Energy supply system combining energy station equipment and wireless energy supply unmanned aerial vehicle

Abstract

The invention relates to the technical field of charging, in particular to an energy supply system combining energy station equipment and a wireless energy supply unmanned aerial vehicle. Information such as residual electricity is received from the distributed sensor; when the residual electricity is greater than the endurance critical point, calculating the mode loss speed of the electric quantity of the sensor, performing endurance prediction in combination with the residual electricity, obtaining a to-be-powered timestamp when the residual electricity is less than or equal to the endurance critical point, and performing identification storage in combination with the obtained pre-stored minimum charging quantity of the distributed sensor; a plurality of places with to-be-powered timestamps and a plurality of minimum charging amounts are obtained, and a plurality of unmanned aerial vehicle energy supply scheduling schemes are randomly generated in combination with scheduled tasks of the wireless energy supply unmanned aerial vehicles and deployment positions of energy station equipment; and performing path shortest constraint optimization to obtain a target energy supply scheduling scheme, and scheduling the wireless energy supply unmanned aerial vehicle to perform wireless power taking and wireless power supply scheduling control. According to the invention, the technical problems of high cost and low efficiency of a sensor network energy supply method are solved.

Description

Technical Field

[0001] This invention relates to the field of charging technology, and in particular to an energy replenishment system that combines an energy station device with a wirelessly powered drone. Background Technology

[0002] With the widespread deployment of sensor networks and IoT devices in outdoor environments, the issue of power supply has become increasingly prominent. However, sensors deployed in the field, at high risk, or on a large scale (such as those for environmental monitoring and industrial IoT) not only face difficulties in frequent manual charging but also pose safety hazards due to personnel moving between hazardous environments. Existing solutions include equipping sensors with energy harvesting devices such as solar panels, but these methods often fail to provide continuous power effectively in environments where they are shaded by trees or enclosed in burial enclosures. Traditional sensor network power supply methods suffer from high costs and low efficiency. Summary of the Invention

[0003] This invention addresses the technical problems of high cost and low efficiency in existing sensor network energy replenishment methods by providing an energy replenishment system that combines an energy station device with a wirelessly powered drone.

[0004] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: In a first aspect, the present invention provides an energy replenishment system combining an energy station device with an unmanned aerial vehicle (UAV) providing power, comprising: The sensor status monitoring module is used to receive information on remaining power, location, environmental status, and operating status from distributed sensors. The sensor battery life prediction module is used to retrieve the mode power loss rate of a sample set of sensors of the same model that meet the environmental conditions and the working conditions and have a service life of less than or equal to 0.2 years when the remaining power is greater than the battery life critical point, and combine it with the remaining power to predict the battery life and obtain the power-on-demand timestamp when the remaining power is less than or equal to the battery life critical point. A power demand sensor identification module is used to obtain the pre-stored minimum charging amount of the distributed sensor and identify and store the location in combination with the power demand timestamp; The power supply scheme generation module is used to obtain several locations with the timestamp of the power supply to be supplied and several minimum charging amounts, and combine the scheduled tasks of the wireless power supply drones and the deployment locations of the energy station equipment to execute random configuration of the scheduling scheme and generate several drone power supply scheduling schemes. The power supply scheme optimization module is used to optimize the path shortest constraint based on the aforementioned UAV power supply scheduling schemes, obtain the target power supply scheduling scheme, and schedule wireless power supply UAVs for wireless power acquisition and wireless power supply scheduling control.

[0005] Optionally, the sensor battery life prediction module is further configured to: when the remaining battery power is less than or equal to the battery life threshold, set the current time as a power supply timestamp, and identify and store the location in conjunction with the minimum charging amount.

[0006] Optionally, the sensor battery life prediction module is further configured to: retrieve the sensor power mode loss rate of a sample set of sensors of the same model that satisfies the environmental state and the operating state, including: performing multi-valued processing on the environmental state to generate an initial environmental state matrix; receiving a pre-configured environmental factor power loss influence weight distribution, weighting the initial environmental state matrix to obtain a modified environmental state matrix; performing multi-valued processing on the operating state to generate an initial operating state matrix; receiving a pre-configured operating factor power loss influence weight distribution, weighting the initial operating state matrix to obtain a modified operating state matrix; and retrieving the sensor power mode loss rate of a sample set of sensors of the same model that satisfies the modified environmental state matrix and the modified operating state matrix.

[0007] The environmental state is multi-valued to generate an initial environmental state matrix, including: Step 1: Configure the first environmental factor fluctuation value through the user terminal; Step 2: Using the first environmental factor fluctuation value as a single variable, collect a set of power fluctuation values ​​for sensors of the same model; Step 3: Perform a centralized value evaluation on the set of power fluctuation values ​​of the same type of sensor to obtain the fitted value of the power fluctuation of the same type of sensor; Step 4: When the power fluctuation fitting value of the same model sensor is less than the power fluctuation consistency deviation predefined by the user terminal, increase the first environmental factor fluctuation value by 2 times and return to Step 2; Step 5: When the fitted value of the power fluctuation of the same model sensor is greater than the power fluctuation consistency deviation, increase the first environmental factor fluctuation value by 0.5 times and return to Step 2; Step 6: When the power fluctuation fitting value of the same model sensor is equal to the power fluctuation consistency deviation, set the first environmental factor fluctuation value as the first environmental factor fluctuation consistency deviation. Based on the consistent deviation of the first environmental factor fluctuation, the first environmental factor is partitioned from small to large according to the pre-configured first environmental factor rated range on the user end, and a unique mapping value is configured for each partition. The first environmental factor multi-value mapping function is constructed and added to the environmental factor multi-value mapping function library. Based on the environmental factor multi-valued mapping function library, the environmental state is multi-valued to generate the initial environmental state matrix.

[0008] The process of receiving the pre-configured weight distribution of the impact of environmental factors on power loss includes: based on a set of environmental factors, collecting a dataset of first environmental factor fluctuation values ​​up to the Nth environmental factor fluctuation value dataset corresponding to the same type of sensor, and a dataset of power fluctuation values; traversing the dataset of first environmental factor fluctuation values ​​up to the Nth environmental factor fluctuation value dataset and performing dimensionless processing to obtain a sequence of first environmental factor fluctuation values ​​up to the Nth environmental factor fluctuation value sequence; performing dimensionless processing on the dataset of power fluctuation values ​​to obtain a sequence of power fluctuation values; using the sequence of power fluctuation values ​​as a reference sequence and the sequence of first environmental factor fluctuation values ​​up to the Nth environmental factor fluctuation value sequence as a comparison sequence, performing grey relational analysis to obtain the correlation degree of first environmental factors up to the Nth environmental factor correlation degree, and comparing each correlation degree with the sum of correlation degrees to generate the weight distribution of the impact of environmental factors on power loss.

[0009] The process of retrieving the sensor power mode loss rate of a sample set of sensors of the same model that meet the environmental and operational conditions includes: collecting the power loss rate of sensors with a baseline service life and a second service life, based on a preset sensor model and the same environmental and operational conditions, wherein the baseline service life is less than or equal to 0.2 years; calculating a first ratio of the power loss rate of the second service life to the power loss rate of the baseline service life, and calculating a second ratio of the second service life to the baseline service life; training a power loss ratio prediction model using the second ratio as input and the first ratio as regression prediction supervision; obtaining the ratio of the service life of distributed sensors to the baseline service life, inputting it into the power loss ratio prediction model to obtain a predicted power loss ratio value, and correcting the initial sensor power mode loss rate fitted to the sample set of sensors of the same model to obtain the sensor power mode loss rate.

[0010] Optionally, the power supply scheme generation module is further configured to: construct a wireless power supply expectation constraint based on the plurality of locations and the plurality of minimum charging amounts; construct a drone recharging location constraint based on the deployment location of the energy station equipment; obtain a list of idle drone locations and a list of idle drones with remaining power at the timestamps of the wireless power supply drones based on the scheduled tasks of the wireless power supply drones, wherein the remaining power of the idle drones is equal to the actual remaining power minus the rated remaining power; and randomly configure the plurality of drone power supply scheduling schemes that satisfy the wireless power supply expectation constraint and the drone recharging location constraint based on the list of idle drone locations and the list of idle drones with remaining power.

[0011] Optionally, the power supply optimization module is further configured to: when the wireless power supply drone enters a first preset range centered on the deployment location of the energy station equipment, activate the medium-range wireless power transmission module of the energy station equipment and activate the first wireless energy receiving unit of the wireless power supply drone to charge the wireless power supply drone; when the wireless power supply drone enters a second preset range centered on the location of the distributed sensor, activate the short-range wireless power supply module of the wireless power supply drone and activate the second wireless energy receiving unit of the distributed sensor to draw power from the wireless power supply drone.

[0012] By implementing this invention, it is possible to receive residual power, location, environmental status, and operating status from distributed sensors, ensuring the real-time nature and accuracy of the data. By implementing this invention, when the remaining power is greater than the endurance critical point, the mode power loss rate of a sample set of sensors of the same model that meet the environmental and operating conditions and have a service life of less than or equal to 0.2 years can be retrieved. The endurance can then be predicted in conjunction with the remaining power to obtain the time stamp of the power-on-demand point when the remaining power is less than or equal to the endurance critical point. This allows for the prediction of the remaining power by combining environmental factor weights (determined through grey relational analysis) and operating conditions, and utilizing the mode power loss rate of historical samples, thereby reducing the prediction error rate. By implementing this invention, the minimum charging amount of the pre-stored distributed sensor can be obtained, and the location can be identified and stored in combination with the power-on-demand timestamp. The minimum charging amount of the sensor is bound to the power-on-demand timestamp and the location to form a unique identifier, which facilitates quick retrieval and management in the database. By implementing this invention, it is possible to obtain several locations with the timestamps of the power supply to be supplied and several minimum charging amounts. Combined with the scheduled tasks of the wireless power supply drones and the deployment locations of the energy station equipment, the scheduling scheme is randomly configured to generate several drone power supply scheduling schemes. The random configuration algorithm can generate dozens to hundreds of scheduling schemes, covering different combinations of drones, stations and paths, thereby improving the robustness of the system. By implementing this invention, path shortest constraint optimization can be achieved based on the aforementioned several UAV power supply scheduling schemes to obtain the target power supply scheduling scheme, and wireless power supply UAVs can be scheduled for wireless power acquisition and wireless power supply scheduling control, thereby reducing UAV energy consumption and mission time and improving efficiency.

[0013] In summary, by implementing this invention, the technical effects of reducing the energy supply cost of sensor networks and improving energy supply efficiency can be achieved. Attached Figure Description

[0014] Figure 1 The present invention provides a structural schematic of an energy replenishment system that combines an energy station device with a wirelessly powered drone. Figure 2 This invention provides a schematic diagram of the process for generating an initial environmental state matrix by performing multi-valued processing of environmental states in an energy replenishment system that combines an energy station device with a wireless power supply drone.

[0015] In the attached diagram, the components represented by each number are as follows: Sensor status monitoring module 11, sensor endurance prediction module 12, power demand sensor identification module 13, power supply scheme generation module 14, power supply scheme optimization module 15. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0018] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0019] Example 1, as Figure 1 As shown, this embodiment of the invention provides an energy replenishment system combining an energy station device and a wirelessly powered drone, comprising: Sensor status monitoring module 11 is used to receive residual power, location, environmental status and working status from distributed sensors; The sensor battery life prediction module 12 is used to retrieve the mode power loss rate of a sample set of sensors of the same model that meet the environmental conditions and the working conditions and have a service life of less than or equal to 0.2 years when the remaining power is greater than the battery life critical point, and combine the remaining power to make a battery life prediction to obtain a power supply timestamp when the remaining power is less than or equal to the battery life critical point. The power demand sensor identification module 13 is used to obtain the pre-stored minimum charging amount of the distributed sensor and identify and store the location in combination with the power demand timestamp. The power supply scheme generation module 14 is used to obtain several locations with the timestamp of the power supply to be supplied and several minimum charging amounts, and combine the scheduled tasks of the wireless power supply drones and the deployment locations of the energy station equipment to execute random configuration of the scheduling scheme and generate several drone power supply scheduling schemes. The power supply scheme optimization module 15 optimizes the path shortest constraint based on the aforementioned several UAV power supply scheduling schemes to obtain the target power supply scheduling scheme, and schedules the wireless power supply UAVs to perform wireless power acquisition and wireless power supply scheduling control.

[0020] In this embodiment, the sensor status monitoring module 11 receives information on remaining power, location, environmental status, and operational status from distributed sensors. These distributed sensors are various sensors installed in outdoor environments (such as stress sensors used for earthquake and landslide monitoring, deployed within mountains). Due to the complex environment or remote location, manual battery replacement is inconvenient. To recharge these sensors using the energy station equipment provided by this invention combined with a wirelessly powered drone, it is first necessary to collect basic information about these distributed sensors. This basic information includes remaining power (i.e., the remaining power of the sensor's built-in power supply), location (latitude and longitude coordinates, altitude, etc.), environmental status (such as temperature and humidity), and operational status (such as data acquisition frequency, transmission power, and computational load).

[0021] In this embodiment, the energy station equipment includes solar panels, a battery- or supercapacitor-based energy storage system, a medium-range 100-watt wireless power supply module, a control module, high-precision positioning, and long-range wireless communication components. The energy station collects solar energy through the solar panels, converts it into electrical energy, and stores it in the battery / supercapacitor. When needed, the medium-range wireless power supply module—a magnetic resonance coupling wireless charging device—wirelessly transmits the stored electrical energy to drones located on or within a certain distance above it, thus charging the drones. The energy station acts as a charging base for drones, enabling unattended continuous operation after deployment in the field, providing renewable energy replenishment for drones.

[0022] In this embodiment, the wirelessly powered drone is equipped with wireless energy receiving and supply capabilities. The drone carries a lightweight wireless energy receiving module to obtain power from the wireless power supply module of the energy station equipment and charge its own battery; it also carries a wireless power supply module to transmit its carried electrical energy to the sensor equipment in a contactless manner when near the sensor equipment. Furthermore, the drone is equipped with high-precision positioning components, including a GPS / Beidou module and a UWB (Ultra-Wideband) centimeter-level precision positioning module. The GPS module provides global navigation and positioning, while the UWB module provides precise relative positioning, enabling the drone to hover and dock accurately above the energy station or sensor equipment, ensuring efficient wireless energy transmission. The drone also has the necessary communication module for real-time communication with the cloud platform to obtain sensor status and upload its own status.

[0023] In this embodiment, the sensor device refers to various sensor nodes deployed on the ground, underground, or in enclosed spaces, such as environmental monitoring sensors buried in soil or stress sensors placed inside concrete structures. Each sensor device contains a wireless charging receiver (such as an induction coil and rectifier), a local energy storage element (battery / capacitor), and a long-range communication module (such as LoRa), and can optionally be equipped with a high-precision positioning module (such as a UWB beacon). The wireless charging receiver is used to receive power transmission from nearby drones, thereby charging the sensor's battery; the positioning module is used to align with the nearby drone when power is needed (if this module is not available, the sensor's position needs to be accurately determined during installation); the long-range communication module is used to periodically upload the sensor's own sensing data and current remaining power status data to a cloud platform. Through the above functional module design, the sensor device can still receive wireless power from drones in enclosed or inaccessible environments, without requiring on-site battery replacement.

[0024] The cloud platform described in this embodiment serves as the information scheduling and management center for the entire system. The cloud platform connects to the communication modules of all sensor devices, collecting sensing data, location information, remaining battery power, and operational status data from each sensor. When some sensors are low on power, the cloud platform generates power supply tasks based on geographical location and urgency, and plans the optimal power supply route and scheduling scheme for the drones. The cloud platform then issues task instructions to the wirelessly powered drones, such as instructing them to go to a designated energy station to charge or directly power specific sensors.

[0025] In this embodiment, the sensor battery life prediction module 12 is used to retrieve the mode power loss rate of a sample set of sensors of the same model that meet the environmental conditions and the working conditions and have a service life of less than or equal to 0.2 years when the remaining power is greater than the battery life critical point, and combine the remaining power to perform battery life prediction to obtain the time stamp of the battery life critical point to be powered.

[0026] The sensor battery life prediction module 12 is also used to set the current time as a power supply timestamp and identify and store the location in conjunction with the minimum charging amount when the remaining power is less than or equal to the battery life threshold.

[0027] The aforementioned endurance critical point is the minimum amount of power required to ensure the sensor maintains normal operation before the next recharging operation (e.g., 10 hours later), such as 20Wh. When the sensor's remaining power is greater than this critical point, it indicates that the sensor's recharging requirement is not urgent. Based on the mode power consumption rate of a sample set of sensors of the same model that meet the aforementioned environmental and operational conditions and have a service life of less than or equal to 0.2 years, the sensor's power consumption rate is estimated (e.g., 2Wh / h), and the sensor's current remaining power (e.g., 30Wh). The time required for the sensor's remaining power to reach the endurance critical point is estimated. In the example above, this time is (30Wh-20Wh) / 2Wh / h=5h. Therefore, adding the current time (e.g., 13:00 on May 28, 2025) to the time required for the sensor's remaining power to reach the endurance critical point (e.g., 5h) gives the timestamp for the remaining power being less than or equal to the endurance critical point (i.e., 18:00 on May 28, 2025).

[0028] When the remaining power is below the critical point, the sensor's remaining power is insufficient to sustain its operation until the next recharging operation. Therefore, recharging needs to be arranged as soon as possible. Thus, the current time needs to be set as the power supply timestamp (e.g., 13:00 on May 28, 2025). The minimum amount of power required to raise the sensor's remaining power above the endurance critical point (e.g., 10Wh), along with the aforementioned power supply timestamp and the sensor's location (latitude and longitude coordinates, altitude, etc.), should be identified and stored together. The identification information of the aforementioned sensor to be powered will serve as the basis for scheduling the drone to provide power.

[0029] In this embodiment, the sensor battery life prediction module 12's function of retrieving the sensor power mode loss rate of a sample set of sensors of the same model that meet the environmental state and the operating state includes: The environmental state is multi-valued to generate an initial environmental state matrix; The system receives a pre-configured weight distribution of environmental factors affecting power loss, and then weights the initial environmental state matrix to obtain a modified environmental state matrix. The working states are multi-valued to generate an initial working state matrix; The pre-configured working factor power loss influence weight distribution is received, and the initial working state matrix is ​​weighted to obtain the corrected working state matrix; Retrieve the sensor power mode loss rate of a sample set of sensors of the same model that satisfy the modified environmental state matrix and the modified operating state matrix.

[0030] Among them, Figure 2 As shown, the environmental state is multi-valued to generate an initial environmental state matrix, including: Step 1: Configure the first environmental factor fluctuation value through the user terminal; Step 2: Using the first environmental factor fluctuation value as a single variable, collect a set of power fluctuation values ​​for sensors of the same model; Step 3: Perform a centralized value evaluation on the set of power fluctuation values ​​of the same type of sensor to obtain the fitted value of the power fluctuation of the same type of sensor; Step 4: When the power fluctuation fitting value of the same model sensor is less than the power fluctuation consistency deviation predefined by the user terminal, increase the first environmental factor fluctuation value by 2 times and return to Step 2; Step 5: When the fitted value of the power fluctuation of the same model sensor is greater than the power fluctuation consistency deviation, increase the first environmental factor fluctuation value by 0.5 times and return to Step 2; Step 6: When the power fluctuation fitting value of the same model sensor is equal to the power fluctuation consistency deviation, set the first environmental factor fluctuation value as the first environmental factor fluctuation consistency deviation. Based on the consistent deviation of the first environmental factor fluctuation, the first environmental factor is partitioned from small to large according to the pre-configured first environmental factor rated range on the user end, and a unique mapping value is configured for each partition. The first environmental factor multi-value mapping function is constructed and added to the environmental factor multi-value mapping function library. Based on the environmental factor multi-valued mapping function library, the environmental state is multi-valued to generate the initial environmental state matrix.

[0031] Steps one through six above describe the specific process for multi-value processing of environmental factors (such as temperature and humidity). The purpose is to transform continuous environmental states (such as temperature values) into discrete numerical matrices, enabling quantitative analysis of the impact of environmental factors on sensor power consumption. The specific operation method is as follows: The purpose of configuring the first environmental factor fluctuation value in step one is to control the range of variable variation. In practice, a fluctuation value can be pre-selected based on the normal fluctuation range of the environmental factor in real-world conditions, serving as the range for testing the environmental factor's variation. For example, when the environmental factor under study is temperature, the fluctuation value can be set to 20℃±5℃ (i.e., the temperature fluctuates around 20℃ by 5℃).

[0032] In step two, the fluctuation value of the first environmental factor is used as a single variable to collect a set of power fluctuation values ​​for sensors of the same model. This is to control for variables, keeping other environmental factors constant while retaining only one variable (the fluctuation value of a certain environmental factor, such as temperature), and to study the impact of this variable on the power fluctuation of the sensors (recording the power consumption data of the same model sensors only when this environmental factor changes). For example, when the temperature fluctuates between 20±5℃, power consumption data of 100 sensors of the same model are collected (e.g., power consumption of 2Wh or 3Wh per hour, etc., to obtain the relationship between environmental factor fluctuation and power consumption).

[0033] Step three involves evaluating the central tendency of the power fluctuation values ​​from the same type of sensor to obtain a fitted value for the power fluctuation of the same type of sensor. This is achieved by statistically analyzing the collected power data and calculating its central tendency value, which represents the typical power loss rate under that fluctuation value. This fitted value can be calculated by taking the average or median of the power fluctuation values ​​from the same type of sensor. For example, if the temperature fluctuation range is 20℃±5℃, this value is the fitted value for the power fluctuation when the temperature fluctuates within 20℃±5℃.

[0034] In step four, if the power fluctuation fitting value of the same model sensor is less than the power fluctuation consistency deviation predefined by the user terminal, the first environmental factor fluctuation value is increased by 2 times and the process returns to step two.

[0035] The power fluctuation consistency deviation is the maximum deviation of the measured power loss data of the same model sensor from the fitted value of the power fluctuation. For example, assuming the fitted value of the power fluctuation is 2.6W, and the power loss data of the same model sensor is {2.4Wh / h, 2.6Wh / h, 2.7Wh / h}, the maximum deviation of the actual data is 0.2Wh / h (i.e., 2.6Wh / h - 2.4Wh / h). The power fluctuation consistency deviation predefined by the user terminal is the maximum deviation that can be tolerated without affecting the power loss prediction. Deviations within this range have negligible impact on the prediction of the sensor's remaining usage time. For example, the power fluctuation consistency deviation predefined by the user terminal can be defined as 0.3Wh / h.

[0036] If the maximum deviation between the fitted value (e.g., 2.0 Wh / h) and the actual data (e.g., 0.2 Wh / h) is less than the preset allowable error (0.3 Wh / h), it indicates that the current environmental factor fluctuations (e.g., ±5℃) do not have a significant impact on power loss. It is necessary to increase the range of environmental factor variations to enhance their impact on power loss until the user-set deviation standard is reached. Specifically, increase the fluctuation value by a factor of 2 (e.g., from ±5℃ to ±10℃), and return to step two to recollect data.

[0037] In step five, if the fitted value of the power fluctuation of the same model sensor is greater than the power fluctuation consistency deviation, the first environmental factor fluctuation value is increased by 0.5 times, and the process returns to step two. This means that when the maximum deviation between the fitted value (e.g., 2.0 Wh / h) and the actual data (e.g., 0.4 Wh / h) is less than the preset allowable error (0.3 Wh / h), it indicates that the current environmental factor fluctuation (e.g., ±5℃) has a significant impact on power loss. It is necessary to reduce the range of environmental factor variation to weaken its impact on power loss until the user-set deviation standard is met. Specifically, the fluctuation value is reduced to 0.5 times its original value (e.g., from ±10℃ to ±5℃), and the process returns to step two to re-collect data.

[0038] In step six, when the fitted value of the power fluctuation of the same type of sensor is equal to the power fluctuation consistency deviation, the first environmental factor fluctuation value is set as the first environmental factor fluctuation consistency deviation. This means that when the fitted value is exactly equal to the predefined deviation, it indicates that the current fluctuation value is a critical value, and at this time, the change in the environmental factor can just cause the expected power fluctuation.

[0039] Furthermore, the current fluctuation value is set as the first environmental factor fluctuation consistency deviation (i.e., the standard fluctuation range). Then, based on the user-configured environmental factor rated range (i.e., the normal operating temperature range of the sensor, such as -20℃ to 50℃), the range is divided from small to large fluctuation consistency deviation (e.g., 0℃±15℃, 15℃±15℃, etc.). A multi-valued mapping function (e.g., mapping power consumption values ​​through temperature) is constructed and stored in the environmental factor multi-valued mapping function library. This process is similar to "digitizing" continuous environmental data, enabling the system to quantify environmental factors through mathematical models, thereby more accurately predicting the sensor's power consumption rate.

[0040] Using the methods described above, we can further establish the mapping relationship between other environmental factors (such as humidity) and sensor power consumption values, which will not be elaborated here.

[0041] For example, temperature can be classified according to thresholds as high temperature (>30℃, mapped to 3), medium temperature (20℃~30℃, mapped to 2), and low temperature (<20℃, mapped to 1). Humidity can be classified according to percentage as high humidity (>70%, mapped to 3), medium humidity (40%~70%, mapped to 2), and low humidity (<40%, mapped to 1).

[0042] Next, based on the environmental factor multi-valued mapping function library, the environmental state needs to be multi-valued to generate the initial environmental state matrix. The initial environmental state matrix is ​​formed by mapping real-time collected environmental state parameters (such as temperature 25℃ and humidity 60%) into discrete numerical vectors through the function library, creating a one-dimensional or multi-dimensional matrix. For example, if the environmental state is "temperature 25℃, humidity 60%", after mapping through the function library, it becomes a numerical vector [2,2], which is a component of the initial environmental state matrix.

[0043] The pre-configured environmental factor power loss influence weight distribution includes: Based on the set of environmental factors, collect the first environmental factor fluctuation value dataset corresponding to the same type of sensor up to the Nth environmental factor fluctuation value dataset, as well as the power fluctuation value dataset. The first environmental factor fluctuation value dataset is traversed up to the Nth environmental factor fluctuation value dataset and then subjected to dimensionless processing to obtain the first environmental factor fluctuation value sequence up to the Nth environmental factor fluctuation value sequence. The power fluctuation value dataset is subjected to dimensionless processing to obtain a power fluctuation value sequence; Using the power fluctuation value sequence as the baseline sequence and the first environmental factor fluctuation value sequence up to the Nth environmental factor fluctuation value sequence as the comparison sequence, grey relational analysis is performed to obtain the correlation degree of the first environmental factor up to the Nth environmental factor. The correlation degree is then compared with the sum of the correlation degrees to generate the weight distribution of the impact of power loss on the environmental factors.

[0044] In this embodiment, the aforementioned environmental factors refer to various environmental parameters that may affect power loss during production, including but not limited to temperature and humidity, where N is the number of environmental factors. For each environmental factor, fluctuation data of the same type of sensor under different operating conditions are collected to form N datasets. For example, the temperature factor fluctuation value dataset may contain [20℃, 25℃, 30℃,...], and the humidity factor fluctuation value dataset may contain [40%, 60%, 80%,...].

[0045] Next, for each environmental factor fluctuation, the sensor's power fluctuation value is recorded. For example, when the temperature is 20℃, the power fluctuation value is 0.3Wh / h; when the temperature is 25℃, the power fluctuation value is 0.2Wh / h. Since different environmental factors have different physical dimensions (e.g., temperature is measured in °C, humidity in %), direct comparison will produce deviations, so dimensionless conversion is necessary. Specifically, the maximum and minimum values ​​in each environmental factor fluctuation value dataset can be calculated first, and then each data point can be converted into a relative value within the range of the maximum and minimum values. For example, the calculation method can be relative value = (original value - minimum value) / (original value - minimum value), where the original value is the baseline value used to calculate the environmental factor fluctuation value, and the converted relative value ranges from 0 to 1. This eliminates the influence of dimensions, thus obtaining the i-th environmental factor fluctuation value sequence, where i is a positive integer less than or equal to N. Performing the same relative value conversion on the power fluctuation value dataset yields the power fluctuation value sequence.

[0046] Further, grey relational analysis is required to calculate the degree of correlation between each environmental factor and power consumption. In this embodiment, the baseline sequence for grey relational analysis is the power consumption fluctuation value sequence (i.e., the power consumption data processed in step two above), and the comparison sequence is the fluctuation value sequence of each environmental factor (a total of N sequences). Specifically, for each environmental factor sequence, the absolute difference between it and the baseline sequence at each data point is calculated. For example, the difference between the k-th data point of the temperature sequence and the k-th data point of the power consumption sequence is |xtemperature(k)−xpower(k)|. Next, the maximum and minimum differences need to be calculated. The minimum difference is the minimum value among all absolute differences, and the maximum difference is the maximum value among all absolute differences.

[0047] Next, the correlation coefficient needs to be calculated. For each data point in each environmental factor sequence, the correlation coefficient is calculated as follows: Correlation coefficient = (Minimum difference + ρ × Maximum difference) / (|x0(k)−xi(k)| + ρ × Maximum difference). Here, ρ is the resolution coefficient (usually taken as 0.5), used to adjust the sensitivity of the correlation coefficient.

[0048] Finally, the correlation coefficients of each environmental factor sequence are averaged to obtain the correlation degree r of that factor. The higher the correlation degree, the greater the impact of the environmental factor on power loss.

[0049] In this embodiment, the correlation degree of each environmental factor also needs to be converted into weights. First, the total correlation degree needs to be calculated: Total correlation degree = r1 + r2 + ... + r N For each environmental factor, its weight is w. i =r i / Sum of correlations. It is not difficult to see that the sum of all weights is 1, and thus the weight distribution of the environmental factor power loss can be calculated.

[0050] Based on the weighted distribution of the environmental factors affecting power loss, the initial environmental state matrix can be weighted to obtain a corrected environmental state matrix. For example, if the temperature weight is 0.6 and the humidity weight is 0.4, the initial environmental state matrix is ​​[2,2] (corresponding to temperature and humidity respectively), and the corrected environmental state matrix is ​​[1.2,0.8]. In this embodiment, a multi-valued working state processing is also required. First, the working states (such as data acquisition frequency, transmission power, computational load, etc.) need to be converted into quantifiable parameters, such as acquisition frequency (times / second) and transmission power (W). Then, using the same processing method as for environmental factors, a mapping relationship between the working state parameters and discrete values ​​is established. For example, for data acquisition frequency, ≤10 times / second corresponds to matrix parameter 1; 10~50 times / second corresponds to matrix parameter 2; 10~50 times / second corresponds to matrix parameter 3. Similarly, for transmission power, ≤0.5W corresponds to matrix parameter 1; 0.5~1W corresponds to matrix parameter 2; >1W corresponds to matrix parameter 3, and so on. Finally, when the data acquisition frequency = 20 times / second and the transmission power = 0.8W, the initial working state matrix can be [2,2]. If its weight vector is [0.7,0.3], then the corrected working state matrix is ​​[2×0.7,2×0.3] = [1.4,0.6]. Through the above method, corrected working state matrices for multiple working states can be obtained.

[0051] In this embodiment of the application, sensor sample retrieval and mode loss rate calculation are also required.

[0052] First, the corrected environment and working state matrices need to be used as search criteria. For example, the corrected environment state matrix is ​​[1.8, 0.6], and the corrected working state matrix is ​​[1.4, 0.6].

[0053] Next, sensor samples that meet the following conditions need to be selected from the historical database (i.e., the same model sensor sample set): same sensor model, service life ≤ 0.2 years, matching environmental conditions, and matching operating conditions. The matching criteria for environmental and operating conditions can be that the Euclidean distance between the corrected environmental / operating matrix and the sample matrix is ​​≤ a threshold (e.g., 0.5, 0.3, etc., set according to actual needs). The historical database contains pre-collected actual power consumption data of the same model sensors under different environmental and operating conditions. For example, one set of data could be: temperature 25℃, humidity 60%, data acquisition frequency 20 times / second, transmission power 0.8W, and real-time power consumption rate 2.2Wh / h. Collecting data from multiple such same model sensors, covering the possible ranges of various parameters, constitutes the aforementioned historical database, i.e., the same model sensor sample set.

[0054] Finally, for the selected samples that meet the above conditions, their power consumption rate (Wh / h) is statistically analyzed, and then the mode of the consumption rate (i.e. the value that appears most frequently, such as 2.3Wh / h) is calculated.

[0055] Further, retrieving the sensor power mode loss rate of a sample set of sensors of the same model that satisfy the environmental state and the operating state includes: Based on a preset model sensor, the power consumption rate of a baseline service life sensor and a second service life sensor under the same environment and working conditions are collected, where the baseline service life is less than or equal to 0.2 years. Calculate the first ratio of the power consumption rate of the second service life sensor to the power consumption rate of the reference service life sensor, and calculate the second ratio of the second service life to the reference service life. Using the second ratio data as input and the first ratio data as regression prediction supervision, train the power loss ratio prediction model; The service life of the distributed sensor is compared with the baseline service life. The power loss ratio prediction model is input to obtain the predicted power loss ratio value. The initial sensor power loss rate fitted to the sample set of the same type of sensor is corrected to obtain the sensor power loss rate.

[0056] In this embodiment, to achieve the above steps, data acquisition for a specific sensor model is first required. Ensuring the collected sensor data is collected under the same environmental conditions (e.g., temperature, humidity) and operating states (e.g., data frequency, transmission power), a sensor with a service life ≤ 0.2 years is selected as a benchmark (new sensors are less affected by battery aging). Its power consumption rate V0 is recorded. Then, sensors with different service lives (e.g., 0.5 years, 1 year, 2 years, etc., i.e., the second service life) are selected, and their power consumption rate is recorded as V0.i Next, the first ratio is calculated, which is the ratio of the wear rate of each sensor at different service years to the wear rate of the reference sensor. This reflects the multiplier effect of battery aging on the wear rate. Specifically, the first ratio = V i / V0. Then calculate the second ratio data, which reflects the relative service time. The calculation method is: second ratio = T i / T0. Where T i T0 represents the service life of the i-th sensor, and T0 represents the baseline service life (e.g., 0.2 years).

[0057] Next, a power loss ratio prediction model needs to be trained. This model takes the second ratio data (relative service time) as input and the first ratio data (power loss ratio) as output. It uses machine learning algorithms to learn the relationship between service time and the rate of power loss, thus achieving the effect of predicting the output rate of power loss based on service time.

[0058] Specifically, since the power consumption of the same type of sensor increases with the increase in service life, a linear regression model can be used for this prediction model. The input feature of this model is the second ratio data (i.e., relative service time), and the output label is the first ratio data (i.e., power loss ratio). The training sample data consists of multiple sets of relative service time and corresponding power loss ratio data of the same type of sensor under the same environment and operating conditions, for example (relative service time 1.5, power loss ratio 1.2). At least 1000 sets of the aforementioned data are collected as sample data.

[0059] In the selection of parameters for the power loss ratio prediction model, the learning rate α is set to 0.01; the number of iterations is 1000; the parameter update method is batch gradient descent (BGD); the mean squared error (MSE) is used as the loss function; the gradient is calculated and the parameters are updated through backpropagation; the convergence criterion can be the absolute loss threshold, and the model is considered to have converged when the MSE of the training set is less than 0.0001.

[0060] The service life of the distributed sensor is compared with the baseline service life (e.g., 1.5, 2, etc.), and then input into the power loss ratio prediction model to obtain the predicted power loss ratio value (e.g., 1.2, 1.3, etc.).

[0061] Finally, the initial sensor power mode loss rate fitted to the sample set of the same model of sensor needs to be corrected to obtain the sensor power mode loss rate. For example, if the initial sensor power mode loss rate (calculated based on the baseline service life) is 2.0 Wh / h and the predicted power loss ratio is 1.8, then the corrected loss rate corresponding to the sensor under the actual service life is 2.0 × 1.8 = 3.6 Wh / h.

[0062] In this embodiment of the application, the power demand sensor identification module 13 obtains the pre-stored minimum charging amount of the distributed sensor, and identifies and stores the location in combination with the power demand timestamp; Since the predicted corrected wear rate of the sensor under its actual service life has been obtained through the above steps, it is only necessary to substitute the corrected wear rate of the sensor under its actual service life into the sensor power mode wear rate estimated in the sensor endurance prediction module 12 to calculate the power consumption rate of the sensor. This will yield the corresponding sensor's power-requiring timestamp. Combined with the sensor's location information, the sensor identifier can be obtained and stored. Further details are omitted here.

[0063] In this embodiment, the power supply scheme generation module 14 is used to obtain several locations with the timestamps of the power supply to be supplied and several minimum charging amounts, and, in conjunction with the scheduled tasks of the wireless power supply drones and the deployment locations of the energy station equipment, execute random configuration of the scheduling scheme to generate several drone power supply scheduling schemes, including: Based on the aforementioned locations and the aforementioned minimum charging amounts, construct the desired wireless power supply constraints; Based on the deployment locations of the energy station equipment, construct location constraints for drone recharging; Based on the scheduled tasks of wirelessly powered drones, obtain the list of idle drone locations and the list of idle drones with remaining power at the timestamps of the timestamps to be powered, wherein the remaining power of the idle drones is equal to the actual remaining power minus the rated remaining power. Based on the list of idle drone locations and the list of idle drones with remaining power, several drone power supply scheduling schemes are randomly configured to satisfy the wireless power supply expectation constraints and the drone recharging location constraints.

[0064] The expected constraint for wireless power supply is that the drone must carry enough power to reach the location and complete the charging process, i.e.: the drone's remaining power capacity ≥ the power capacity of the round trip route + the minimum charging amount.

[0065] The drone recharging location constraint requires that the drone must obtain energy from the energy station before performing the power supply task. That is, the drone path must include the energy station node, and the drone path in the scheduling plan must be current location → energy station → power supply location to ensure that the drone has power to recharge.

[0066] After establishing constraints, it is necessary to screen the idle drones corresponding to the power-requiring timestamps of the power-requiring sensors determined through the aforementioned steps. The constraint is that the drone is idle at the power-requiring timestamp and still has available power after deducting the rated remaining power (i.e., the round-trip power + minimum charging amount) to ensure that the drone can return home. A set of drones that meet the conditions is selected as usable drones.

[0067] Based on the above constraints, several UAV power supply scheduling schemes that satisfy the wireless power supply expectation constraint and the UAV recharging location constraint are randomly configured. For example, Scheme 1 is that UAV U001 departs from energy station S1, charges sensor L1, and then returns to S1; Scheme 2 is that UAV U002 departs from energy station S2, charges sensor L2, and then returns to S2, and so on, generating multiple similar schemes.

[0068] In this embodiment, the power supply scheme optimization module 15 is used to perform path shortest constraint optimization based on the aforementioned several UAV power supply scheduling schemes, obtain the target power supply scheduling scheme, and schedule wireless power supply UAVs for wireless power acquisition and wireless power supply scheduling control.

[0069] Among them, the shortest path constraint optimization is performed based on the aforementioned UAV power supply scheduling schemes to obtain the target power supply scheduling scheme. Specifically, a greedy algorithm can be used to implement this, selecting the current optimal solution (shortest path) each time to gradually build a complete scheme.

[0070] For example, in Scheme 1, the total path length for U001 is 10km, and the total path length for U002 is 12km, for a total path length of 22km. In Scheme 2, the total path length for U001 is 8km, and the total path length for U002 is 15km, for a total path length of 23km. Scheme 1 is ultimately chosen, and so on, iteratively calculating to obtain the optimal path. The method of calculating the optimal path using algorithms is existing technology, easy to obtain and implement, and will not be elaborated here.

[0071] This includes the scheduling and control of wirelessly powered drones for wireless power acquisition and supply, including: When the wireless power supply drone enters the first preset range centered on the deployment location of the energy station equipment, the medium-range wireless power transmission module of the energy station equipment is activated, and the first wireless energy receiving unit of the wireless power supply drone is activated to charge the wireless power supply drone. When the wirelessly powered drone enters the second preset range centered on the location of the distributed sensor, the short-range wireless power supply module of the wirelessly powered drone is activated, and the second wireless energy receiving unit of the distributed sensor is activated to draw power from the wirelessly powered drone.

[0072] In this embodiment, the energy station equipment is typically installed in an open outdoor location to maximize solar energy absorption. A wireless power supply module is also included. Optionally, when the wirelessly powered drone flies to a circular area centered on the energy station (e.g., with a radius of 500 meters), the energy station activates its medium-range wireless power transmission module (e.g., a magnetic resonance coupling wireless charging device, with a transmission distance within 500 meters), and the wirelessly powered drone activates its first wireless energy receiving unit (e.g., a receiving coil) to begin charging.

[0073] Among them, magnetic resonance coupling wireless power transfer technology can generate an alternating magnetic field within a certain range. Its transmitting antenna can be designed to face directly upwards to form an energy coverage area. Wireless power supply is activated and deactivated according to the drone's request, and parameters such as transmission power and frequency are adjusted to adapt to the drone's reception. When the wirelessly powered drone needs charging, the wireless power supply module is activated to transmit power. The energy station can also be equipped with environmental sensing sensors and an emergency stop switch. For example, if a foreign object is detected entering the power supply area or the wirelessly powered drone deviates from its position, the power supply will be cut off in time to prevent energy leakage or safety accidents.

[0074] The wirelessly powered drone is preferably a multi-rotor aircraft with vertical takeoff and landing (VTOL) and hovering capabilities. It is equipped with a dedicated wireless power receiving unit to harvest energy from the electromagnetic field emitted by the power station equipment. The drone also carries a short-range wireless power supply module for charging sensor devices. This module can employ an electromagnetic induction coil or a magnetic resonance transmitting coil, mounted beneath the drone's fuselage. It generates a strong local magnetic field when the drone approaches the sensor device, coupling energy to the sensor's receiving unit (the second wireless power receiving unit).

[0075] For example, when a wirelessly powered drone flies to a circular area (e.g., within a radius of 100 meters) centered on the sensor, the drone activates its short-range wireless power supply module (which can use electromagnetic induction technology with a transmission distance of up to 10 meters). The sensor then activates its second wireless energy receiving unit to begin drawing power until it has finished charging the sensor and returns to the energy station to recharge.

[0076] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0077] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0078] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0079] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0080] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0081] Although preferred embodiments of the invention have been described, those skilled in the art, once they have learned the basic inventive concept, can make other changes and modifications to these embodiments.

[0082] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. An energy replenishment system combining an energy station device with a wirelessly powered drone, characterized in that, Applied to a cloud platform, the cloud platform is communicatively connected to distributed sensors, energy station equipment, and wirelessly powered drones, including: The sensor status monitoring module is used to receive information on remaining power, location, environmental status, and operating status from distributed sensors. The sensor battery life prediction module is used to retrieve the mode power loss rate of a sample set of sensors of the same model that meet the environmental conditions and the working conditions and have a service life of less than or equal to 0.2 years when the remaining power is greater than the battery life critical point, and combine it with the remaining power to predict the battery life and obtain the power-on-demand timestamp when the remaining power is less than or equal to the battery life critical point. A power demand sensor identification module is used to obtain the pre-stored minimum charging amount of the distributed sensor and identify and store the location in combination with the power demand timestamp; The power supply scheme generation module is used to obtain several locations with the timestamp of the power supply to be supplied and several minimum charging amounts, and combine the scheduled tasks of the wireless power supply drones and the deployment locations of the energy station equipment to execute random configuration of the scheduling scheme and generate several drone power supply scheduling schemes. The power supply scheme optimization module optimizes the path shortest constraint based on the aforementioned UAV power supply scheduling schemes to obtain the target power supply scheduling scheme, and schedules the wireless power supply UAVs for wireless power acquisition and wireless power supply scheduling control.

2. The system as described in claim 1, characterized in that, Also includes: When the remaining battery power is less than or equal to the endurance threshold, the current time is set as the power supply timestamp, and the location is identified and stored in conjunction with the minimum charging amount.

3. The system as described in claim 1, characterized in that, Retrieving the sensor power mode loss rate from a sample set of sensors of the same model that satisfy the stated environmental and operating conditions includes: The environmental state is multi-valued to generate an initial environmental state matrix; The system receives a pre-configured weight distribution of environmental factors affecting power loss, and then weights the initial environmental state matrix to obtain a modified environmental state matrix. The working states are multi-valued to generate an initial working state matrix; The pre-configured working factor power loss influence weight distribution is received, and the initial working state matrix is ​​weighted to obtain the corrected working state matrix; Retrieve the sensor power mode loss rate of a sample set of sensors of the same model that satisfy the modified environmental state matrix and the modified operating state matrix.

4. The system as described in claim 3, characterized in that, The environmental state is multi-valued to generate an initial environmental state matrix, including: Step 1: Configure the first environmental factor fluctuation value through the user terminal; Step 2: Using the first environmental factor fluctuation value as a single variable, collect a set of power fluctuation values ​​for sensors of the same model; Step 3: Perform a centralized value evaluation on the set of power fluctuation values ​​of the same type of sensor to obtain the fitted value of the power fluctuation of the same type of sensor; Step 4: When the power fluctuation fitting value of the same model sensor is less than the power fluctuation consistency deviation predefined by the user terminal, increase the first environmental factor fluctuation value by 2 times and return to Step 2; Step 5: When the fitted value of the power fluctuation of the same model sensor is greater than the power fluctuation consistency deviation, increase the first environmental factor fluctuation value by 0.5 times and return to Step 2; Step 6: When the power fluctuation fitting value of the same model sensor is equal to the power fluctuation consistency deviation, set the first environmental factor fluctuation value as the first environmental factor fluctuation consistency deviation. Based on the consistent deviation of the first environmental factor fluctuation, the first environmental factor is partitioned from small to large according to the pre-configured first environmental factor rated range on the user end, and a unique mapping value is configured for each partition. The first environmental factor multi-value mapping function is constructed and added to the environmental factor multi-value mapping function library. Based on the environmental factor multi-valued mapping function library, the environmental state is multi-valued to generate the initial environmental state matrix.

5. The system as described in claim 4, characterized in that, Receive the pre-configured weight distribution of environmental factors affecting power loss, including: Based on the set of environmental factors, collect the first environmental factor fluctuation value dataset corresponding to the same type of sensor up to the Nth environmental factor fluctuation value dataset, as well as the power fluctuation value dataset. The first environmental factor fluctuation value dataset is traversed up to the Nth environmental factor fluctuation value dataset and then subjected to dimensionless processing to obtain the first environmental factor fluctuation value sequence up to the Nth environmental factor fluctuation value sequence. The power fluctuation value dataset is subjected to dimensionless processing to obtain a power fluctuation value sequence; Using the power fluctuation value sequence as the baseline sequence and the first environmental factor fluctuation value sequence up to the Nth environmental factor fluctuation value sequence as the comparison sequence, grey relational analysis is performed to obtain the correlation degree of the first environmental factor up to the Nth environmental factor. The correlation degree is then compared with the sum of the correlation degrees to generate the weight distribution of the impact of power loss on the environmental factors.

6. The system as described in claim 1, characterized in that, Retrieving the sensor power mode loss rate from a sample set of sensors of the same model that satisfy the stated environmental and operating conditions includes: Based on a preset model sensor, the power consumption rate of a baseline service life sensor and a second service life sensor under the same environment and working conditions are collected, where the baseline service life is less than or equal to 0.2 years. Calculate the first ratio of the power consumption rate of the second service life sensor to the power consumption rate of the reference service life sensor, and calculate the second ratio of the second service life to the reference service life. Using the second ratio data as input and the first ratio data as regression prediction supervision, train the power loss ratio prediction model; The service life of the distributed sensor is compared with the baseline service life. The power loss ratio prediction model is input to obtain the predicted power loss ratio value. The initial sensor power loss rate fitted to the sample set of the same type of sensor is corrected to obtain the sensor power loss rate.

7. The system as described in claim 1, characterized in that, Obtain several locations with the timestamps of the pending power supply and several minimum charging amounts. Combined with the scheduled tasks of the wireless power supply drones and the deployment locations of the energy station equipment, execute random configuration of the scheduling scheme to generate several drone power supply scheduling schemes, including: Based on the aforementioned locations and the aforementioned minimum charging amounts, construct the desired wireless power supply constraints; Based on the deployment locations of the energy station equipment, construct location constraints for drone recharging; Based on the scheduled tasks of wirelessly powered drones, obtain the list of idle drone locations and the list of idle drones with remaining power at the timestamps of the timestamps to be powered, wherein the remaining power of the idle drones is equal to the actual remaining power minus the rated remaining power. Based on the list of idle drone locations and the list of idle drones with remaining power, several drone power supply scheduling schemes are randomly configured to satisfy the wireless power supply expectation constraints and the drone recharging location constraints.

8. The system as described in claim 1, characterized in that, Dispatching wirelessly powered drones for wireless power acquisition and supply control includes: When the wireless power supply drone enters the first preset range centered on the deployment location of the energy station equipment, the medium-range wireless power transmission module of the energy station equipment is activated, and the first wireless energy receiving unit of the wireless power supply drone is activated to charge the wireless power supply drone. When the wirelessly powered drone enters the second preset range centered on the location of the distributed sensor, the short-range wireless power supply module of the wirelessly powered drone is activated, and the second wireless energy receiving unit of the distributed sensor is activated to draw power from the wirelessly powered drone.

Images