Internet of things based charging pile operation efficiency evaluation and optimization management system

Abstract

The present application relates to the technical field of intelligent management of charging facilities, in particular to a charging pile operation efficiency evaluation and optimization management system based on Internet of Things, which determines false occupation events by acquiring ground lock state, parking space image and electrical parameters, calculates opportunity loss value based on historical same period service probability, average single service income and false occupation duration, and then generates occupation efficiency loss index, when the occupation efficiency loss index exceeds the first loss threshold, sends integral reward reminding related to the occupation efficiency loss index, when the occupation efficiency loss index exceeds the second loss threshold, adds floating additional fee related to the overrun proportion linearly on the basis of the occupation fee, until the occupation ends, realizes accurate identification of false occupation, quantitative evaluation of opportunity loss and hierarchical flexible intervention, and improves the operation efficiency of charging piles and resource utilization.

Description

Technical Field

[0001] This invention relates to the field of intelligent management technology for charging facilities, and more specifically, to an Internet of Things-based management system for evaluating and optimizing the operational efficiency of charging piles. Background Technology

[0002] With the rapid growth of new energy vehicle ownership, existing technologies have explored various approaches to address the issue of charging pile occupancy. Chinese patent CN118840713A discloses a method, device, and computer equipment for detecting illegal vehicle occupancy. By acquiring parking space image information, it identifies the target charging space in an occupied state and the type of vehicle on it. Then, based on the occupancy duration and vehicle type, it determines whether illegal occupancy has occurred, thus initially realizing the visual detection of occupancy events. Chinese patent CN120198169A discloses a charging station occupancy management method, device, electronic terminal, and storage medium. By collecting historical operating data and using a charging demand prediction model, it obtains dynamic free parking time and dynamic pricing, and then calculates the occupancy fee, introducing a dynamic pricing mechanism into occupancy management. Furthermore, Chinese patent CN119379492A discloses an IoT-based intelligent management method and system for charging piles, which aims to optimize the operational efficiency and profitability of charging stations from the perspective of charging pile status analysis and processing. However, considering the aforementioned technological achievements and the overall development status of charging station operation and management, existing solutions have significant shortcomings in the following three aspects. First, the detection of parking space occupation mostly relies on single conditions, such as judging violations solely based on the duration of parking space occupancy and vehicle type. In false occupation scenarios where the charging gun is inserted but there is no effective charging current, the parking lock may be locked and the complete vehicle outline may be present, but electrical parameters indicate no effective charging. Such behavior of occupying a space under the guise of inserting the charging gun is very easy to miss by existing detection mechanisms. Just as image recognition can determine vehicle type and occupancy duration, it is difficult to establish a correlation between the insertion status and effective charging current, creating a blind spot for misjudgment of someone occupying a space but not actually charging. Second, even if an occupation event is detected, operation and management methods mostly remain at the level of single economic penalties, such as charging a fixed standard occupation fee. This cannot dynamically adjust the billing intensity according to time and location, and lacks a complete chain from quantifying to responding to opportunity loss. The first issue is the lack of objective quantitative data on the actual revenue loss caused by vehicles occupying charging spaces, how to calculate the loss, and the appropriate level of intervention. Secondly, current disposal strategies are mostly rigid charging models, failing to provide tiered and flexible intervention based on the degree of efficiency loss. There is a lack of positive guidance mechanisms for minor occupancy, and insufficient economic constraints for serious occupancy, leading to repeated instances where fines are ineffective. Therefore, how to build a comprehensive management mechanism based on multi-source IoT data fusion, encompassing accurate determination of occupancy events, quantification of opportunity revenue loss, generation of tiered intervention strategies, and closed-loop feedback optimization, enabling charging pile operation to shift from passively responding to occupancy to proactively assessing losses and flexibly adjusting levels, has become a pressing issue in the field of intelligent operation and maintenance of charging facilities. To address this problem, we provide an IoT-based charging pile operation efficiency evaluation and optimization management system. Summary of the Invention

[0003] The purpose of this invention is to provide an Internet of Things-based charging pile operation efficiency evaluation and optimization management system to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, an IoT-based charging pile operation efficiency evaluation and optimization management system is provided, including: The IoT data acquisition unit is used to acquire charging pile ground lock status signals, parking space images, and electrical parameters including plug status and real-time charging current. The false parking space determination unit determines whether a false parking space event has occurred based on the charging pile ground lock status signal, parking space image, and electrical parameters including plug status and real-time charging current. The efficiency loss index generation unit is used to obtain the probability of a charging pile successfully charging a parking space in the same week and time period in history after determining that a false parking space occupancy event has occurred. This probability is used as an opportunity service probability factor. The unit also obtains the recent average single service revenue of the charging pile. The unit multiplies the duration of the false parking space occupancy, the opportunity service probability factor, and the average single service revenue to obtain the opportunity revenue loss value. Finally, the unit calculates the proportion of the opportunity revenue loss value in the theoretical maximum service revenue of the charging pile within a preset evaluation period to generate the parking space occupancy efficiency loss index. The optimization strategy execution unit is set with a first loss threshold and a second loss threshold. When the occupancy efficiency loss index exceeds the first loss threshold but does not exceed the second loss threshold, a parking reminder instruction containing points rewards is generated. The amount of the points rewards is positively correlated with the occupancy efficiency loss index. When the occupancy efficiency loss index exceeds the second loss threshold, a dynamic rate adjustment instruction is generated, which adds a floating surcharge to the current occupancy fee unit price on the base rate. The amount of the floating surcharge is determined linearly according to the proportion by which the occupancy efficiency loss index exceeds the second loss threshold, and the billing module bills the current occupancy event in real time until the false occupancy determination unit determines that the false occupancy event has ended.

[0005] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention collects parking lock status, parking space images, and electrical parameters through the Internet of Things (IoT), and correlates the charging gun status with the charging current for verification. It accurately identifies false parking space occupancy events where the gun is present but the battery is not charged, solving the problem that traditional single-image detection cannot distinguish between real charging and false occupancy. Based on this, it quantifies opportunity revenue loss and generates an occupancy efficiency loss index according to the historical service probability and average single-use revenue. This provides an objective measurement of the actual impact of occupancy on operations. Then, it implements tiered intervention based on the degree of index exceedance. For mild exceedance, it pushes a car-moving reminder with points rewards to guide users to move their cars. For severe exceedance, it activates a dynamic fee rate, adding a floating surcharge linearly related to the exceedance ratio to the basic occupancy fee until the occupancy ends. This effectively reduces resource waste and improves the turnover efficiency and operational revenue of charging piles. Attached Figure Description

[0006] Figure 1 This is an overall block diagram of the present invention. Detailed Implementation

[0007] 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.

[0008] This invention provides an IoT-based charging pile operation efficiency evaluation and optimization management system. Please refer to [link / reference]. Figure 1 As shown, it includes: The IoT data acquisition unit is used to acquire charging pile ground lock status signals, parking space images, and electrical parameters including plug status and real-time charging current. The false parking space determination unit determines whether a false parking space event has occurred based on the charging pile ground lock status signal, parking space image, and electrical parameters including plug status and real-time charging current. The efficiency loss index generation unit is used to obtain the probability of a charging pile successfully charging a parking space in the same week and time period in history after a false parking space occupancy event is determined to have occurred. This probability is used as an opportunity service probability factor. The unit also obtains the recent average single service revenue of the charging pile. The unit multiplies the duration of the false parking space occupancy, the opportunity service probability factor, and the average single service revenue to obtain the opportunity revenue loss value. Finally, the unit calculates the proportion of the opportunity revenue loss value in the theoretical maximum service revenue of the charging pile within the preset evaluation period to generate the parking space occupancy efficiency loss index. The optimization strategy execution unit is set with a first loss threshold and a second loss threshold. When the parking space efficiency loss index exceeds the first loss threshold but does not exceed the second loss threshold, a parking space relocation reminder instruction containing points rewards is generated. The amount of points rewards is positively correlated with the parking space efficiency loss index. When the occupancy efficiency loss index exceeds the second loss threshold, a dynamic rate adjustment instruction is generated, which adds a floating surcharge to the current occupancy fee unit price on the base rate. The amount of the floating surcharge is determined linearly according to the proportion of the occupancy efficiency loss index exceeding the second loss threshold, and the billing module bills the current occupancy event in real time until the false occupancy determination unit determines that the false occupancy event has ended.

[0009] After acquiring the parking space image, the IoT data acquisition unit performs vehicle contour extraction processing on the parking space image to obtain a contour image. At the same time, it extracts the vehicle color features from the parking space image and correlates the plug status in the electrical parameters with the real-time charging current to determine whether the vehicle corresponding to the contour image and the electrical parameters belong to the same charging pile service object. The charging pile ground lock status signal, contour image, vehicle color features and judgment results are sent together to the false occupancy determination unit.

[0010] The false occupancy determination unit generates an initial false occupancy event determination result based on the charging pile ground lock status signal being locked, the presence of a complete vehicle outline in the contour image, and the judgment result indicating no effective charging current in the electrical parameters. It then compares the result with the vehicle color characteristics in the historical database. If the vehicle corresponding to the color characteristics has a short-term occupancy behavior in the historical records, the confidence level of the initial false occupancy event determination result is increased; otherwise, the confidence level is decreased. When the confidence level exceeds the dynamically adjusted determination threshold, a false occupancy event is finally determined to have occurred. The determination threshold is obtained by periodically adjusting in reverse based on the number of misjudgment records in the charging pile's historical determination.

[0011] From the historical operation database, all historical time period records with the same weekday type attribute and within the same time period as the current time are selected. The number of times the charging pile was idle within the historical time period records is counted, and the number of times charging service orders were actually generated by the charging pile during the idle period is further counted. The ratio of the number of charging service orders generated to the number of times the charging pile was idle is calculated as the basic probability factor. Historical weather data and holiday marking data for the same period are obtained, and the basic probability factor is weighted and corrected according to the preset mapping rules. The mapping rules define the influence weight of different weather conditions and whether it is a holiday on users' charging intentions. The weighted and corrected basic probability factor is then subjected to a time decay smoothing process to obtain the opportunity service probability factor.

[0012] The process of obtaining the recent average revenue per service for charging stations includes: Extract all charging service orders completed by charging stations within a preset recent time period. For each charging service order, obtain the difference in charging capacity of the vehicle battery from the start to the end of charging, and obtain the applicable rate unit price for the charging period of the charging service order. Multiply the charging capacity difference by the rate unit price to obtain the basic electricity fee revenue of the charging service order. Then, add the service fee paid for the charging service order to the basic electricity fee revenue to obtain the total service revenue of the charging service order. Sum the total service revenue of all charging service orders and divide it by the total number of charging service orders to obtain the initial average revenue. Finally, multiply the initial average revenue by a preset coefficient reflecting the recent revenue change trend to obtain the recent average single service revenue of the charging station. The coefficient of the recent revenue change trend is calculated based on the comparison between the historical average revenue of the previous period and the initial average revenue.

[0013] Within a preset evaluation period, the predefined charging pile is always in a serviceable and occupied state, which is taken as the ideal state. The theoretical maximum number of services is obtained by dividing the total duration of the preset evaluation period by the historical average duration of a single service of the charging pile. Then, the theoretical maximum number of services is multiplied by the average revenue per service to obtain the theoretical maximum serviceable revenue. Finally, the opportunity revenue loss value is divided by the theoretical maximum serviceable revenue to obtain the original occupancy efficiency loss ratio. The original occupancy efficiency loss ratio is then input into a transformation function that is normalized based on historical occupancy efficiency loss data to output the final occupancy efficiency loss index.

[0014] The process of setting the first loss threshold and the second loss threshold includes: Collect all occupancy efficiency loss indices generated by charging piles during their historical normal operating cycles, sort all occupancy efficiency loss indices by numerical value, select the occupancy efficiency loss index located at the first specific quantile position in the sorted sequence as the initial value of the first loss threshold, and select the occupancy efficiency loss index located at the second specific quantile position in the sorted sequence as the initial value of the second loss threshold, where the second specific quantile position is higher than the first specific quantile position, and introduce a time-based decay factor into each threshold. After weighted sorting and calculation, update the specific values ​​of the first loss threshold and the second loss threshold.

[0015] When the optimization strategy execution unit generates a parking reminder instruction that includes a points reward, the process for determining the amount of the points reward is as follows: The system calculates the excess amount of the parking space occupancy efficiency loss index that exceeds the first loss threshold, and calculates the proportion of the excess amount to the first loss threshold. A base reward value is defined, and the base reward value is multiplied by the proportion to obtain the calculated reward value. At the same time, the system queries the historical credit records of the user account that triggered the false parking space occupancy event, sets a reward coefficient based on the historical credit records, multiplies the calculated reward value by the reward coefficient to obtain the final reward amount, and sends a parking space relocation reminder instruction containing the reward amount to the user terminal associated with the charging pile.

[0016] When the optimization strategy execution unit generates a dynamic rate adjustment instruction, the process for determining the floating surcharge amount includes: Calculate the second excess amount of the occupancy efficiency loss index exceeding the second loss threshold, and calculate the excess ratio of the second excess amount to the second loss threshold. Obtain the occupancy fee unit price corresponding to the basic rate as a benchmark. Multiply the occupancy fee unit price by the excess ratio to obtain the surcharge unit price. From the moment it is determined that the occupancy efficiency loss index exceeds the second loss threshold, update the real-time occupancy fee unit price in the charging pile's billing module to the sum of the basic rate and the surcharge unit price, and continuously bill the current occupancy event according to the updated real-time occupancy fee unit price.

[0017] The optimization strategy execution unit continuously monitors the user terminal's response operations and the updated data uploaded by the IoT data acquisition unit. If it detects that the user has performed a vehicle relocation operation or that the electrical parameters show that effective charging has started, it triggers the false parking space determination unit to re-determine the current status. If the re-determination result is that the false parking space event has ended, the optimization strategy execution unit sends a stop billing instruction to the billing module and records the complete data of this false parking space event, including the duration, the final parking space efficiency loss index, and the result, into the historical operation database for subsequent updates and calculations of the opportunity service probability factor, average single service revenue, and the first and second loss thresholds.

[0018] Further explanation is needed regarding the IoT data acquisition unit's standardized preprocessing of parking space status data through multi-dimensional image analysis and spatiotemporal linkage verification of electrical parameters. This improves the accuracy of matching vehicle presence status with charging behavior, providing unambiguous and traceable basic data support for identifying false parking occupancy. After acquiring real-time parking space images through high-definition cameras mounted on the side of the parking space, the IoT data acquisition unit immediately performs vehicle contour extraction processing. This contour extraction processing employs a step-by-step execution method based on a combination of Canny edge detection and constrained region growing. First, the color parking space image is converted to a single-channel grayscale image. Then, an image filtering operation is performed using a Gaussian convolution kernel to remove interference noise such as ground reflections, shadows, and ambient light. After filtering, the Canny edge detection algorithm is used to traverse all pixels of the image, calculate the gradient magnitude and gradient direction of each pixel, and retain edge points with gradient magnitudes higher than the low threshold and connected to pixels with high thresholds. Discrete edge points are connected according to the eight-neighbor continuity rule to form closed edge curves. Using the four preset physical corner points of the charging pile parking space as spatial constraint boundaries, a region growing algorithm is executed, starting from the seed pixel at the geometric center of the parking space and gradually expanding outwards. Adjacent pixels with grayscale differences from the seed pixel within a preset fixed range are merged. Parking space lines, ground markings, manhole covers, and other debris are automatically removed, retaining only the complete vehicle-occupied area inside the parking space. Finally, morphological closing operations are performed on the edges of the vehicle-occupied area, using rectangular structuring elements to fill edge gaps and smooth contour lines, eliminating burrs and breakpoint interference, resulting in a contour image containing only the complete external contour of the vehicle. Simultaneously, the IoT data acquisition unit starts vehicle color feature extraction, which adopts a multi-region uniform sampling method based on the CIELAB color space. First, twenty sampling points are uniformly selected in a row-column matrix within the effective area of ​​the vehicle defined by the contour image. These sampling points cover different main parts of the vehicle, such as the front, body, and rear. Non-painted areas such as windows, lights, tires, and bumpers are removed using a pixel grayscale threshold, retaining only the sampling pixels corresponding to the painted surfaces of the vehicle body. The RGB color values ​​of each effective sampling pixel in the original parking space image are extracted. These RGB color values ​​are then converted to CIELAB color space values ​​using the International Commission on Illumination (CLI) standard conversion formula. The CIELAB color space includes a luminance channel and two independent chrominance channels. The two chrominance channel values ​​of each sampling pixel are extracted, and the chrominance mean of all effective sampling points is calculated. This chrominance mean is then compared with the built-in standard color sample library. Sample-by-sample comparison determines the vehicle's main color characteristics, output in standard color label format with no ambiguity. The IoT data acquisition unit performs spatiotemporal correlation judgment between the charging gun status and real-time charging current in the electrical parameters. The charging gun status includes two states: charging gun in place and not in place. The real-time charging current includes two values: zero mA and effective charging current greater than ten mA. The correlation judgment first verifies the acquisition timestamp of the charging gun status with the acquisition timestamp of the parking space image, ensuring that the time difference between the two is less than one hundred milliseconds. Then, it verifies the correspondence between the physical number of the charging pile and the physical number of the parking space to confirm that the two types of data come from the same charging pile service object. Finally, a fixed state matching rule is established, which is set as charging gun in place and real-time charging current... A current greater than 10 mA is considered valid charging activity; a current not in place or zero mA is considered invalid charging activity. The system compares the vehicle presence result corresponding to the contour image with the valid charging activity determination result of the electrical parameters. If the vehicle exists but there is no valid charging activity, the vehicle in the contour image and the electrical parameters belong to the same charging pile service object. If the vehicle does not exist or there is valid charging activity, they do not belong to the same service object. Finally, the IoT data acquisition unit packages the charging pile ground lock status signal, the processed contour image, the extracted vehicle color features, and the service object attribution determination result according to a unified acquisition timestamp and charging pile number, and transmits the data via a low-power IoT system. The communication module sends the data to the false parking space determination unit. For example, the timestamp of the parking space image acquisition is 9:10:20, the timestamp of the charging gun status acquisition is 9:10:25, the time difference meets the requirements, the charging pile number and the parking space number are both 05, the average color value of the vehicle body sampling points corresponds to the standard white label, the contour image shows that the vehicle exists, the electrical parameters show that the charging gun is not plugged in and the real-time charging current is zero mA, and it is determined that the vehicle belongs to the same service object. The four types of data are uploaded synchronously, and the entire processing flow is automatically executed by the embedded image processing chip. Contour extraction and color extraction are both accurately performed for the parking space area, and the association judgment strictly follows the spatiotemporal consistency rule to ensure that the data transmitted to the false parking space determination unit has high credibility and high matching degree.

[0019] The false parking space determination unit significantly improves the accuracy of false parking space event identification through a hierarchical determination logic that employs multi-condition rigid verification, historical behavior matching correction, dynamic confidence adjustment, and adaptive reverse updating of the determination threshold. This reduces misjudgments caused by environmental interference and temporary parking, providing a stable and reliable basis for subsequent operational efficiency loss calculations and optimization strategy triggering. After receiving the charging pile ground lock status signal, contour image, vehicle color features, and service object attribution determination results transmitted by the IoT data acquisition unit, the false parking space determination unit generates the initial false parking space event determination result, including the following steps: The system verifies the encoded value of the charging pile's parking lock status signal. The locked state corresponds to a preset fixed code of zero. When the signal is parsed as code zero, the parking lock is considered to be in the locked state. The system calculates the total pixel area of ​​the closed vehicle outline in the contour image and divides this area by the total pixel area of ​​the effective area of ​​the charging pile parking space to obtain the outline percentage. If the outline percentage is greater than 80%, a complete vehicle outline is considered to exist. If the charging gun is not inserted or inserted but the real-time charging current is less than or equal to 10 mA, the electrical parameters indicate no effective charging current. When all three verification conditions are met, the false occupancy determination unit immediately generates an initial false occupancy event determination result and assigns a system default initial confidence value to the initial determination result. The confidence level is fixed at 0.6, representing the basic judgment confidence. After the initial judgment is completed, the false occupancy judgment unit extracts the standard color label corresponding to the vehicle color feature. It then performs a search operation within the historical operation database, limiting the search to the records of the same charging pile over the past 30 days. This filters out all historical occupancy event records that perfectly match the vehicle color feature. For each matching record, a short-term occupancy behavior judgment is performed. The start and end timestamps of the occupancy are extracted from the record, and the difference between the two timestamps gives the duration of a single occupancy. The preset threshold for short-term occupancy behavior is five minutes. If the duration of a single occupancy is less than or equal to five minutes and there is no valid charging current data in the corresponding record, the record is judged as a short-term occupancy behavior, and the process is complete. After judgment, the total number of occurrences of short-term occupancy behavior is counted. The false occupancy judgment unit performs a confidence increase operation based on the number of occurrences of short-term occupancy behavior. A confidence correction rule is set: the confidence correction coefficient increases by 0.1 for one occurrence of short-term occupancy behavior, and increases by 0.2 for two or more occurrences. The initial confidence value and the correction coefficient value are added together to obtain the increased confidence value. If there is no matching history for the vehicle color feature or no short-term occupancy behavior in the matching history, the initial confidence value is subtracted by 0.1 to obtain the decreased confidence value. The judgment threshold adopts a dynamic reverse adjustment method with a 24-hour cycle. The initial value of the basic judgment threshold is set to 0.75. Within each adjustment cycle, the total number of occurrences of false occupancy behavior is counted. The system counts the number of misjudged records in the charging pile's historical assessment. These misjudged records include successful user appeals and incorrectly determined instances of normal temporary parking. A reverse adjustment calculation rule is established: for each additional misjudged record, the judgment threshold increases by 0.05 from its base value. When the number of misjudged records is zero, the judgment threshold remains unchanged. The judgment threshold is automatically updated and stored at the end of the cycle. The false parking determination unit compares the increased or decreased confidence value with the dynamically adjusted judgment threshold. When the confidence value exceeds the judgment threshold, a false parking event is ultimately determined for the current charging pile. Examples of false parking events include: the charging pile's ground lock status signal is locked with a zero code; the vehicle outline occupancy in the contour image is 85%; and there is no effective charging current in the electrical parameters.The initial confidence level is 0.6, the vehicle color is white, and there are two short-term occupancy behaviors in the historical matching records. The correction coefficient is increased by 0.2, and the confidence level is raised to 0.8. The number of false judgment records in the current period is zero, and the judgment threshold remains at 0.75. Since 0.8 is greater than 0.75, a false occupancy event is finally determined to have occurred, ensuring that the false occupancy judgment result closely matches the actual operation scenario of the charging pile.

[0020] The opportunity service probability factor is generated through a full-process quantitative calculation method involving historical data filtering in the same dimension, basic probability calculation, weighted correction by external environmental factors, and time-series decay smoothing. It reflects the potential probability of a charging pile being used for normal charging in the current time period, providing probability parameters that fit the actual operation scenario for the calculation of opportunity revenue loss value. The same week type attribute refers to the system uniformly dividing the seven days of the week into two categories: weekdays and rest days. Monday to Friday belong to the weekday type, and Saturday and Sunday belong to the rest day type. The current time belongs to which category attribute by filtering all date records that belong to the same category in history. The efficiency loss index generation unit performs a hierarchical data filtering operation from the historical operation database, first extracting the week type attribute corresponding to the current time and the specific time. The time period is divided into independent intervals, each hour. Historical time period records that match the current time's weekday type and fall within the same hour are selected. After filtering, a precise calculation of the basic probability factor is initiated. First, the total number of times the charging station was idle in all selected historical time period records is counted. Idle state refers to a state where the charging station's ground lock is raised, no vehicle occupies the space, and no charging service is in progress. Then, from these idle state records, the total number of actual charging service orders is counted. Charging service orders refer to formal order records where users complete the plug-in operation and generate effective charging current. The ratio of the actual number of charging service orders to the number of times the charging station was idle is the basic probability factor. For example, historical... The total number of idle states in the same period was fifty, with thirty-five valid charging orders generated. The basic probability factor was calculated to be 0.7. The preset mapping rule is a pre-defined weighting system for external factors. This rule divides weather conditions into four levels: sunny, cloudy, rainy, and snowy, corresponding to charging intention influence weights of 1.0, 0.9, 0.7, and 0.5, respectively. It also divides holiday markers into statutory holidays and non-holidays, corresponding to charging intention influence weights of 0.8 and 1.0, respectively. The two weights are combined using a product method to form the comprehensive corrected weight for a single record. The efficiency loss index generation unit obtains weather condition data and holiday marker data corresponding to all historical filtered records from the same period, extracting each record one by one. Record the corresponding weather weights and holiday weights, and calculate the comprehensive correction weight. Calculate the arithmetic mean of the comprehensive correction weights for all records to obtain the overall weighted correction coefficient. Multiply the base probability factor by the overall weighted correction coefficient to complete the weighted correction operation. For example, if the base probability factor is 0.7 and the average comprehensive correction weight is 0.9, the weighted correction value will be 0.63. The weighted correction base probability factor needs to undergo time decay smoothing. This process uses an exponential decay algorithm, taking the historical occurrence time as the baseline and the time difference between the current moment and the historical time as the decay variable. Set the decay half-life to 30 days. The decay coefficient for historical records with a time difference of 30 days or less is 1.0. For every additional 30 days in the time difference, the decay coefficient is multiplied by 0.5.After assigning a corresponding attenuation coefficient to each historical record, a weighted average is calculated between the weighted adjusted value and the attenuation coefficients of all records. This yields a smooth and stable opportunity service probability factor. For example, if the weighted adjusted value is 0.63 and the weighted average of the attenuation coefficients is 0.95, the final opportunity service probability factor is 0.5985, ensuring that the opportunity service probability factor accurately matches the actual charging demand probability for the current period.

[0021] The recent average revenue per service for charging piles is generated through comprehensive calculation of recent orders, basic revenue breakdown, mean calculation, and dynamic correction of trend coefficients. This improves the real-time performance and trend alignment of revenue calculation, accurately reflects the current actual profitability of charging piles, and provides a reliable revenue benchmark for calculating opportunity revenue loss. The preset recent time period refers to a complete time interval of seven consecutive days, starting from the current time point and tracing back to the previous time point. This interval covers a complete weekday and weekend cycle, and can stably reflect the recent true operating revenue level of charging piles. The efficiency loss index generation unit first extracts all charging service orders completed by charging piles within the preset recent time period. For each charging service order, it retrieves the vehicle battery charging start time from the charging pile operation database. The charging service order calculates the difference between the initial charging capacity and the final charging capacity at the end of the charging process. Simultaneously, it retrieves the unit rate for the actual charging period, a pre-set rate based on peak / off-peak hours. Multiplying this difference by the unit rate yields the basic electricity revenue for the order. Then, it extracts the service fee from the order payment records, representing equipment usage and maintenance costs. Adding this to the basic electricity revenue gives the total service revenue for the single charging service order. This process completes the total service revenue for all orders. After calculating service revenue, the total service revenue of all orders is summed to obtain the recent total service revenue. This total service revenue is then divided by the total number of charging service orders completed within a preset recent time period to obtain the initial average revenue. The coefficient for the recent revenue trend is calculated through cross-period revenue comparison and linear mapping. Using the previous seven-day period as a comparison period, the total service revenue of all charging service orders for the previous period is extracted, and the historical average revenue for that period is calculated. The initial average revenue is subtracted from the historical average revenue for the previous period to obtain the revenue difference. This revenue difference is then divided by the historical average revenue for the previous period to obtain the revenue change rate. A benchmark value for the revenue change trend coefficient is set at [value missing]. The baseline value is added to the rate of change in revenue to obtain the final recent revenue trend coefficient. When revenue increases, the trend coefficient is greater than 1.0; when revenue decreases, the trend coefficient is less than 1.0; and when revenue remains stable, the trend coefficient equals 1.0. Finally, the efficiency loss index generation unit multiplies the initial average revenue by the coefficient of the recent revenue trend to obtain the recent average revenue per service for the charging pile. For example, if ten orders are completed within a recent time period, the total service revenue is 200 yuan, the initial average revenue is 20 yuan, the historical average revenue from the previous period is 16 yuan, the revenue difference is 4 yuan, the rate of change in revenue is 0.25, the recent revenue trend coefficient is 1.25, and the final recent average revenue per service for the charging pile is 25 yuan.Ensure that the revenue figures accurately reflect the recent actual operating conditions of the charging stations.

[0022] The charging station occupancy efficiency loss index is generated through a full-process quantitative process, including ideal operating condition modeling, theoretical benefit calculation, loss ratio calculation, and normalization transformation. It can standardize the measurement of the impact of false occupancy on charging station operational efficiency, providing a unified quantitative benchmark for triggering tiered optimization strategies. The preset evaluation period is a fixed one-hour evaluation interval. The predefined state that a charging station is always available and occupied means that within this evaluation period, the charging station hardware has no downtime or maintenance operations, the charging station ground lock remains locked and vehicles are continuously and stably parked, the charging station electrical module remains in a standby and startable state, and effective charging service is immediately initiated after a vehicle parks without waiting. Throughout the entire evaluation period, there were no idle periods, no interruptions, and no equipment unavailability, maintaining an ideal operational state of continuous charging service. The efficiency loss index generation unit first divides the total duration of the preset evaluation period by the historical average service duration of the charging pile. The historical average service duration is the arithmetic mean of the actual service durations of all valid charging orders for the charging pile in the past thirty days, yielding the theoretical maximum number of services under ideal conditions. Then, the theoretical maximum number of services is multiplied by the recent average revenue per service of the charging pile to obtain the theoretical maximum serviceable revenue within the preset evaluation period. Finally, the opportunity revenue loss value caused by false occupancy is divided by the theoretical maximum serviceable revenue to obtain... The transformation function, which normalizes the original occupancy efficiency loss ratio based on historical occupancy efficiency loss data, adopts a min-max linear normalization form. First, it extracts all original occupancy efficiency loss ratio data for the charging pile within the past ninety days, calculates the maximum and minimum values ​​of this data set, and then calculates the occupancy efficiency loss index based on the difference between the original occupancy efficiency loss ratio and the minimum historical data value. This difference is then divided by the difference between the maximum and minimum historical data values, and multiplied by one hundred. This maps the original ratio to a standardized index range of zero to one hundred, ultimately outputting a standardized occupancy efficiency loss index, for example, based on a preset evaluation period. The total duration is 60 minutes, the historical average service duration is 15 minutes, the theoretical maximum number of services is 4, the average revenue per service is 25 yuan, the theoretical maximum service revenue is 100 yuan, the opportunity revenue loss is 20 yuan, the original occupancy efficiency loss ratio is 0.2, the historical original ratio of the past 90 days has a minimum of 0.05 and a maximum of 0.5. After calculation by the conversion function, the occupancy efficiency loss index is equal to (0.2 minus 0.05) divided by (0.5 minus 0.05) multiplied by 100, and the result is 33.33. This ensures that the occupancy efficiency loss index can be compared horizontally across time periods and sites, accurately reflecting the degree of operational efficiency loss caused by false occupancy.

[0023] The optimization strategy execution unit dynamically calibrates the first and second loss thresholds by taking historical operational data quantile values, correcting time-series decay weights, and iterating through a two-stage weighted sorting process. This ensures that the thresholds align with the long-term operational efficiency distribution patterns of charging piles, avoiding false or missed triggers of strategies caused by fixed thresholds and improving the accuracy of hierarchical optimization management. The historical normal operating period is defined as a continuous and effective operating period for the charging pile within the past ninety days, characterized by no hardware failures, no platform maintenance downtime, and no regional power outages. The optimization strategy execution unit first collects all effective occupancy efficiency loss indices generated by the charging piles during this period, then removes abnormal extreme values ​​caused by extreme weather or system failures to form a standard historical sample set. The first specific quantile is the 60th percentile (P60) coordinate position set after the standard historical sample set is arranged in ascending order of numerical values. This quantile corresponds to the critical level of mild efficiency loss under normal operation and is used to define the trigger point for mild intervention. The second specific quantile is the 85th percentile (P85) coordinate position, which is higher than the first specific quantile and corresponds to the critical level of severe efficiency loss under normal operation. It is used to define the trigger point for severe intervention. The time-based decay factor is a time-series weighting coefficient constructed using an exponential decay model. It is calculated based on the time difference between the time when the historical data was generated and the current time, and the decay half-life is set to thirty days. The decay factor for recent historical data within a day is 1.0. For every 30 days exceeding this period, the decay factor is multiplied by 0.5. The older the historical data, the lower its weight, thus mitigating the interference of outdated data on threshold calibration. After sample sorting, the occupancy efficiency loss index corresponding to the 60th percentile in the ascending sequence is selected as the initial value for the first loss threshold, and the occupancy efficiency loss index corresponding to the 85th percentile is selected as the initial value for the second loss threshold. A corresponding time decay factor is matched for each historical sample index. The original sample index is multiplied by the corresponding decay factor to obtain the weighted sample index, constructing a new weighted sorting sequence. The 60th percentile and the 85th percentile are then located again within the weighted sorting sequence. The values ​​corresponding to the percentiles are used to obtain the weighted correction values ​​for the first and second loss thresholds, respectively. Finally, the initial values ​​and the weighted correction values ​​are calculated by performing an arithmetic average with equal weights of 50% each. The result is the final updated values ​​of the first and second loss thresholds. For example, after sorting the standard historical sample set, the initial value of the 60th percentile is 26 and the initial value of the 85th percentile is 43. In the weighted sorting sequence, the correction value of the 60th percentile is 24 and the correction value of the 85th percentile is 41. After equal weight averaging, the final first loss threshold is updated to 25 and the second loss threshold is updated to 42, ensuring that the thresholds are dynamically adapted to the operating status of the charging piles.

[0024] The optimization strategy execution unit employs a layered calculation logic—including efficiency loss over-calculation, fixed basic points calibration, and dynamic adjustment of user credit coefficients—to determine the points reward amount. This differentiated incentive approach precisely guides users to move their vehicles promptly, balancing charging pile operational efficiency losses with platform user experience. Subsequently, it initiates a precise full-process calculation of the points reward amount. First, it subtracts the first loss threshold from the real-time generated occupancy efficiency loss index to obtain the excess amount exceeding the first loss threshold. Then, it divides the excess amount by the first loss threshold to obtain the proportion of the excess amount to the first loss threshold. The basic points reward value is determined by the system. Based on the charging pile operation incentive budget and the platform's overall user incentive strategy, a fixed benchmark value is preset. This value serves as the basic incentive amount under mild efficiency loss intervention and is uniformly set at 100 points by the platform's operations team. It is the fixed benchmark base for point calculation. The basic point reward value is multiplied by the excess ratio value to obtain the calculated point value without credit correction. At the same time, the optimization strategy execution unit retrieves the complete historical credit record of the user account that triggered the false parking space event from the platform's historical operation database using the user's identity identifier corresponding to this false parking space event. The historical credit record consists of the user's historical vehicle relocation response timeliness rate, cumulative... The system comprehensively evaluates three indicators—number of false parking attempts, charging order fulfillment—into four fixed credit levels: Excellent, Good, Average, and Poor. The reward coefficient is a pre-set differential adjustment coefficient based on the user's credit level: Excellent corresponds to a reward coefficient of 1.2, Good to 1.0, Average to 0.8, and Poor to 0.5. The calculated points are multiplied by the reward coefficient matching the user's credit level to obtain the final reward amount. Finally, the optimization strategy execution unit combines this final reward amount with standardized parking relocation procedures. The system generates a vehicle relocation reminder by combining prompt texts, which is then sent via an IoT communication link to the user's terminal device uniquely associated with the current charging station. For example, if the occupancy efficiency loss index is 35, the first loss threshold is 25, the excess is 10, the excess ratio is 0.4, the base reward is 100 points, the calculated reward is 40 points, the user's credit rating is excellent, the reward coefficient is 1.2, and the final reward is 48 points. The entire point calculation is based on quantitative rules and the user's historical credit data, without any subjective human adjustment deviations, achieving a precise match between the reward level and the degree of occupancy efficiency loss and the user's credit level.

[0025] The dynamic rate adjustment mechanism implements linear floating billing based on the degree of occupancy efficiency loss. It addresses severe false occupancy through differentiated economic constraints, quickly recovering charging pile service resources and compensating for the revenue gap caused by operational efficiency losses. The optimization strategy execution unit initiates the dynamic rate adjustment calculation process. First, it subtracts the second loss threshold value from the real-time acquired occupancy efficiency loss index to obtain the second excess amount exceeding the second loss threshold. Then, it divides the second excess amount by the second loss threshold value to obtain the excess ratio of the second excess amount to the second loss threshold. The basic rate is a standard occupancy fee benchmark uniformly set by the charging pile operation platform based on site costs, regional electricity prices, and equipment operation and maintenance costs. The occupancy fee unit price corresponding to the basic rate is the basic occupancy fee value executed per unit time. The optimization strategy execution unit retrieves the occupancy fee unit price corresponding to the current applicable basic rate of the charging pile as the calculation benchmark, multiplies the occupancy fee unit price by the excess ratio, and obtains the surcharge unit price. The surcharge unit price changes linearly with the excess ratio. The optimization strategy execution... From the precise moment when the system determines that the occupancy efficiency loss index exceeds the second loss threshold, the unit sends a rate update command to the billing module built into the charging pile. After receiving the command, the billing module performs an addition operation on the basic rate value and the surcharge unit price value to complete the dynamic update of the real-time occupancy fee unit price. The updated real-time occupancy fee unit price serves as the sole billing standard for the current false occupancy event. The billing module uses a fixed unit time as the billing cycle and performs uninterrupted continuous billing for the current false occupancy event according to the updated real-time occupancy fee unit price until the false occupancy determination unit determines that the false occupancy event has terminated. For example, if the occupancy efficiency loss index is fifty, the second loss threshold is forty, the second excess is ten, the excess ratio is 0.25, the occupancy fee unit price corresponding to the basic rate is two yuan per hour, the surcharge unit price is 0.5 yuan per hour, and the real-time occupancy fee unit price is updated to two and a half yuan per hour, the billing module will continuously bill at two and a half yuan per hour from the determination time. The entire rate calculation strictly follows the linear floating rule to ensure that the billing intensity is completely matched with the degree of occupancy efficiency loss.

[0026] The optimized strategy execution unit establishes a full-time closed-loop monitoring and event archiving mechanism to achieve fully automated management of false parking space occupancy events, from intervention and termination to data feedback. This continuously improves the system's evaluation accuracy and strategy adaptability. After issuing a vehicle relocation reminder or dynamic rate adjustment instruction, the optimized strategy execution unit immediately enters a continuous real-time monitoring state. The monitoring content includes two core types of information: one is all real-time response operations fed back by the user terminal through the charging pile operation platform's interaction interface, including vehicle relocation confirmation, vehicle departure reporting, and charging start triggering; the other is the charging pile's continuously updated on-site data uploaded by the IoT data acquisition unit, including charging pile ground lock status signals, parking space outline images, and vehicle storage information. When the optimization strategy execution unit detects that either of the two types of information—namely, the charging pile status, the charging gun status, and the real-time charging current—meets the termination trigger condition, specifically when the user terminal reports that the vehicle relocation operation has been performed and the parking space has been cleared, or when the electrical parameters uploaded by the IoT data acquisition unit show that the charging gun status is in place and the real-time charging current value is greater than 10 mA, the optimization strategy execution unit immediately sends a trigger command to the false occupancy determination unit to re-determine the current status. After receiving the command, the false occupancy determination unit re-calls the latest collected charging pile ground lock status signal, contour image, vehicle color features, and electrical parameter attribution judgment results, and strictly performs a full-dimensional secondary verification according to the initial false occupancy event judgment rules. If the verification result shows that the parking space is not occupied or the vehicle has entered a valid charging state, the current false parking space occupancy event is finally determined to be over. After receiving the official determination result of the end of the false parking space occupancy event, the optimization strategy execution unit immediately sends a control command to stop billing to the billing module built into the charging pile. After receiving the command, the billing module immediately terminates the entire billing process of the current false parking space occupancy event and fixes the final billing amount. Subsequently, the optimization strategy execution unit performs standardized collection processing on all data of this false parking space occupancy event. The collected data includes the total duration of the false parking space occupancy event from the start to the end of the determination. The total duration is calculated by subtracting the start timestamp from the end timestamp of the event. The entire event is finally generated. The system collects the occupancy efficiency loss index, the final termination type of the event, and the handling result. After collection, the complete data of this false occupancy event is synchronously written into the historical operation database for permanent storage. The stored data will serve as core samples in subsequent iterative calculations of the system. Specifically, it will be used to update the probability factor of the opportunity service, update the recent average single service revenue of the charging pile, and update and calibrate the first and second loss thresholds. This will enable the single event handling data to continuously feed back and optimize the parameters of the entire system's evaluation model and optimization strategy. For example, if a user moves their car within 15 minutes after receiving the points reward reminder, the system will detect the parking space emptying signal and trigger a re-judgment. If the false occupancy event is confirmed to have ended, the billing module will stop billing.This incident lasted fifteen minutes, resulting in a final parking space occupancy efficiency loss index of 32. The incident ended with the user actively moving their vehicle. All data was completely stored in the historical operation database for subsequent model parameter updates. The entire process—monitoring, triggering, judgment, billing, and archiving—was automated, forming a complete closed loop for handling false parking space occupancy incidents. This continuously strengthens the self-learning capabilities and operational control accuracy of the IoT-based charging pile operation efficiency assessment and optimization management system.

[0027] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A charging pile operation efficiency evaluation and optimization management system based on the Internet of Things, characterized in that, include: The IoT data acquisition unit is used to acquire charging pile ground lock status signals, parking space images, and electrical parameters including plug status and real-time charging current. The false parking space determination unit determines whether a false parking space event has occurred based on the charging pile ground lock status signal, parking space image, and electrical parameters including plug status and real-time charging current. The efficiency loss index generation unit is used to obtain the probability of a charging pile successfully charging a parking space in the same week and time period in history after determining that a false parking space occupancy event has occurred. This probability is used as an opportunity service probability factor. The unit also obtains the recent average single service revenue of the charging pile. The unit multiplies the duration of the false parking space occupancy, the opportunity service probability factor, and the average single service revenue to obtain the opportunity revenue loss value. Finally, the unit calculates the proportion of the opportunity revenue loss value in the theoretical maximum service revenue of the charging pile within a preset evaluation period to generate the parking space occupancy efficiency loss index. The optimization strategy execution unit is set with a first loss threshold and a second loss threshold. When the occupancy efficiency loss index exceeds the first loss threshold but does not exceed the second loss threshold, a parking reminder instruction containing points rewards is generated. The amount of the points rewards is positively correlated with the occupancy efficiency loss index. When the occupancy efficiency loss index exceeds the second loss threshold, a dynamic rate adjustment instruction is generated, which adds a floating surcharge to the current occupancy fee unit price on the base rate. The amount of the floating surcharge is determined linearly according to the proportion by which the occupancy efficiency loss index exceeds the second loss threshold, and the billing module bills the current occupancy event in real time until the false occupancy determination unit determines that the false occupancy event has ended.

2. The IoT-based charging pile operation efficiency evaluation and optimization management system according to claim 1, characterized in that: After acquiring the parking space image, the IoT data acquisition unit performs vehicle contour extraction processing on the parking space image to obtain a contour image. At the same time, it extracts the vehicle color features from the parking space image and associates the plug status in the electrical parameters with the real-time charging current to determine whether the vehicle corresponding to the contour image and the electrical parameters belong to the same charging pile service object. The charging pile ground lock status signal, the contour image, the vehicle color features, and the judgment result are sent together to the false occupancy determination unit.

3. The IoT-based charging pile operation efficiency evaluation and optimization management system according to claim 2, characterized in that: The false occupancy determination unit generates an initial false occupancy event determination result based on the charging pile's ground lock status signal being locked, the presence of a complete vehicle outline in the contour image, and the judgment result indicating no effective charging current in the electrical parameters. It then compares the result with the vehicle's color characteristics in a historical database. If the vehicle corresponding to the color characteristics has a short-term occupancy history, the confidence level of the initial false occupancy event determination result is increased; otherwise, the confidence level is decreased. When the confidence level exceeds a dynamically adjusted determination threshold, a false occupancy event is ultimately determined to have occurred. This threshold is obtained by periodically adjusting in reverse based on the number of misjudgment records in the charging pile's historical determinations.

4. The IoT-based charging pile operation efficiency evaluation and optimization management system according to claim 1, characterized in that: The process of obtaining the opportunity service probability factor includes: From the historical operation database, all historical time period records with the same weekday type attribute and within the same time period as the current time are selected. The number of times the charging pile was idle within the historical time period records is counted, and the number of times the charging pile actually generated charging service orders during the idle period is further counted. The ratio of the number of charging service orders generated to the number of times the charging pile was idle is calculated as the basic probability factor. Historical weather condition data and holiday marking data for the same period are obtained. The basic probability factor is weighted and corrected according to a preset mapping rule. The mapping rule defines the influence weight of different weather conditions and whether it is a holiday on users' charging intentions. The weighted and corrected basic probability factor is then subjected to a time decay smoothing process to obtain the opportunity service probability factor.

5. The IoT-based charging pile operation efficiency evaluation and optimization management system according to claim 4, characterized in that: The process of obtaining the recent average revenue per service for charging piles includes: Extract all charging service orders completed by the charging pile within a preset recent time period. For each charging service order, obtain the difference in charging capacity of the vehicle battery from the start to the end of charging, and obtain the applicable rate unit price for the charging period of the charging service order. Multiply the charging capacity difference by the rate unit price to obtain the basic electricity revenue of the charging service order. Then, add the service fee paid for the charging service order to the basic electricity revenue to obtain the total service revenue of the charging service order. Sum the total service revenue of all charging service orders and divide it by the total number of charging service orders to obtain the initial average revenue. Finally, multiply the initial average revenue by a preset coefficient reflecting the recent revenue change trend to obtain the recent average single service revenue of the charging pile. The coefficient of the recent revenue change trend is calculated based on the comparison between the historical average revenue of the previous period and the initial average revenue.

6. The IoT-based charging pile operation efficiency evaluation and optimization management system according to claim 5, characterized in that: Within the preset evaluation period, the predefined charging pile is always in a serviceable and occupied state, which is taken as the ideal state. The theoretical maximum number of services is obtained by dividing the total duration of the preset evaluation period by the historical average service duration of the charging pile. Then, the theoretical maximum number of services is multiplied by the average service revenue per service to obtain the theoretical maximum serviceable revenue. Finally, the opportunity revenue loss value is divided by the theoretical maximum serviceable revenue to obtain the original occupancy efficiency loss ratio. The original occupancy efficiency loss ratio is input into a transformation function that is normalized based on historical occupancy efficiency loss data to output the final occupancy efficiency loss index.

7. The IoT-based charging pile operation efficiency evaluation and optimization management system according to claim 6, characterized in that: The process of setting the first loss threshold and the second loss threshold includes: Collect all occupancy efficiency loss indices generated by the charging pile during its historical normal operating cycle, sort all occupancy efficiency loss indices by numerical value, select the occupancy efficiency loss index located at the first specific quantile position in the sorted sequence as the initial value of the first loss threshold, select the occupancy efficiency loss index located at the second specific quantile position in the sorted sequence as the initial value of the second loss threshold, wherein the second specific quantile position is higher than the first specific quantile position, and introduce a time-based decay factor for each threshold. After weighted sorting and calculation, update the specific values ​​of the first loss threshold and the second loss threshold.

8. The IoT-based charging pile operation efficiency evaluation and optimization management system according to claim 7, characterized in that: When the optimization strategy execution unit generates a parking reminder instruction that includes a points reward, the process for determining the amount of the points reward is as follows: Calculate the amount by which the occupancy efficiency loss index exceeds the first loss threshold, and calculate the proportion of the excess amount to the first loss threshold. Define a base reward value, multiply the base reward value by the proportion to obtain the calculated reward value. Simultaneously, query the historical credit records of the user account that triggered the false occupancy event, set a reward coefficient based on the historical credit records, multiply the calculated reward value by the reward coefficient to obtain the final reward amount, and send a vehicle relocation reminder instruction containing the reward amount to the user terminal associated with the charging pile.

9. The IoT-based charging pile operation efficiency evaluation and optimization management system according to claim 8, characterized in that: When the optimization strategy execution unit generates a dynamic rate adjustment instruction, the process for determining the floating surcharge amount includes: Calculate the second excess amount of the occupancy efficiency loss index exceeding the second loss threshold, and calculate the excess ratio of the second excess amount to the second loss threshold. Obtain the occupancy fee unit price corresponding to the basic rate as a benchmark. Multiply the occupancy fee unit price by the excess ratio to obtain the surcharge unit price. From the moment it is determined that the occupancy efficiency loss index exceeds the second loss threshold, update the real-time occupancy fee unit price in the charging pile's billing module to the sum of the basic rate and the surcharge unit price, and continuously bill the current occupancy event according to the updated real-time occupancy fee unit price.

10. The IoT-based charging pile operation efficiency evaluation and optimization management system according to claim 9, characterized in that: The optimization strategy execution unit continuously monitors the user terminal's response operations and the updated data uploaded by the IoT data acquisition unit. If it detects that the user has performed a vehicle relocation operation or that the electrical parameters show that effective charging has started, it triggers the false parking space determination unit to re-determine the current state. If the re-determination result is that the false parking space event has ended, the optimization strategy execution unit sends a stop billing instruction to the billing module and records the complete data of this false parking space event, including the duration, the final parking space efficiency loss index, and the result, into the historical operation database for subsequent updates and calculations of the opportunity service probability factor, the average single service revenue, and the first and second loss thresholds.

Images