adaptive energy-saving system includes a sensing and acquisition module, an edge processing module, a sleep decision module, and an energy-saving management module.

[0068] The sensing and acquisition module is used to acquire sensing data and complete the automatic networking of sensor nodes and the transmission of sensing data through a self-organization mechanism. When an abnormality occurs in the sensor nodes, it triggers the reconstruction of the network topology.

[0069] The sensing data includes equipment operation data, environmental data, and network data.

[0070] Specifically, in a certain intelligent park, perception data of users and electromechanical devices is obtained through sensors. Among them, device operation data refers to the operation frequency, operation type (such as click, slide, and input), and operation duration of the recorded users; environmental data refers to the environmental light changes and sound intensity around the electromechanical devices. The environmental light changes are used to infer whether the user is in the usage area of the electromechanical device, while the sound intensity is used to determine whether there are signs of user activities; network data refers to monitoring the network request frequency and request type of the electromechanical device, such as whether large amounts of data are being transmitted, in order to infer whether the user is performing active operations.

[0071] As Figure 2 shown, the logic of the self-organizing mechanism includes:

[0072] Generate a node identifier for each sensor node based on the geographical location of the sensor nodes, and perform neighbor discovery based on the sensor nodes to generate a neighbor table for each sensor node;

[0073] Based on the link quality index and energy reserve in the neighbor table, construct an initial network topology, and elect a parent node according to the node degree, signal strength, and energy reserve to form a hierarchical network topology;

[0074] Construct a routing table according to the hierarchical network topology structure, and automatically select a single-hop or multi-hop transmission mode according to the data priority;

[0075] Determine the data transmission path for each sensor node according to the routing table, and start the hybrid retransmission mechanism;

[0076] During the data transmission process, each sensor node periodically performs network topology scanning, real-time monitors the status information of each sensor node, and determines whether each sensor node has an abnormality to send an abnormal signal to trigger network topology reconstruction.

[0077] To achieve the automatic networking of sensor nodes, a unique identifier is assigned to each sensor node and a neighbor table is established, laying a foundation for subsequent network topology construction. Combining the geographical location of sensor nodes, a unique node identifier is generated for each sensor node. This node identifier not only includes the location encoding of the sensor node but also incorporates timestamp information to ensure uniqueness and traceability in a large-scale dynamic network environment. After each sensor node completes the generation of its node identifier, each sensor node periodically (such as every 1 minute) broadcasts a beacon frame containing its own node identifier, geographical location, energy reserve (remaining energy value), signal strength, and working status (including idle, busy, and abnormal). After other sensor nodes receive the beacon frame, they check the signal strength in the beacon frame. If the signal strength is greater than a preset strength threshold, the sensor node is considered a neighbor node of its own. Each sensor node records the information of the neighbor node (node identifier, geographical location, energy reserve, signal strength, and working status) in its own neighbor table and sorts the sensor nodes in the neighbor table according to the signal strength. In this way, each sensor node can quickly discover the surrounding neighbor nodes and receive the information of the neighbor nodes, thereby generating the neighbor table of each sensor node. The sensor node can master the information of the surrounding neighbor nodes, providing a basis for subsequent network organization and data transmission, thus realizing the mutual discovery and preliminary communication ability between sensor nodes and providing data support for network topology construction.

[0078] To optimize the data transmission path, it is necessary to construct a hierarchical network topology and elect a parent node to ensure the efficient operation of the sensor network. After obtaining the neighbor table, calculate the link quality index of each neighbor node. The link quality index can be obtained through signal strength and signal-to-noise ratio. According to the link quality index and energy reserve of the neighbor nodes, construct an initial network topology through the minimum spanning tree algorithm. During the execution of the minimum spanning tree algorithm, preferentially select neighbor nodes with high link quality index and sufficient energy reserve for connection. The initial network topology constructed in this way can ensure the stability and sustainability of the network, thereby determining the connection relationship and data transmission path between sensor nodes and providing a framework for subsequent data transmission.

[0079] Electing a parent node can construct a hierarchical network structure and improve the management and data transmission efficiency of the sensor network. The parent node is responsible for coordinating and forwarding data. A reasonable selection of the parent node can enhance the overall performance of the sensor network. Therefore, after obtaining the initial network topology, each sensor node calculates its own node degree, signal strength, and energy reserve. The node degree refers to the number of neighbor nodes directly connected to the sensor node. Here, a parent node election function can be defined, such as , where represents the parent node score, represents the node degree of the sensor node, Indicates the signal strength of the sensor node, Indicates the percentage of energy reserve of the sensor node, 、 and respectively indicate the influence degree of the node degree of the sensor node on the parent node election, the influence degree of the signal strength of the sensor node on the parent node election, and the influence degree of the percentage of energy reserve of the sensor node on the parent node election. 、 and The value ranges of are between 0 and 1, and

[0080] Each sensor node calculates the parent node scores of itself and its neighbor nodes according to the parent node election function. Among them, the sensor node with the highest parent node score is elected as the parent node, so as to determine the parent node of each subnet. The parent node broadcasts its own information to the neighbor nodes, thus forming a hierarchical network topology and constructing a well-defined network structure, which is convenient for the network management and data routing of sensors, and improves the reliability and scalability of the sensor network.

[0081] Each sensor node calculates the shortest path to other sensor nodes according to the hierarchical network topology through the distance vector routing algorithm. Each sensor node records the information of the shortest path in the routing table, including the information of the next-hop node, transmission delay, number of hops, and energy consumption, which provides path planning for data transmission and ensures the orderly transmission of data in the sensor network.

[0082] Data with different data priorities have different requirements for transmission timeliness and reliability. Selecting the transmission mode according to the data priority can optimize the utilization of network resources and ensure the transmission of key data. Define the data priorities of different types of data. For example, the abnormal data of the sensor node is of high priority, and the relevant monitoring data of the electromechanical equipment is of low priority. When the sensor node has data to transmit, it makes a judgment according to the data priority. If it is high-priority data, it preferentially selects single-hop transmission and directly sends the data to the target sensor node. If it is low-priority data, it selects multi-hop transmission and forwards the data through intermediate sensor nodes. When selecting multi-hop transmission, it selects the shortest data transmission path according to the routing table, which improves the utilization efficiency of the sensor network resources and ensures the fast transmission of high-priority data.

[0083] To ensure that data can be accurately transmitted along the planned path, the hybrid retransmission mechanism can improve the reliability of data transmission. The sensor node determines the next-hop node for data transmission according to the routing table. When transmitting data, the hybrid retransmission mechanism is adopted, combining forward error correction coding and selective automatic repeat request protocol. Before sending data, forward error correction coding is performed on the data to add redundant information. After receiving the data, the receiving node first attempts to correct the error using forward error correction coding. If the error cannot be corrected, the sending node is requested to retransmit the data through the selective automatic repeat request protocol, ensuring reliable data transmission in a complex network environment and reducing data loss and errors.

[0084] The node status in the sensor network can change at any time, such as energy depletion and signal weakening. Each sensor node periodically (e.g., every 5 minutes) sends a probe message to its neighbor nodes. The probe message contains the node identifier and timestamp of the sensor node. After receiving the probe message, the neighbor node immediately replies with a response message. The sending sensor node judges the status information of the neighbor node according to the reception situation and signal strength of the response message. Each sensor node records information such as the energy reserve, signal strength, link quality, working status, and transmission success rate of the monitored neighbor nodes, and real-time monitors the operating status of the sensor nodes in the sensor network, providing data support for judging whether a sensor node is abnormal.

[0085] Thresholds are set for each monitoring index (energy reserve, signal strength, link quality, working status, and transmission success rate). The set thresholds are determined according to the actual situation. The sensor node compares the monitoring index with the set threshold. If a certain monitoring index is greater than the set threshold, it is judged that the sensor node is abnormal. When a sensor node judges that itself or a certain neighbor node is abnormal, it sends an abnormal signal to the sensor network, triggering network topology reconstruction to timely respond to abnormal situations in the sensor network and ensure the continuous and stable operation of the sensor network.

[0086] The logic of network topology reconstruction includes:

[0087] Receiving the abnormal signal, transmitting the abnormal data of the abnormal sensor node to the sink node through the network and integrating it to form an abnormal data set;

[0088] Judging the abnormal type of the sensor node based on the abnormal data set. The abnormal types include local node failure, local network failure, and global network failure;

[0089] Triggering a multi-level reconstruction mechanism according to the abnormal type of the sensor node and verifying the reconstruction effect to optimize the network topology reconstruction.

[0090] The sink node is responsible for collecting and managing the information of the entire sensor network, and transmitting the abnormal data to the sink node for unified analysis and processing to comprehensively understand the abnormal situation of the sensor network. The abnormal sensor nodes encapsulate their own abnormal data (including their own node identifiers, geographical locations, and monitored indicators with abnormalities) into abnormal data packets, and send the abnormal data packets to the sink node according to the routing table. During the transmission process, if the path in the routing table is unavailable, the abnormal sensor nodes can find a new transmission path through broadcasting. When the sink node receives the abnormal data packet, it stores the abnormal data packet in the database and classifies and organizes it according to the node identifier and the monitored indicator with abnormalities to form an abnormal data set, providing a data basis for subsequent abnormal type judgment.

[0091] Different abnormal types require different network topology reconstructions. Accurately judging the abnormal type can take targeted reconstruction measures to improve the reconstruction efficiency and network recovery ability. The sink node analyzes the abnormal data set and classifies the abnormal type according to the distribution of the abnormal sensor nodes. If one or a few isolated sensor nodes have abnormalities, and the abnormalities of these sensor nodes do not significantly affect the normal communication and data transmission of their neighbor nodes, and at the same time, through checking the routing table, it is found that only the paths related to these sensor nodes are abnormal while the transmission paths of most other subnets are normal, it is determined as local node failure; if multiple adjacent sensor nodes in a certain area simultaneously have abnormalities, and the link quality index of the subnet where these sensor nodes are located generally decreases, there are a large number of unreachable paths in the subnet routing, and the communication between this subnet and other subnets is also seriously affected, but there are still some subnets operating normally, it is determined as local network failure at this time; when it shows that a large number of sensor nodes simultaneously have abnormalities, the link quality index of the entire network seriously decreases, the routing table completely fails, and the sensor nodes cannot communicate normally with each other, and the overall function of the network is paralyzed, it is determined as global network failure, clarifying the severity and scope of the abnormalities of the sensor nodes and providing a basis for selecting an appropriate reconstruction mechanism.

[0092] Different types of exceptions require reconstruction mechanisms of different intensities and methods. When it is determined that a local node fails, the sink node notifies the neighbor nodes of the sensor node with the exception, selects the sensor nodes with sufficient energy reserves and high link quality index among the neighbor nodes to replace the sensor node with the exception, and updates the neighbor table and routing table of the sensor node; when it is determined that a local network fails, the sink node re-plans the topology of the local network, based on the boundary nodes of the failed subnet, allocates more standby nodes from the standby node pool and supplements them into the failed subnet to form a new subnet structure, and at the same time re-plans the routing path between the failed subnet and other normal subnets to ensure the overall connectivity of the network; when it is determined that the global network fails, the sink node initiates network topology initialization, including re-assigning node identifiers to all sensor nodes, performing neighbor discovery and constructing a hierarchical network topology, and preferentially restoring the core data transmission link, thereby restoring the normal function of the sensor network and improving the network performance and stability.

[0093] Verify that the reconstruction effect can ensure that the network returns to normal and the performance is optimized. After the reconstruction is completed, the sink node verifies the reconstruction effect by monitoring the performance metrics of the entire network (such as transmission delay, packet loss rate, and throughput). If the performance metrics do not meet the expectations, the sink node can adjust the network topology reconstruction and perform re-reconstruction.

[0094] The edge processing module is used to extract the perception features of the perception data to dynamically evaluate the user activity, and comprehensively judge whether to trigger the projection interaction mode according to the activity interval, activity trend, and the duration of the user in the low activity state.

[0095] The evaluation logic of user activity includes:

[0096] Extract the perception features of the perception data, and the perception features include device operation frequency, light change rate, and network request frequency;

[0097] Calculate the activity score according to the perception features to divide the activity interval;

[0098] Calculate the change rate of the activity score to judge the activity trend.

[0099] Extract the device operation frequency from the device operation data, such as the number of user operations within a unit of time (e.g., 10 minutes), which can directly reflect the degree of interaction between the user and the electromechanical device; extract the light change rate from the environmental data, such as the standard deviation of the environmental light intensity within a time window of 30 seconds, which can assist in determining whether the user is in the area where the electromechanical device is used; extract the network request frequency from the network data, such as monitoring the number of network requests initiated by the electromechanical device per second, which can reflect the usage activity of the electromechanical device, and map the values corresponding to each perception feature to between 0 and 1, so as to convert the original perception data into quantifiable perception features for subsequent calculation and analysis.

[0100] The influence degrees of different perception features on user activity are different. Reflect the importance differences of different perception features through weight assignment, so as to calculate the activity score more accurately. Assign weights to each perception feature and calculate the activity score through weighted summation, where the sum of the weights of all perception features is 1; divide the activity interval through a preset activity threshold. For example, set the activity thresholds to 0.4 and 0.7. The activity intervals include a low activity interval (e.g., the activity score is less than 0.4), a medium activity interval (e.g., the activity score is greater than or equal to 0.4 and less than 0.7), and a high activity interval (e.g., the activity score is greater than or equal to 0.7). Match the calculated activity score with the activity interval to determine the activity interval to which the current user activity belongs, providing key information for judging whether to trigger the projection interaction mode.

[0101] At the same time, it is necessary to predict the change trend of user activity to avoid mis-triggering the projection interaction mode due to short-term fluctuations. Calculate the change rate of the activity score, such as , where represents the change rate of the activity score, represents the activity score of the user at time represents the activity score of the user at time represents the time interval between time and time (e.g., 5 minutes). Thus, it can reflect the user activity trend. The activity trend is an upward trend (e.g., ), a downward trend (e.g., ), and a stable trend (e.g.,

[0102] ), providing a basis for judging the direction of user activity, improving the accuracy of active state determination, and reducing the mis-trigger probability of the projection interaction mode.

[0103] Determine whether the user activity level is in the low-activity range and determine the trend of the user's activity level to obtain whether the user is in a low-activity state;

[0104] Record the duration for which the user is in a low-activity state to obtain whether to trigger the projection interaction mode;

[0105] If the projection interaction mode is triggered, obtain the user's historical preferences based on the device operation data to generate personalized projection content.

[0106] When the user activity level is in the low-activity range and the activity trend shows a downward or stable trend, it is determined that the user is in a low-activity state, so as to accurately judge whether the user is in a low-activity state, avoid false triggers caused by short-term low activity, improve the reliability of the determination, and provide a basis for subsequent trigger decisions.

[0107] At the same time, it is necessary to decide whether to trigger the projection interaction mode according to the duration of the low-activity state. When the user is in a low-activity state, start the timing mechanism, record the duration for which the user is in a low-activity state, set a basic projection threshold (such as 15 minutes), and then adjust it according to the historical behavior data to obtain the projection duration threshold, such as , where represents the projection duration threshold, represents the basic projection threshold, represents the number of times in the low-activity state, represents the total number of times the user operates the electromechanical device. When the duration for which the user is in a low-activity state is greater than or equal to the projection duration threshold, trigger the projection interaction mode; otherwise, do not trigger the projection interaction mode, and continuously monitor the user activity level to decide whether to enter the projection interaction mode, so as to achieve energy conservation of the electromechanical device and optimization of the user experience.

[0108] Extract the user's historical operation records from the device operation data, such as the frequently used functions and click frequencies, etc. Analyze the data of the user's historical operation records through machine learning algorithms to mine the user's historical preferences. According to the user's historical preferences, screen relevant content from the content library to produce personalized projection content, so as to provide interactive content that meets the user's needs and enhance the user's acceptance and satisfaction of the projection interaction mode.

[0109] The sleep decision module is used to continuously monitor the duration for which the electromechanical device is in the projection interaction mode when the electromechanical device is in the projection interaction mode. Judge whether it is necessary to switch to the sleep mode according to the duration and the task priority. When it is necessary to switch to the sleep mode, comprehensively weigh the power consumption and performance of the electromechanical device and the user's historical preferences, formulate and optimize the intelligent sleep strategy through the genetic algorithm, determine the sleep depth and sleep time of the electromechanical device, and display the operation simulation state of the electromechanical device under different intelligent sleep strategies through holographic projection, including the power consumption change and performance fluctuation of the electromechanical device.

[0110] As Figure 3 shown, the conversion judgment logic for the sleep mode includes:

[0111] Obtain the operation tasks of the electromechanical device and evaluate the task priorities;

[0112] Monitor the continuous duration of the electromechanical device in the projection interaction mode;

[0113] Judge whether it is necessary to convert to the sleep mode according to the operation task priorities and the continuous duration of the electromechanical device in the projection interaction mode. If it is necessary to convert to the sleep mode, generate a sleep signal.

[0114] Different operation tasks have different requirements and impacts on the operation of the electromechanical device. Evaluating the task priorities helps to comprehensively consider the operation status of the electromechanical device and the importance of the operation tasks when judging whether to enter the sleep mode, avoiding affecting the execution of important tasks due to entering the sleep mode. Obtain the list of all currently running operation tasks from the operating system of the electromechanical device, and assign task priorities to each operation task according to the type, importance, and urgency of the operation task. For example, device critical tasks (such as calibration and data transmission of the electromechanical device) can be set as high priorities, device backup tasks (such as file backup and logging) can be set as medium priorities, and background non-essential tasks (such as standby status and idle tasks) can be set as low priorities to generate a task priority table and mark the minimum running duration of each operation task, so as to ensure that device critical tasks are not affected by sleep and maintain the normal operation of the electromechanical device.

[0115] The continuous duration of the user in the projection interaction mode is one of the important factors for judging whether to convert to the sleep mode. When the electromechanical device is in the projection interaction mode, start the timer to time, continuously monitor and record the continuous duration of the electromechanical device in the projection interaction mode, set a basic sleep threshold (such as 10 minutes), and adjust it according to the user's historical behavior to obtain a sleep duration threshold. For example , where represents the sleep duration threshold, represents the basic sleep threshold, represents the number of times the user has recently used the electromechanical device to make the electromechanical device in the projection interaction mode. When the continuous duration of the electromechanical device in the projection interaction mode is greater than or equal to the sleep duration threshold and there are no high-priority tasks in the current operation tasks of the electromechanical device, it is determined that it is necessary to convert to the sleep mode to ensure the energy saving of the electromechanical device and generate a sleep signal. Otherwise, it is not necessary to convert to the sleep mode, and continuously monitor the continuous duration of the electromechanical device in the projection interaction mode. According to the actual operation situation of the electromechanical device and the continuous duration in the projection interaction mode, accurately judge whether it is necessary to convert to the sleep mode and generate a sleep signal in a timely manner, providing a trigger condition for formulating an intelligent sleep strategy in the future.

[0116] As Figure 4 shown, the intelligent sleep strategy includes:

[0117] Receiving a sleep signal, encoding the sleep depth, sleep time, and wake-up condition to form an individual, and randomly generating several individuals to form an initial population, where each individual in the initial population represents an intelligent sleep strategy;

[0118] Determining a fitness function based on the power consumption and performance of the electromechanical device and the user's historical preferences, and calculating the fitness value of each individual;

[0119] Updating the initial population according to the fitness value of each individual to generate a sleep population;

[0120] Selecting the individual with the highest fitness value from the sleep population and decoding it into an intelligent sleep strategy, including the sleep depth, sleep time, and wake-up condition.

[0121] The genetic algorithm requires an initial population as the basis for optimization. By encoding and randomly generating individuals, it can generate a variety of different intelligent sleep strategies, providing diversity for subsequent optimization. When a sleep signal is received, the sleep depth, sleep time, and wake-up condition are encoded. For example, the sleep depth can be divided into light sleep, moderate sleep, and deep sleep, represented by the numbers 1, 2, and 3 respectively. The sleep time is in minutes and represented by an integer. The wake-up condition can be encoded according to different events (such as user operations and timed wake-up). 1 represents user operation, 2 represents timed wake-up. Randomly generate multiple individuals containing the encoded sleep depth, sleep time, and wake-up condition. Each individual represents an intelligent sleep strategy. Combine these individuals together to form an initial population. For example, an initial population of 100 individuals is generated. The encoding example of each individual is as follows: [2, 60, 1]. The meaning of this encoding example is that the current intelligent sleep strategy is moderate sleep, the sleep time is 60 minutes, and the wake-up condition is user operation. A diverse initial population is created, providing a rich set of candidate strategies for the optimization of the intelligent sleep strategy and avoiding falling into a local optimum.

[0122] Calculating the fitness function to quantify the quality of each intelligent sleep strategy represented by an individual. By comprehensively considering the power consumption and performance of the electromechanical device and the user's historical preferences, the optimal sleep strategy that is both energy-saving and meets the user's needs can be found. The fitness function comprehensively considers the power consumption and performance of the electromechanical device and the user's historical preferences. For example, the expression of the fitness function is as follows:

[0123] ;

[0124] In the formula, represents the fitness value, represents the power consumption weight coefficient, Represents the power consumption value of the electromechanical device, Represents the performance weight coefficient, Represents the performance loss rate of the electromechanical device, Represents the preference weight coefficient, Represents the score of the user's historical preference.

[0125] It should be noted that: the power consumption weight coefficient refers to the degree of influence of the power consumption of the electromechanical device on the fitness value, and the performance weight coefficient refers to the degree of influence of the performance of the electromechanical device on the fitness value, and the preference weight coefficient refers to the degree of influence of the user's historical preference on the fitness value, 、 and The value ranges from 0 to 1, and ; the power consumption value of the electromechanical device 、the performance loss rate of the electromechanical device and the score of the user's historical preference are obtained by each individual simulating the operation of the electromechanical device under this intelligent sleep strategy according to its encoded sleep depth, sleep time, and wake-up conditions, so as to obtain the power consumption value and performance loss rate of the electromechanical device and the score of the user's historical preference.

[0126] According to the fitness function, the fitness value of each individual is obtained, thereby measuring the quality of this individual as an intelligent sleep strategy, providing a basis for subsequent population selection and optimization.

[0127] Through the iterative optimization of the genetic algorithm, the intelligent sleep strategy is continuously screened and improved, so that the individuals in the initial population gradually tend to the optimal solution, thereby finding the most suitable intelligent sleep strategy for the electromechanical device. According to the fitness value of each individual, operations such as selection, crossover, and mutation are used to update the initial population. Individuals with higher fitness values are selected as parent individuals, and offspring individuals are generated through crossover and mutation operations. The offspring individuals are merged with the parent individuals to form a new population. The above process is repeated until the preset number of iterations (such as 100 times) is satisfied to obtain the sleep population. Through iterative optimization, the intelligent sleep strategy in the initial population has been continuously improved, making it contain a better intelligent sleep strategy.

[0128] After optimization by the genetic algorithm, the intelligent sleep strategy represented by the individual with the highest fitness value has achieved the relative optimality in terms of power consumption, performance and user preference. Taking it as the final intelligent sleep strategy can achieve a balance between energy saving, performance and meeting user needs. The individual with the highest fitness value is found in the sleep population, and the encoding of the individual is decoded to obtain the corresponding sleep depth, sleep time and wake-up conditions to form the final intelligent sleep strategy. The final intelligent sleep strategy is determined, which provides specific parameters and conditions for the sleep operation of electromechanical equipment.

[0129] The generation logic of dormant population includes:

[0130] Determine the probability of each individual being selected according to the fitness value of each individual and arrange them in descending order to select the parent individuals;

[0131] Perform crossover operation on the selected parent individuals to generate offspring individuals;

[0132] Perform mutation operation on offspring individuals, merge the mutated offspring individuals with the parent individuals to obtain a new population;

[0133] Repeatedly calculate the fitness value of each individual in the new population and update the new population until the number of iterations is met to obtain a dormant population.

[0134] Individuals with high fitness values ​​will produce excellent offspring individuals. Selecting parent individuals according to probability can ensure that the population is optimized in a better direction. The fitness value of each individual is calculated as the ratio of the total fitness value of the initial population. As the probability of the individual being selected, the individuals are sorted from high to low according to the probability of being selected. According to the preset selection ratio, the individuals with the highest ranking are selected as parent individuals, thereby screening out relatively excellent individuals as parent individuals (for example, 50), which provides a good foundation for subsequent crossover and mutation operations.

[0135] The crossover operation can combine the good genes of parent individuals to generate new individuals, increase the diversity and adaptability of the population, randomly select two parent individuals, randomly select a crossover point in the code of each parent individual, exchange the parts after the crossover point of the two parent individuals, generate two offspring individuals, repeat the above process until a sufficient number of offspring individuals (for example, 50) are generated. For example, select parent individuals [2,60,1] and [3,90,2], and the randomly selected crossover point is in the second position. After crossover, offspring individuals [2,90,2] and [3,60,1] are generated. Through the crossover operation, new individuals are generated, enriching the initial population, and providing more possibilities for finding a better intelligent hibernation strategy.

[0136] The mutation operation can introduce new genes, avoid the initial population falling into local optimal solutions, and ensure the evolution ability of the initial population. For each offspring individual, a coding bit is randomly selected for mutation with a certain mutation probability, and the mutated offspring individual is merged with the parent individual to generate a new population. For example, the offspring individual [2, 90, 2] mutates with a mutation probability of 0.01, changing the dormancy depth from 2 to 3, obtaining the mutated offspring individual [3, 90, 2]. The mutated offspring individual is merged with the parent individual to obtain a new population of 100 individuals. Through the mutation operation, the diversity of the population is increased, and the possibility of finding the global optimal solution is improved.

[0137] By iteratively updating the population multiple times, the intelligent dormancy strategy can be continuously optimized, making the individuals in the population gradually tend to the optimal solution. For each individual in the new population, the fitness value of each individual is recalculated. According to the new fitness values, the above selection, crossover, and mutation operations are repeated to update the population. This process is continuously repeated until the preset number of iterations (such as 100 times) is reached, obtaining the final dormant population. Through iterative optimization, a relatively optimal combination of intelligent dormancy strategies is found, providing a rich set of candidate solutions for finally selecting the best intelligent dormancy strategy.

[0138] The logic of holographic projection display includes:

[0139] Simulate and record the power consumption changes and performance fluctuations of electromechanical devices under different intelligent dormancy strategies to generate operation simulation data of the electromechanical devices;

[0140] Construct a holographic projection model of the electromechanical device through 3D modeling, and map the operation simulation data of the electromechanical device into the holographic projection model of the electromechanical device to display the operation simulation states of the electromechanical device under different intelligent dormancy strategies;

[0141] Add interactive functions to the holographic projection model of the electromechanical device and monitor user interaction operations to update the holographic projection display.

[0142] By simulating the operation states under different intelligent dormancy strategies, it is possible to intuitively understand the impact of each intelligent dormancy strategy on the power consumption and performance of the electromechanical device, providing data support for users to select appropriate intelligent dormancy strategies. For the intelligent dormancy strategies represented by each individual in the dormant population, simulate the operation states of the electromechanical device under different intelligent dormancy strategies, record the power consumption changes and performance fluctuations during the dormancy and wake-up processes of the electromechanical device, and organize these data into operation simulation data, thereby obtaining the operation information of the device under different intelligent dormancy strategies, providing a data basis for holographic projection display.

[0143] Holographic projection display can enable users to understand the impact of different intelligent sleep strategies on electromechanical devices in an intuitive way, helping users better understand and select appropriate intelligent sleep strategies. Use 3D modeling software to construct a holographic projection model based on the actual structure and appearance of the electromechanical device, and associate the power consumption changes and performance fluctuation data in the operation simulation data with the holographic projection model. For example, use different colors to represent different power consumption levels and the length of the line to represent performance fluctuations. Display the holographic projection model mapped with the operation simulation data through a holographic projection device to present the operation simulation state of the electromechanical device under different intelligent sleep strategies, visually showing the operation state of the device under different intelligent sleep strategies and improving the efficiency of users' understanding and selection of intelligent sleep strategies.

[0144] The curve function of the power consumption change is as follows:

[0145] ;

[0146] In the formula, represents the total power consumption of the electromechanical device at time , represents the power consumption of the electromechanical device in the non-sleep mode, represents the sleep depth coefficient, represents the power consumption of the electromechanical device in the sleep mode.

[0147] It should be noted that: the sleep depth coefficient refers to the power consumption ratio of the electromechanical device in the sleep mode, and the value range of is between 0 and 1. represents that the electromechanical device is completely non-sleeping, represents that the electromechanical device is completely sleeping; the power consumption of the electromechanical device in the non-sleep mode and the power consumption of the electromechanical device in the sleep mode are obtained from the specification of the electromechanical device. Among them,

[0148] The curve function of the performance fluctuation is as follows:

[0149] ;

[0150] In the formula, represents the performance of the electromechanical device at time , represents the maximum performance of the electromechanical device in the non-sleep mode, represents the performance loss rate of the electromechanical device.

[0151] It should be noted that: the performance loss rate of the electromechanical device It refers to the percentage of performance degradation of electromechanical equipment in the sleep mode. The value ranges from 0 to 1. It indicates that there is no performance loss of the electromechanical equipment. It indicates that the performance of the electromechanical equipment is completely lost. , where represents the performance of the electromechanical equipment in the sleep mode, and the performance loss rates under different sleep depths are also different.

[0152] Adding an interactive function can enhance the user's sense of participation and experience, enabling the user to obtain more detailed information according to their own needs and preferences, so as to make more appropriate decisions. Adding an interactive function to the holographic projection model, such as gesture recognition and voice control, etc., the user can select different intelligent sleep strategies and view detailed operation simulation data through gestures or voice operations, etc., and monitor the user's interactive operations in real time, and update the holographic projection display according to the operation content. For example, when the user selects another intelligent sleep strategy, the operation simulation state of the electromechanical equipment under this intelligent sleep strategy is displayed, improving the interactivity between the user and the holographic projection display, enabling the user to more actively understand the characteristics of different intelligent sleep strategies, and optimizing the user experience.

[0153] The energy-saving management module is used to generate the wake-up signal of the electromechanical equipment and perform the safety protection of the electromechanical equipment, and manage the power supply of the electromechanical equipment through distributed power supply and energy recovery in a coordinated manner.

[0154] When the wake-up condition is met, generate the wake-up signal of the electromechanical equipment. After receiving the wake-up signal, the electromechanical equipment immediately checks the permissions of the current user or operation, and restricts or allows the corresponding operations according to the preset permission policy. At the same time, during the operation of the electromechanical equipment, monitor the activities such as calls and network connections in the electromechanical equipment in real time, so as to identify and prevent the abnormal behaviors and attacks of the user, thereby ensuring the security and data integrity of the electromechanical equipment during operation.

[0155] The power supply management logic of the electromechanical equipment includes:

[0156] Real-time monitor the status data of the distributed power supply and energy recovery to obtain the power supply status;

[0157] Generate the power supply priority according to the operation tasks and power consumption of the electromechanical equipment;

[0158] Optimize the power supply of the electromechanical equipment based on the power supply status and power supply priority.

[0159] Understanding the real-time status of distributed power sources and energy recovery is the basis for the reasonable management of the power supply of electromechanical devices. Through sensors and monitoring circuits, the status data of the output voltage, current, and remaining power (for batteries) of distributed power sources (such as batteries, solar panels, and small generators), as well as the status data of the energy recovery rate and the total current recovered energy of energy recovery, are obtained in real time. Based on these status data, the power supply status is evaluated, including whether the distributed power source is working properly and whether the energy is sufficient, etc., accurately grasping the real-time situation of the power supply, providing a basis for subsequent power supply management and optimization.

[0160] Different operating tasks in electromechanical devices have different requirements and importance for power supply. Reasonably allocating power supply priorities can ensure the normal operation of critical tasks and improve energy utilization efficiency. Obtain information on the current operating tasks of electromechanical devices, including task types and task priorities. At the same time, based on the power consumption monitoring data of electromechanical devices, determine the power consumption requirements of each operating task, and generate power supply priorities according to task priorities and power consumption requirements. For example, emergency tasks and critical tasks with high power consumption are given priority in power supply, and the power supply priority of background non-essential tasks can be reduced when energy is scarce, thus clarifying the power supply priorities of each operating task in electromechanical devices and providing a basis for optimizing the power supply management of electromechanical devices.

[0161] According to the power supply status and power supply priorities, adjust the output of the distributed power source and the working mode of energy recovery. When the distributed power source is working properly and the energy is sufficient, give priority to meeting the power supply requirements of high-priority tasks and appropriately supply power to low-priority tasks; when the energy is scarce, give priority to ensuring high-priority tasks, reduce or suspend the power supply of low-priority tasks, optimize the energy distribution, ensure the power supply of critical tasks, and improve energy utilization efficiency.

[0162] Embodiment 2

[0163] As Figure 5 shown, the present application embodiment provides a method flowchart of an electromechanical device adaptive sleep method driven by user activity. The electromechanical device adaptive sleep method driven by user activity includes:

[0164] Obtain sensing data and complete the automatic networking of sensor nodes and the transmission of sensing data through a self-organization mechanism;

[0165] When an abnormality occurs in the sensor node, trigger network topology reconstruction;

[0166] Extract the sensing features of the sensing data to dynamically evaluate user activity;

[0167] Comprehensively judge whether to trigger the projection interaction mode according to the activity interval, activity trend, and the duration of the user being in a low-active state;

[0168] When the electromechanical device is in the projection interaction mode, continuously monitor the duration of the electromechanical device in the projection interaction mode;

[0169] Judge whether it is necessary to switch to the sleep mode according to the duration and task priority;

[0170] When it is necessary to switch to the sleep mode, comprehensively weigh the power consumption and performance of the electromechanical device and the user's historical preferences, formulate and optimize an intelligent sleep strategy through a genetic algorithm, and determine the sleep depth and sleep time of the electromechanical device;

[0171] Display the operation simulation status of the electromechanical device under different intelligent sleep strategies through holographic projection, including the power consumption change and performance fluctuation of the electromechanical device;

[0172] Generate a wake-up signal for the electromechanical device and perform the safety protection of the electromechanical device, and coordinate the power supply of the electromechanical device through distributed power supply and energy recovery.

[0173] Since the principle of solving problems in the method in the embodiments of the present application is similar to that of the above-mentioned system in the embodiments of the present application, the implementation of the method can refer to the implementation of the system, and the repeated parts will not be elaborated.

Claims

1. An electromechanical device adaptive energy-saving system driven by user activity, characterized in that Including: A perception acquisition module, an edge processing module, and a sleep decision module; The perception acquisition module is used to acquire perception data, complete the automatic networking of sensor nodes and the transmission of perception data through a self-organization mechanism, and trigger network topology reconstruction when a sensor node has an abnormality; The edge processing module is used to extract the perception features of the perception data to dynamically evaluate the user activity level, and comprehensively judge whether to trigger the projection interaction mode according to the activity level interval, activity trend, and the duration of the user being in a low-activity state; The sleep decision module is used to continuously monitor the duration of the electromechanical device being in the projection interaction mode when the electromechanical device is in the projection interaction mode, judge whether to switch to the sleep mode according to the duration and task priority, and when it is necessary to switch to the sleep mode, comprehensively weigh the power consumption and performance of the electromechanical device as well as the user's historical preferences, formulate and optimize an intelligent sleep strategy through a genetic algorithm, determine the sleep depth and sleep time of the electromechanical device, and display the operation simulation state of the electromechanical device under different intelligent sleep strategies through holographic projection, including the power consumption change and performance fluctuation of the electromechanical device.

2. The user activity-driven adaptive energy-saving system for electromechanical devices according to claim 1, wherein The perception data includes device operation data, environmental data, and network data, and the logic of the self-organization mechanism includes: Generating a node identifier for each sensor node based on the geographical location of the sensor node, and performing neighbor discovery according to the sensor node to generate a neighbor table for each sensor node; Constructing an initial network topology based on the link quality index and energy reserve in the neighbor table, and electing a parent node according to the node degree, signal strength, and energy reserve to form a hierarchical network topology; Constructing a routing table according to the hierarchical network topology structure, and automatically selecting a single-hop or multi-hop transmission mode according to the data priority; Determining the data transmission path for each sensor node according to the routing table, and starting a hybrid retransmission mechanism; During the data transmission process, each sensor node periodically performs network topology scanning, real-time monitors the status information of each sensor node, and judges whether each sensor node has an abnormality to send an abnormality signal to trigger network topology reconstruction.

3. The user activity-driven electromechanical device adaptive energy-saving system according to claim 2, characterized in that The logic of the network topology reconstruction includes: Receiving the abnormality signal, transmitting the abnormality data of the sensor node with the abnormality through the network to the aggregation node and integrating it to form an abnormality data set; Judging the abnormality type of the sensor node based on the abnormality data set, and the abnormality type includes local node failure, local network failure, and global network failure; Triggering a multi-level reconstruction mechanism according to the abnormality type of the sensor node, and verifying the reconstruction effect to optimize the network topology reconstruction.

4. The user activity-driven electromechanical device adaptive energy-saving system according to claim 3, characterized in that The evaluation logic of the user activity level includes: Extracting the perception features of the perception data, and the perception features include device operation frequency, light change rate, and network request frequency; Calculating an activity score according to the perception features to divide the activity level interval; Calculating the change rate of the activity score to judge the activity trend.

5. The user activity-driven adaptive energy-saving system for electromechanical equipment according to claim 4, wherein, The triggering logic of the projection interaction mode includes: Judging whether the user activity level is in a low-activity interval and judging the activity trend of the user to obtain whether the user is in a low-activity state; Recording the duration of the user being in a low-activity state to obtain whether to trigger the projection interaction mode; If the projection interaction mode is triggered, user historical preferences are obtained based on device operation data to generate personalized projection content.

6. The user activity-driven electromechanical device adaptive energy-saving system according to claim 5, wherein, The conversion judgment logic of the sleep mode includes: Obtain the operation tasks of the electromechanical device and evaluate the task priorities; Monitor the continuous duration of the electromechanical device in the projection interaction mode; Judge whether it is necessary to convert to the sleep mode according to the operation task priority and the continuous duration of the electromechanical device in the projection interaction mode. If it is necessary to convert to the sleep mode, a sleep signal is generated.

7. The user activity-driven adaptive energy-saving system for electromechanical equipment according to claim 6, wherein The intelligent sleep strategy includes: Receive the sleep signal, encode the sleep depth, sleep time, and wake-up conditions to form individuals, and randomly generate multiple individuals to form an initial population. Each individual in the initial population represents an intelligent sleep strategy; Determine the fitness function according to the power consumption and performance of the electromechanical device and user historical preferences, and calculate the fitness value of each individual; Update the initial population according to the fitness value of each individual to generate a sleep population; Select the individual with the highest fitness value from the sleep population and decode it into an intelligent sleep strategy, including sleep depth, sleep time, and wake-up conditions.

8. The user activity-driven adaptive energy-saving system for electromechanical equipment according to claim 7, characterized in that, The generation logic of the sleep population includes: Determine the selection probability of each individual according to the fitness value of each individual and sort them in descending order to select parental individuals; Perform crossover operations on the selected parental individuals to generate offspring individuals; Perform mutation operations on the offspring individuals, and merge the mutated offspring individuals with the parental individuals to obtain a new population; Repeat calculating the fitness value of each individual in the new population to update the new population until the iteration times are met to obtain the sleep population.

9. The user activity-driven adaptive energy-saving system for electromechanical equipment according to claim 8, wherein The logic of the holographic projection display includes: Simulate and record the power consumption changes and performance fluctuations of the electromechanical device under different intelligent sleep strategies to generate the operation simulation data of the electromechanical device; Construct a holographic projection model of the electromechanical device through 3D modeling, and map the operation simulation data of the electromechanical device into the holographic projection model of the electromechanical device to display the operation simulation state of the electromechanical device under different intelligent sleep strategies; Add an interaction function to the holographic projection model of the electromechanical device and monitor user interaction operations to update the holographic projection display.

10. The method for adaptively sleeping an electromechanical device driven by user activity is implemented based on the user activity-driven electromechanical device adaptive energy-saving system according to any one of claims 1-9, and is characterized in that, Include: Obtain sensing data, and complete the automatic networking of sensor nodes and the transmission of sensing data through a self-organization mechanism; When an abnormality occurs in the sensor node, trigger network topology reconstruction; Extract the sensing features of the sensing data to dynamically evaluate user activity; Comprehensively judge whether to trigger the projection interaction mode according to the activity interval, activity trend, and the continuous duration of the user in the low-activity state; When the electromechanical device is in the projection interaction mode, continuously monitor the continuous duration of the electromechanical device in the projection interaction mode; Judge whether it is necessary to convert to the sleep mode according to the continuous duration and task priority; When it is necessary to convert to the sleep mode, comprehensively weigh the power consumption and performance of the electromechanical device and user historical preferences, formulate and optimize an intelligent sleep strategy through a genetic algorithm, and determine the sleep depth and sleep time of the electromechanical device; Display the operation simulation state of the electromechanical device under different intelligent sleep strategies through holographic projection, including power consumption changes and performance fluctuations of the electromechanical device; Generate a wake-up signal for the electromechanical device and perform safety protection for the electromechanical device, and cooperate with the distributed power supply and energy recovery to manage the power supply of the electromechanical device.

Images