Multi-stage hibernation energy management method for intelligent parking space management device

Abstract

The application discloses a multi-level hibernation energy management method of an intelligent parking space management device, relates to the field of multi-level hibernation energy management of a parking space management device, and comprises the following steps: the application predicts an energy supply expected value based on a solar panel photoelectric conversion parameter and a historical light change trend, and dynamically divides a hibernation state level in combination with an energy storage battery state of charge and a vehicle access frequency; the duration of a parking space without a vehicle is monitored in real time, and a time-sensitive threshold is calculated in combination with the energy storage battery state of charge and the vehicle access frequency, so that the hibernation state level is dynamically adjusted; when the energy supply expected value is in a low supply state and the energy storage battery is in a low power state, a state reduction priority strategy is started, and subsequent hibernation state switching is predicted based on historical records; the energy utilization efficiency of the intelligent parking space management device in a complex environment such as weak light or low temperature is realized, and the operation demand in a long-term unattended scene is met.

Description

Technical Field

[0001] This invention belongs to the field of multi-level dormant energy management technology for parking space management devices, and relates to a multi-level dormant energy management method for intelligent parking space management devices. Background Technology

[0002] With the rapid development of smart cities and intelligent parking systems, solar-powered intelligent parking management devices have been widely used in open-air parking lots due to their advantages such as no wiring required and energy saving.

[0003] However, the current multi-level sleep energy management scheme for intelligent parking space management devices has the following shortcomings in practical applications: In low light or low temperature environments, the efficiency of solar energy collection and battery energy storage performance drop significantly, easily leading to energy supply interruptions, making it difficult for the device to work stably and causing frequent power outages; secondly, vehicle entry and exit from parking spaces are highly random, while solar energy replenishment is greatly affected by weather and time of day, and the existing fixed sleep strategy is prone to power waste or detection failure; at the same time, there is a lack of emergency degradation mechanism in low power and low replenishment scenarios. If graded sleep regulation is not implemented for prediction in the future, it may lead to low energy utilization and fail to meet the long-term unattended operation requirements. Summary of the Invention

[0004] In view of this, in order to solve the problems mentioned in the background art, the present invention provides a multi-level dormant energy management method for intelligent parking space management devices.

[0005] The objective of this invention can be achieved through the following technical solution: a multi-level dormant energy management method for intelligent parking space management devices, comprising: S100, predicting the expected energy replenishment value in the future time period based on the photoelectric conversion parameters of solar panels and historical light change trends.

[0006] S200. Based on the current state of charge of the energy storage battery and the expected energy replenishment value, the intelligent parking space management device is divided into hibernation state levels, including zero hibernation state, shallow hibernation state and deep hibernation state.

[0007] The S300 monitors the duration of time a parking space remains empty in real time, and calculates time-sensitive thresholds based on the current state of charge of the energy storage battery and the frequency of vehicle entry and exit, dynamically adjusting the device's sleep state level switching.

[0008] S400 If the expected energy supply value is detected to be in a low supply state and the state of charge of the energy storage battery is in a low charge state, then the state priority reduction strategy is activated.

[0009] S500 records the state of charge of the energy storage battery, the expected energy replenishment value, and the actual dormancy state after each state adjustment, and predicts the subsequent dynamic adjustment device dormancy state switching based on these records.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention predicts the expected value of energy replenishment based on the photoelectric conversion parameters of the solar panel and the historical light change trend, and dynamically divides the hibernation state level by combining the state of charge of the energy storage battery and the frequency of vehicle entry and exit. This overcomes the problem of energy supply interruption caused by the decline in solar energy collection efficiency and battery energy storage performance under weak light or low temperature conditions, avoids frequent power outages of the device, realizes the active prediction of energy supply and matching of hibernation level under harsh environmental conditions, and improves the stability of system operation.

[0011] (2) This invention monitors the duration of parking space vacancy in real time and calculates the time-sensitive threshold by combining the state of charge of the energy storage battery and the frequency of vehicle entry and exit. It dynamically adjusts the hibernation state level switching, overcomes the defects of existing fixed hibernation strategies that are difficult to adapt to the randomness of vehicle entry and exit and the fluctuation of energy supply, avoids power consumption waste or detection failure, and realizes dynamic adaptation of hibernation strategy to real-time scenario requirements.

[0012] (3) This invention overcomes the shortcomings of existing technologies that cannot predict and regulate future hibernation by starting a downgrade priority strategy when the expected energy replenishment value is in a low replenishment state and the energy storage battery is in a low power state, and predicts the subsequent hibernation state switching based on historical records, thereby avoiding the problem of low energy utilization, realizing hibernation and predictive regulation in low power and low replenishment scenarios, and meeting the operation requirements in long-term unattended scenarios. Attached Figure Description

[0013] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a diagram illustrating the implementation steps of the method of the present invention; Figure 2 This is a flowchart of the energy replenishment expectation prediction process of the present invention; Figure 3 This is a flowchart illustrating the hierarchical division of the dormant state in this invention. Detailed Implementation

[0015] 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.

[0016] Please see Figure 1 As shown, the present invention provides a multi-level dormant energy management method for intelligent parking space management devices, including: S100, predicting the expected energy replenishment value in the future time period based on the photoelectric conversion parameters of the solar panel and the historical light change trend.

[0017] Since solar power generation depends on sunlight intensity, it can be supplemented with energy during the day or when there is sunlight, but at night or when there is no sunlight, it relies entirely on the stored energy in energy storage batteries. By predicting the expected future energy supply, it is possible to identify risk scenarios where current power is sufficient but future supply will be insufficient before energy shortages occur.

[0018] refer to Figure 2 As shown, the expected energy supply value for the future time period includes: S100.1, obtaining the geographical location of the intelligent parking space management device and the timestamp of the future time period, and calculating the theoretical maximum light reference value for each moment in the future time period through solar radiation.

[0019] S100.2 Calculate the ratio of the actual power of the same period in history to the theoretical maximum illumination reference value at the corresponding time to obtain the historical attenuation factor of the degree of meteorological interference.

[0020] S100.3 Calculate the ratio of the actual power at the current moment to the theoretical maximum illumination reference value at the current moment in real time, and perform a moving average filter on the comparison value to obtain the real-time trend correction coefficient; the moving average filter process calculates the moving average value by setting the moving window length N, such as N=5, that is, taking the instantaneous ratio of the current moment and the past N-1 sampling periods.

[0021] S100.4 Multiply the theoretical maximum illumination reference value, the historical attenuation factor at the corresponding time, and the current real-time trend correction coefficient to obtain the instantaneous power prediction value at each time.

[0022] S100.5 Integrate the instantaneous power prediction value within the future time period to obtain the energy replenishment expectation value; the energy replenishment expectation value refers to the total predicted amount of electrical energy that the solar panels can replenish to the energy storage battery within the predicted time period; the future time period is a complete natural cycle as the prediction time period, such as predicting to the start of the next solar cycle.

[0023] The geographical location of the intelligent parking space management device in S100.1 is calculated in real time by the satellite positioning module upon initial power-on, determining the relative angle of the sun. The future time period timestamp is obtained by the real-time clock module integrated within the device, which acquires the current date and time and, based on the current moment until the next predicted vehicle arrival time, obtains a sequence of discrete time points within the future time period, determining the solar declination angle and hour angle.

[0024] The theoretical maximum irradiance reference value refers to the theoretical ground irradiance under conditions of no atmospheric interference, representing the maximum light intensity that the device can obtain under ideal clear sky conditions. The higher the solar altitude angle, the greater the theoretical irradiance; the higher the altitude, the greater the theoretical irradiance.

[0025] The ratio in S100.2 above is not a single calculated value, but the result of statistical analysis based on the ratios of multiple historical time periods. Generally, the average or median value is taken. By statistically analyzing multiple historical data from the same period, the influence of single outliers can be filtered out.

[0026] The historical period refers to a certain number of days before and after the same date within at least one calendar year in the past, such as historical data within 7 days. During the statistical analysis, outliers caused by equipment failure or extreme weather are first removed, and then the average of the remaining ratios is calculated as the historical attenuation factor.

[0027] The historical attenuation factor of the meteorological interference level is the percentage of the actual light intensity to the theoretical maximum light intensity at the corresponding historical time. When the value is close to 1, it indicates that the meteorological conditions at that time are close to ideal clear skies and the atmospheric interference is minimal. When it is close to 0, it indicates that the meteorological conditions at that time are extremely poor and the light intensity is severely attenuated. When it is in the middle value, it indicates different degrees of meteorological interference, such as thin clouds and light fog.

[0028] The multiplication operation in S100.4 above is essentially a step-by-step attenuation correction, which means that solar radiation first passes through the atmosphere, then through the current cloud layer, and finally reaches the surface of the solar panel.

[0029] S200. Based on the current state of charge of the energy storage battery and the expected energy replenishment value, the intelligent parking space management device is divided into hibernation state levels, including zero hibernation state, shallow hibernation state and deep hibernation state.

[0030] Given that the energy input of solar power devices is intermittent, relying solely on the current state of charge to make dormancy decisions is insufficient to cope with risky scenarios. Furthermore, simply identifying risks is not enough to guide the device's behavioral decisions. Once the risk of energy shortage is confirmed, it is necessary to determine the level of functionality at which the device should operate to adapt to the current energy situation.

[0031] refer to Figure 3 As shown, the process of dividing the intelligent parking space management device into sleep state levels includes: S200.1, obtaining the current state of charge value of the energy storage battery and comparing it with the high charge baseline that is dynamically updated based on the battery health status and historical charge and discharge data.

[0032] S200.2 Obtain the predicted expected energy supply value and compare it with the low supply baseline generated based on the statistical patterns of energy supply in the same period of history.

[0033] S200.3 If the current state of charge is greater than or equal to the high charge baseline, then the partitioning device is in a zero-sleep state regardless of the expected energy replenishment value; the zero-sleep state refers to the working mode in which all functional modules of the intelligent parking space management device are in normal working condition.

[0034] The high-capacity baseline is the critical point between abundant and tight energy. When the current state of charge of the energy storage battery is greater than or equal to the high-capacity baseline, it means that the device is in an energy redundancy state, that is, the existing electrical energy is sufficient to support full-function operation for a considerable period of time, and there is no need to consider the amount of future energy replenishment. It can be directly classified into a zero-dormant state.

[0035] S200.4 If the current state of charge is less than the high charge baseline but the expected energy replenishment value is greater than or equal to the low replenishment baseline, the partitioning device is in a shallow sleep state. The shallow sleep state means that the intelligent parking space management device shuts down some additional functional units configured to improve user experience, such as the wireless communication module and the status indication module, and only retains the most basic functions of sensing vehicles entering or leaving the parking space, such as the working mode of the sensor module and its signal acquisition and processing circuit.

[0036] When the current state of charge of the energy storage battery is less than the high-capacity baseline, the device is out of the energy redundancy state, and the existing power is insufficient, so energy saving needs to be considered; at the same time, the expected energy replenishment value is greater than or equal to the low-replenishment baseline, and effective energy replenishment can be obtained in the future, so the risk of energy shortage is controllable.

[0037] S200.5 If the current state of charge is less than the high charge baseline and the expected energy replenishment value is less than the low replenishment baseline, the partitioning device is in a deep sleep state; deep sleep state refers to the working mode in which the intelligent parking space management device retains only the most basic wake-up function and all other functional modules are turned off.

[0038] When the current state of charge of the energy storage battery is less than the high-charge baseline, it means that the device's existing energy is insufficient; when the expected energy replenishment value is less than the low-replenishment baseline, it indicates that future energy replenishment is impossible. The combination of these two factors indicates that the device's existing energy is insufficient and there is no potential for growth. If the device remains in a shallow dormant state, it will run out of energy due to the lack of replenishment, causing it to be unable to detect the arrival of a vehicle. Therefore, it must enter a deep dormant state.

[0039] The current state of charge (SOC) value of the energy storage battery in S200.1 above is obtained by reading the current remaining charge percentage from the internal register of the energy storage battery in real time through the monitoring interface of the battery management system using the I2C bus protocol.

[0040] The high-capacity baseline is dynamically updated as follows: The actual capacity of the battery is recorded each time it is fully charged; its ratio to the nominal capacity is calculated as a quantitative indicator of battery health; and this ratio is linearly scaled to obtain the current high-capacity baseline. As the battery ages, its actual usable capacity decreases, and the absolute energy corresponding to the same percentage of state of charge decreases. To ensure sufficient energy reserves for the device on aging batteries, the high-capacity baseline percentage value needs to be linearly scaled, such as 80% for lithium iron phosphate batteries and 85% for ternary lithium batteries.

[0041] The low energy replenishment baseline in S200.2 above is dynamically updated as follows: First, the expected daily energy replenishment values ​​for the same historical period are classified according to weather type, and the average value of the expected value under each weather type is calculated as the replenishment baseline value; then, the current weather type is obtained through local sensors or network interfaces, and the corresponding replenishment baseline value is selected as the current low energy replenishment baseline. Because daily energy replenishment is similar under the same weather type, dynamically selecting the baseline value according to weather type allows the low energy replenishment determination to better reflect the actual weather conditions.

[0042] The S300 monitors the duration of time a parking space remains empty in real time, and calculates time-sensitive thresholds based on the current state of charge of the energy storage battery and the frequency of vehicle entry and exit, dynamically adjusting the device's sleep state level switching.

[0043] Given that intelligent parking management devices need to sense vehicle arrival and respond promptly, but vehicle arrival is random, excessive energy consumption during full-function operation and blind hibernation may prevent timely wake-up. Therefore, it is necessary to predict the arrival time of the next vehicle and dynamically adjust the hibernation depth during periods without vehicles. The vehicle entry and exit frequency is obtained as follows: a 24-hour day is divided into 24 consecutive time periods according to natural hours.

[0044] The total parking time of vehicles in each time period of the day is counted, the average parking time of each time period is calculated, and the reciprocal of the average parking time is taken as the vehicle entry and exit frequency.

[0045] For parking space occupancy events, the average parking time per vehicle within a certain time period is the average time each vehicle occupies a parking space. Since only one vehicle can occupy a parking space at a time, the number of vehicles served per unit time is inversely proportional to the average parking time. The shorter the average parking time, the more frequent the vehicle entry and exit; the longer the average parking time, the sparser the vehicle entry and exit.

[0046] The calculation time-sensitive threshold includes: S300.1, obtaining the current state of charge value of the energy storage battery, and calculating the normal working time that the remaining energy can maintain based on the average power consumption of the device in zero sleep state, as the energy depletion time.

[0047] S300.2. Take the reciprocal of the vehicle entry and exit frequency of the current time period and convert it into the average time interval without vehicles, which is used as the statistical arrival interval. The statistical arrival interval refers to the average time interval between the current vehicle leaving the parking space and the next vehicle possibly entering the parking space. It represents the average duration during the current time period when the parking space is vacant, that is, the waiting time required for the next vehicle to arrive.

[0048] S300.3. Take the smaller value between the energy depletion time and the statistical arrival interval as the current time-sensitive threshold.

[0049] The average power consumption in the zero sleep state in S300.1 above refers to the average electrical energy consumed per unit time when the device is in the zero sleep state. It is obtained by monitoring the total current and system voltage in real time through the device's built-in current sampling circuit, calculating the instantaneous power consumption through the power calculation formula, and taking the sliding average value of the instantaneous power consumption over the past 10 minutes.

[0050] The remaining energy can sustain normal operation for the duration that the device can maintain while remaining in a zero-dormancy state. The calculation steps are as follows: calculate the product of the rated energy value provided in the battery specification, the current state of charge value of the energy storage battery, and the current terminal voltage to obtain the remaining energy. Divide the remaining energy by the average power consumption to obtain the remaining energy can sustain normal operation.

[0051] In S300.3 above, the energy depletion time is determined by the current battery level and power consumption. It represents the duration the device can continue operating without replenishing energy. If the period without a vehicle exceeds this time, the device will shut down due to battery depletion and completely lose its sensing capability. The arrival interval represents the time when the next vehicle may arrive. If the period without a vehicle exceeds this interval, a vehicle may arrive at any time, and the device should maintain sufficient wake-up capability to respond immediately. The shorter of the two times will trigger the risk event first: either the battery depletes first, or the vehicle arrives first. This is determined by the critical point of arrival based on the weakest link principle.

[0052] The state of charge (SOC) of an energy storage battery is not constant but continuously depletes over time. Even if the initial energy supply is sufficient, it may enter a state of energy shortage after a period of standby consumption. Therefore, the sleep state level switching needs to be continuously adjusted according to energy consumption. The dynamic adjustment device for the sleep state level switching includes: real-time monitoring of the duration of no vehicles in the parking space. If the duration of no vehicles exceeds a time-sensitive threshold, the control device switches to a deeper sleep state level in a progressive sequence. If the current state is deep sleep, it remains in deep sleep.

[0053] When the duration of vehicle-free operation exceeds a time-sensitive threshold, it indicates that if the threshold is based on the energy depletion time, continuing to maintain the current state will lead to power depletion and device shutdown; if the threshold is based on the statistical arrival interval, continuing to maintain the current state may result in the device not responding in time when a vehicle arrives. Therefore, exceeding the time-sensitive threshold means that the current state is no longer safe and a deeper sleep level must be switched to reduce power consumption.

[0054] If a vehicle is detected entering a parking space, the device is immediately restored to a zero-sleep state, and the count of the duration without a vehicle is reset.

[0055] The entry of a vehicle into a parking space is an external event that marks the end of a car-free period and the beginning of a car-available period. During car-free periods, the device may be in a shallow or deep sleep state for energy conservation, extending its standby time as much as possible while waiting for a vehicle to arrive. Once a vehicle enters the parking space, the device immediately switches from energy-saving standby to providing service.

[0056] S400 If the expected energy supply value is detected to be in a low supply state and the state of charge of the energy storage battery is in a low charge state, then the state priority reduction strategy is activated.

[0057] The state reduction priority strategy has a higher priority than the conventional progressive switching strategy based on time-sensitive thresholds. When both triggering conditions exist, the state reduction priority strategy is executed first.

[0058] Under extreme conditions, where the current power is already insufficient and energy replenishment is very limited in the future, conventional progressive hibernation may still be insufficient to prevent the power from being depleted. Therefore, the activation of the state-priority reduction strategy is as follows: The current remaining energy and arrival interval of the energy storage battery are obtained. Taking the current moment as the starting point and the arrival interval as the projection window, the predicted instantaneous solar power value for each moment within each projection window is obtained. The predicted instantaneous solar power value is obtained according to the method in step S100.4.

[0059] The simulation window is discretized into multiple consecutive time intervals. Taking the current remaining energy as the initial value, the remaining energy at each discrete moment is calculated segment by segment based on the predicted instantaneous solar power and the device's operating power consumption in that time interval.

[0060] The above discretization is divided into multiple consecutive time intervals of equal length Δt, such as 1 minute.

[0061] The change in energy of the device during each discrete time period is calculated using the following formula: .

[0062] in, This refers to the average solar power supply during that time period, which is the average power that the solar panels actually charge the energy storage battery during that period. The power consumption of the device during this time period is dynamically determined based on the sleep state level the device is in during the simulation: if the device has not yet triggered a switch to a deeper sleep level at a certain point in the simulation, the power consumption corresponding to the current sleep level is maintained; if the device switches to a deeper sleep level during the simulation due to a decrease in power or a time-related trigger, the power consumption value corresponding to the new level is used to continue the simulation from the time of the switch. Specifically, This represents the total energy supplied by solar energy during that period. This represents the total energy consumed by the device during that time period. A negative result indicates that the device's energy consumption exceeds its replenishment, resulting in a decrease in remaining energy; a positive result indicates an increase in remaining energy.

[0063] The remaining energy at each discrete moment is calculated by recursively step by step, which is the sum of the remaining energy at the previous moment and the change in energy.

[0064] If, at any discrete moment within the simulation window, the remaining energy is lower than the energy safety threshold, then the expected energy supply value is determined to be in a low supply state and the state of charge of the energy storage battery is in a low charge state, and the state-priority reduction strategy is initiated.

[0065] It should be noted that if the expected energy supply value is less than the low supply baseline, it is determined to be a low supply state; if the current state of charge value is less than the high charge baseline, it is determined to be a low charge state; however, when both the low supply state and the low charge state are met, the risk of energy crisis is further confirmed. Based on this, if the remaining energy is lower than the energy safety threshold at any discrete moment within the simulation window, the final decision is to activate the state reduction priority strategy.

[0066] The energy safety threshold is pre-calibrated based on the specifications of the energy storage battery used in the device and the minimum operating voltage requirements of the hardware platform. Specifically, it is the maximum energy consumption required for the device to complete a full wake-up and vehicle detection under the rated operating voltage.

[0067] If the remaining energy is below the energy safety threshold, it indicates that the device is at risk of running out of energy within the statistical arrival interval. Therefore, a state priority reduction strategy must be initiated to force a switch to a deeper hibernation level in order to maximize the survival time of the core detection function.

[0068] The state priority reduction strategy includes: after the state priority reduction strategy is activated, the system will force a switch to a deeper level step by step from the current level to the deepest level in a progressive order from zero sleep state to deep sleep state, until the deep sleep state is reached.

[0069] In shallow sleep mode, the wireless communication module is turned off first, while the sensor detection function is retained.

[0070] In intelligent parking management devices, sensor detection is essential for the device to sense vehicle entry and complete basic tasks. Without this function, the device cannot detect parking space status or wake up in time. The wireless communication module is used for additional services such as status reporting and remote querying. Turning off this module when there is no vehicle will not affect the device's ability to detect vehicle arrival.

[0071] When the expected energy supply is lower than the low supply baseline, the state of charge (SOC) value corresponding to the transition from shallow to deep sleep is increased. Increasing the SOC value corresponding to the transition from shallow to deep sleep allows for earlier entry into the extreme energy-saving mode, reserving a greater safety margin for energy crises.

[0072] If the expected energy supply is below the low supply baseline, it indicates that energy supply will be severely insufficient in the near future, such as during nighttime or rainy weather. In this case, the device cannot rely on external supply to maintain operation and can only rely entirely on the battery's stored power.

[0073] Without anticipated resupply, the survival time supported by the same amount of electricity is significantly shortened. If deep hibernation is triggered under normal charging conditions, the electricity may be exhausted shortly after reaching the normal charging state, leaving no time to wait for the vehicle to arrive in deep hibernation.

[0074] If the conditions for initiating the downgrade priority strategy are no longer met after the switch, downgrade will stop and the current level will be maintained. If the energy crisis continues, downgrade will continue; if the energy crisis is alleviated, downgrade will stop in a timely manner, ensuring that core functions are maintained at the shallowest possible hibernation level.

[0075] The state of charge (SOC) and expected energy replenishment of the energy storage battery change dynamically over time. During degradation, due to reduced power consumption, energy consumption slows down, and the downward trend of SOC may slow down or even stabilize. As time progresses, improved weather conditions may increase the expected energy replenishment. After each switch, the system enters a new operating state, and the original startup conditions may change, requiring reassessment. During the assessment, the total available time is calculated based on the actual average power consumption of the current hibernation level.

[0076] If a vehicle is detected entering a parking space during the downgrade process, the system immediately reverts to the zero sleep state and exits the downgrade priority strategy.

[0077] In deep sleep mode, the state of charge decreases slowly due to extremely low power consumption, which may meet the conditions for exiting deep sleep mode. Therefore, it is necessary to set the state of charge threshold for entering deep sleep mode to 30% and the state of charge threshold for exiting deep sleep mode and returning to shallow sleep mode to 35% to form a hysteresis interval to prevent repeated state switching.

[0078] S500 records the state of charge of the energy storage battery, the expected energy replenishment value, and the actual dormancy state after each state adjustment, and predicts the subsequent dynamic adjustment device dormancy state switching based on these records.

[0079] By recording historical state adjustment data, long-term trends in energy changes can be identified, allowing for proactive optimization of dormancy switching strategies before future risks actually occur, thus upgrading from passive response to proactive prediction. The prediction of subsequent dynamic adjustment of the device's dormancy state switching includes: obtaining the actual output power of the solar panels within a preset time period from historical records to obtain a light intensity sequence.

[0080] The light intensity sequence is processed using a single exponential smoothing method, and the smoothed value at the current moment is calculated recursively.

[0081] It should be noted that the smoothed value at the current moment is a weighted sum of the actual light intensity at the current moment and the smoothed value at the previous moment. The weight of the actual light intensity at the current moment is determined by the smoothing coefficient, and the weight of the smoothed value at the previous moment is 1 minus this coefficient. For the first moment of the sequence, since there is no smoothed value from the previous moment, the actual light intensity at that moment is taken as the initial smoothed value. Subsequent moments are calculated sequentially according to this recursive relationship, ultimately yielding the smoothed value corresponding to the current moment.

[0082] The smoothing coefficient is between 0 and 1. The closer the smoothing coefficient is to 1, the more sensitive the smoothing value is to recent data changes, resulting in a faster response but poorer smoothness; the closer it is to 0, the smoother the smoothing value, the better the noise filtering effect, but the response is relatively lagging. It is determined as follows: a fixed smoothing coefficient is preset based on empirical values ​​from the application scenario of the device. For example, for areas with drastic changes in lighting, a value of [missing value] is used. Improve response speed; for areas with relatively stable lighting, take Enhance the filtering effect.

[0083] After applying exponential smoothing, the above-mentioned random noise can be effectively filtered out, and the true trend of light intensity variation can be extracted.

[0084] If the current smoothing value is lower than the previous smoothing value, and this trend continues for a preset number of times, then the future light intensity is predicted to continue to decrease.

[0085] When the smoothed value is lower than the smoothed value of the previous moment multiple times consecutively, it indicates that the decrease in light intensity is persistent. The causes of this persistent decrease are often systemic, such as a gradual decrease in the solar altitude angle or continuous changes in weather systems, rather than being caused by instantaneous random factors. Therefore, only a sustained decrease of a predetermined number of times is sufficient to support the prediction that the light intensity will continue to weaken in the future.

[0086] The preset number of times is an empirical value and can be adjusted according to the application scenario of the device. Too few times are easily affected by random fluctuations, while too many times will result in a delayed response. Specifically, the sampling period is 1 minute, and the preset number of times is 5. This 5-minute period can eliminate instantaneous interference such as birds and falling leaves, and allows sufficient response time before the light intensity decreases significantly, thus balancing reliability and timeliness.

[0087] When it is predicted that the future light intensity will continue to decrease, the trigger threshold for deep sleep state is increased in advance, and the sensor sampling interval is extended; the quantitative standard for continuous decrease is the same as the preset number of times used when predicting that the future light intensity will continue to decrease.

[0088] The predicted continued weakening of sunlight indicates that future energy supplies will become increasingly scarce. If we wait until the energy shortage becomes severe before taking action, it may be too late to prevent the equipment from shutting down. Therefore, we need to adjust our strategy in advance.

[0089] Increasing the deep sleep trigger threshold allows the device to enter the extreme power-saving mode earlier, sacrificing some shallow sleep time in exchange for a longer survival time in the deep sleep state; extending the sensor sampling interval can further reduce power consumption without affecting basic sensing capabilities.

[0090] The trigger threshold is the critical value of the energy storage battery's state of charge corresponding to the intelligent parking space management device switching from a shallow sleep state to a deep sleep state. In the sleep level classification rules, when the current state of charge is less than the high charge baseline and the expected energy replenishment value is less than the low replenishment baseline, the device enters a deep sleep state. This determined state of charge threshold is the trigger threshold. Under normal operating conditions, the trigger threshold is 20%. If it is predicted that future sunlight will continue to weaken, it can be adjusted to 30%, allowing the device to enter a deep sleep state earlier when the battery has more power.

[0091] The predicted state of charge (SOC) value is obtained by predicting the historical SOC change trend, and the actual SOC value is monitored in real time to calculate the SOC deviation between the two.

[0092] Preferably, the least squares method is used to fit the linear regression method to perform trend fitting on the historical state of charge data, and extrapolation is used to predict the state of charge value after the current moment.

[0093] Based on the energy deviation between the expected energy supply and the actual energy supply, as well as the charge deviation, a graded compensation wake-up mechanism is triggered.

[0094] Specifically, the graded triggering compensatory wake-up mechanism includes: continuously monitoring the rate of change of charge deviation; if the rate of change of charge deviation is continuously positive and the actual state of charge value shows a continuous downward trend, the triggering device is temporarily woken up from the current dormant state.

[0095] If the rate of change of energy deviation remains positive, it indicates that the actual lighting conditions are worse than expected at an accelerated pace. If the actual lighting intensity continues to decrease, it confirms that the device is in a deteriorating lighting environment. The combination of these two factors indicates that the weather changes are deviating from historical patterns. If the device is not promptly re-predicted, it may be making erroneous decisions based on the incorrect low-supply prediction, thus triggering a temporary wake-up.

[0096] The aforementioned "continuously positive" means that the rate of change of charge deviation is positive across multiple consecutive sampling periods; similarly, the "continuously decreasing" means that the actual state of charge value in each sampling period is lower than that in the previous sampling period. Specifically, when the sampling period is 1 minute, the sensor noise level can be naturally attenuated within 30 seconds, as measured by actual tests. Taking 3 to 5 sampling periods ensures that the cumulative duration of continuous monitoring is 3 to 5 minutes, which can both filter out instantaneous disturbances and intervene before an energy crisis occurs.

[0097] After being awakened, the steps of predicting the expected energy supply value for the future time period, classifying the hibernation state levels, calculating the time-sensitive threshold, and dynamically adjusting the switching of the device's hibernation state levels are re-executed.

[0098] The trigger condition for a temporary wake-up is that the deviation between the original predicted state and the actual state continues to widen. The wake-up only interrupts the current state. If the old predicted data and decision results are still used after the wake-up, the deviation still exists, and the wake-up becomes meaningless. Therefore, the complete decision-making process must be re-executed after the wake-up.

[0099] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0100] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0101] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0102] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0103] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-level dormant energy management method for intelligent parking space management devices, characterized in that: include: Based on the photoelectric conversion parameters of solar panels and historical light change trends, the expected energy supply value in the future time period is predicted. Based on the current state of charge of the energy storage battery and the expected energy replenishment, the intelligent parking space management device is divided into hibernation state levels, including zero hibernation state, shallow hibernation state and deep hibernation state; Real-time monitoring of the duration of parking space vacancy, combined with the current state of charge of the energy storage battery and the frequency of vehicle entry and exit, calculates the time-sensitive threshold and dynamically adjusts the device's sleep state level switching. If the expected energy supply value is detected to be in a low supply state and the state of charge of the energy storage battery is in a low charge state, then the state priority reduction strategy is activated. Record the state of charge, expected energy replenishment, and actual dormancy state of the energy storage battery after each state adjustment, and predict the subsequent dynamic adjustment device dormancy state switching based on these records; The prediction of subsequent dynamic adjustment device dormancy state switching includes: Obtain the actual output power of the solar panel within a preset time period in the historical records to obtain the light intensity sequence; The light intensity sequence is processed using a single exponential smoothing method, and the smoothed value at the current moment is calculated recursively. If the current smoothing value is lower than the previous smoothing value, and this trend continues for a preset number of times, then the future light intensity is predicted to continue to decrease. When it is predicted that the light intensity will continue to decrease in the future, the trigger threshold for deep sleep state is increased in advance, and the sensor sampling interval is extended. The predicted state of charge (SOC) value is obtained based on the historical SOC change trend, and the actual SOC value is monitored in real time to calculate the SOC deviation between the two. Based on the energy deviation between the expected energy supply and the actual energy supply, as well as the charge deviation, a graded compensation wake-up mechanism is triggered.

2. The multi-level dormant energy management method for the intelligent parking space management device according to claim 1, characterized in that: The predicted energy replenishment expectations for the future time period include: Obtain the geographical location and future time period timestamp of the intelligent parking space management device, and calculate the theoretical maximum light reference value for each moment in the future time period through solar radiation calculation; The historical attenuation factor of meteorological interference degree is obtained by calculating the ratio of the actual power of the same period in history to the theoretical maximum irradiance reference value at the corresponding time. The ratio of the actual power at the current moment to the theoretical maximum illumination reference value at the current moment is calculated in real time, and the ratio is filtered by moving average to obtain the real-time trend correction coefficient. The instantaneous power prediction value at each moment is obtained by multiplying the theoretical maximum illumination reference value, the historical attenuation factor at the corresponding time, and the current real-time trend correction coefficient. Integrating the instantaneous power forecast over a future time period yields the expected energy supply value.

3. The multi-level dormant energy management method for the intelligent parking space management device according to claim 1, characterized in that: The method of classifying the intelligent parking space management device into sleep state levels includes: Obtain the current state of charge (SOC) value of the energy storage battery and compare it with a high-charge baseline that is dynamically updated based on the battery's health status and historical charge and discharge data. Obtain the predicted expected energy supply value and compare it with the low supply baseline generated based on the statistical patterns of energy supply in the same period of history. If the current state of charge is greater than or equal to the high charge baseline, then the partitioning device is in a zero sleep state regardless of the expected energy supply value. If the current state of charge is less than the high charge baseline but the expected energy supply is greater than or equal to the low supply baseline, the partitioning device is in a shallow sleep state. If the current state of charge is less than the high charge baseline and the expected energy replenishment is less than the low replenishment baseline, the partitioning device is in a deep sleep state.

4. The multi-level dormant energy management method for the intelligent parking space management device according to claim 1, characterized in that: The method for obtaining the vehicle entry and exit frequency is as follows: Divide a day's 24 hours into 24 consecutive time periods according to the natural hours; The total parking time of vehicles in each time period of the day is counted, the average parking time of each time period is calculated, and the reciprocal of the average parking time is taken as the vehicle entry and exit frequency.

5. The multi-level dormant energy management method for the intelligent parking space management device according to claim 1, characterized in that: The time-sensitive threshold for calculation includes: Obtain the current state of charge value of the energy storage battery, and calculate the normal operation time that the remaining energy can sustain based on the average power consumption of the device in zero sleep state, as the energy depletion time; The reciprocal of the vehicle entry and exit frequency of the current time period is converted into the average vehicle-free time interval, which is used as the statistical arrival interval. The smaller value between the energy depletion time and the statistical arrival interval is taken as the current time-sensitive threshold.

6. The multi-level dormant energy management method for the intelligent parking space management device according to claim 5, characterized in that: The dynamic adjustment device's sleep state level switching includes: The device monitors the duration of time the parking space is empty in real time. If the duration of time the parking space is empty exceeds the time-sensitive threshold, the control device switches to a deeper sleep state level in a progressive order. If the current state is deep sleep, it remains in deep sleep. If a vehicle is detected entering a parking space, the device is immediately restored to a zero-sleep state, and the count of the duration without a vehicle is reset.

7. The multi-level dormant energy management method for the intelligent parking space management device according to claim 6, characterized in that: The process of initiating the downstate priority strategy is as follows: Obtain the current remaining energy and statistical arrival interval of the energy storage battery. Starting from the current time, use the statistical arrival interval as the simulation window to obtain the predicted value of the instantaneous solar power at each time within each simulation window. The simulation window is discretized into multiple consecutive time intervals. Taking the current remaining energy as the initial value, the remaining energy at each discrete moment is calculated segment by segment based on the predicted instantaneous solar power in each time interval and the operating power consumption of the device in that time interval. If, at any discrete moment within the simulation window, the remaining energy is lower than the energy safety threshold, then the expected energy supply value is determined to be in a low supply state and the state of charge of the energy storage battery is in a low charge state, and the state-priority reduction strategy is initiated.

8. The multi-level dormant energy management method for the intelligent parking space management device according to claim 7, characterized in that: The state priority reduction strategy includes: After the state priority reduction strategy is activated, the system will forcibly switch to a deeper level from the current level in a progressive order from zero sleep state to deep sleep state, until deep sleep state is reached. In shallow sleep mode, the wireless communication module is turned off first, while the sensor detection function is retained; When the expected energy supply value is lower than the low supply baseline, increase the state of charge value corresponding to the transition from shallow sleep state to deep sleep state; If the conditions for initiating the downgrade priority strategy are no longer met after the switch, then the downgrade is stopped and the current level is maintained; If a vehicle is detected entering a parking space during the downgrade process, the system immediately reverts to the zero sleep state and exits the downgrade priority strategy.

9. The multi-level dormant energy management method for the intelligent parking space management device according to claim 1, characterized in that: The graded triggering compensatory wake-up mechanism includes: The rate of change of charge deviation is continuously monitored. If the rate of change of charge deviation is continuously positive and the actual state of charge value shows a continuous downward trend, the device is triggered to temporarily wake up from the current dormant state. The rate of change of energy deviation is continuously monitored. If the rate of change of energy deviation remains positive and the actual light intensity shows a continuous decreasing trend, the device is temporarily awakened from its current dormant state. After being awakened, the steps of predicting the expected energy supply value for the future time period, classifying the hibernation state levels, calculating the time-sensitive threshold, and dynamically adjusting the switching of the device's hibernation state levels are re-executed.

Images