A well lid state intelligent perception and reporting system based on cellular internet of things

Abstract

This invention relates to a smart sensing and reporting system for manhole cover status based on cellular IoT, specifically in the field of IoT technology. The system uses a status sensing and quantification module to acquire real-time manhole cover status event flags, energy availability parameters, and signal reliability parameters. An urgency fusion calculation module integrates these parameters into an urgency score. A decision execution and status control module makes intelligent decisions based on this score, triggering immediate reporting, suspending delays, or silent recording operations to ensure priority reporting of critical alarms such as manhole cover damage, while conserving energy and network resources. A parameter adaptive optimization module learns from historical communication results and dynamically adjusts internal weighting coefficients and decision thresholds for continuous optimization. Ultimately, while ensuring timely reporting of critical events, the system significantly extends the device's runtime and achieves an optimal dynamic balance between energy, network resources, and reporting urgency.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, and more specifically, to a smart sensing and reporting system for manhole cover status based on cellular IoT. Background Technology

[0002] With the deepening of smart city construction, intelligent monitoring of municipal infrastructure has become an important development direction. Among them, monitoring the status of various manhole covers throughout the city is a key link in ensuring public safety and efficient urban operation. Traditional manual inspection methods are inefficient and costly. In recent years, a solution has emerged that monitors the status by installing sensor modules under the manhole covers. However, in actual deployment and operation, the aforementioned intelligent manhole cover monitoring system based on image recognition faces a severe technical challenge: how to design an efficient and reliable data reporting strategy under the multiple constraints of massive terminals, limited energy supply, and dynamic wireless environment. Especially for manhole cover status monitoring, the urgency of events varies greatly. Existing solutions lack multi-dimensional collaborative perception and intelligent evaluation of the device's remaining energy, current wireless channel quality, and the priority of manhole cover events to be reported. They cannot make adaptive decisions, making it difficult to balance network load and optimize terminal energy use while ensuring timely reporting of critical alarms such as manhole cover damage. Ultimately, this threatens the long-term sustainable operation of the entire monitoring network. Summary of the Invention

[0003] This invention addresses the technical problems existing in the prior art by providing a smart sensing and reporting system for manhole cover status based on cellular Internet of Things, thereby resolving the issues mentioned in the background section.

[0004] The technical solution of this invention to solve the above-mentioned technical problems is as follows: Specifically, it includes: a state perception and quantification module, an urgency fusion calculation module, a decision execution and state control module, and a parameter adaptive optimization module connected in sequence, wherein; The status awareness and quantization module is activated by the microcontroller unit during a preset awareness period or when triggered by a touch event. It samples the voltage signal of the device's power battery through its analog-to-digital conversion channel, processed by a voltage divider circuit. It also queries the wireless communication unit for the current received signal strength via a serial communication interface and reads the latest determined event flag representing the manhole cover's status from non-volatile memory. The event flag is generated by the microcontroller unit after recognizing images of the manhole cover captured by the image acquisition module, and includes at least three states: normal, slightly abnormal, and severely damaged. Based on a preset mapping relationship, it converts the battery voltage signal into an energy availability parameter representing the device's remaining energy availability, converts the received signal strength into a signal reliability parameter representing the current wireless channel connection reliability, and maps the event flag to an event priority parameter representing the urgency of the manhole cover's status. Urgency fusion calculation module: After obtaining the energy availability parameter, signal reliability parameter and event priority parameter, it performs urgency score fusion calculation. It uses preset calculation rules to perform nonlinear transformation on the energy availability parameter and signal reliability parameter respectively and weighted combination, and then multiplies it with the event priority parameter to generate an urgency score that comprehensively reflects whether the current manhole cover status should be reported immediately. Decision execution and state control module: Compares the urgency score with multiple preset thresholds stored in non-volatile memory. Based on the comparison result, it drives an internal state machine to perform state transitions and triggers corresponding operations. The operations include: the wireless communication unit immediately initiates a network connection and reports an event flag containing the manhole cover status and related data, or suspends the reporting task and enters low-power sleep mode, or only records the current time and status information in non-volatile memory without starting the wireless communication unit; after each communication attempt is completed, a communication result is generated. Parameter adaptive optimization module: After each communication attempt is completed, the communication result is saved as a historical record, and the historical record is statistically analyzed periodically. Based on the analysis results, the weighting coefficient in the urgency score fusion calculation process is dynamically adjusted or the values ​​of multiple preset thresholds are adjusted. In a preferred embodiment, the specific operation of converting the battery voltage signal into an energy availability parameter in the state sensing and quantification module is as follows: The microcontroller unit performs digital filtering on the battery voltage signal obtained by the analog-to-digital conversion channel after it has been processed by the voltage divider circuit; The filtered battery voltage signal is used as input to look up the battery discharge characteristic curve in a pre-stored non-volatile memory to obtain the basic energy estimate; An aging compensation factor based on the cumulative number of battery working cycles is introduced to simulate the degradation of battery performance with the number of uses. The calculation process is as follows: First, calculate the sum of the rated number of working cycles and the cumulative number of working cycles to obtain the calculation result. Second, divide the rated number of working cycles by the calculation result to obtain the aging compensation factor. A preset attenuation coefficient between 0 and 1 is introduced to adjust the intensity of aging compensation; The energy availability parameter is obtained by multiplying the basic energy estimate by the compensation term. The calculation process of the compensation term is as follows: First, calculate the difference between the value and the aging compensation factor. Second, calculate the product of the preset attenuation coefficient and the difference. Finally, subtract the product from the value to get the compensation term.

[0005] In a preferred embodiment, the specific operation of converting the received signal strength into a signal reliability parameter is as follows: The microcontroller unit obtains the received signal strength in decibels and milliwatts from the wireless communication unit; The conversion process uses a piecewise nonlinear function, which predefines three thresholds: a minimum communication threshold, a reliable quality threshold, and an extremely high signal threshold. When the received signal strength is below the minimum communication threshold, the signal reliability parameter is set to zero; When the received signal strength is between the reliable quality threshold and the very high signal threshold, the signal reliability parameter is equal to the value minus the result of the exponent operation with the natural constant as the base, the product of the negative exponent of the exponent operation and the square of the difference between the received signal strength and the reliable quality threshold. When the received signal strength is higher than the extremely high signal threshold, the signal reliability parameter is set to one.

[0006] In a preferred embodiment, the specific operation of mapping the event flag to the event priority parameter is as follows: The microcontroller unit reads the latest determined event flag written by the image recognition algorithm or touch trigger from a specified address in the non-volatile memory. The event flag is an enumerated type flag that represents the state of the manhole cover. Based on the preset mapping relationship, event flags with different enumeration values ​​are directly mapped to different fixed values ​​as event priority parameters; among them, event flags indicating normal state synchronization are mapped to the first value, event flags indicating minor abnormalities in the manhole cover are mapped to the second value greater than the first value, and event flags indicating severe damage to the manhole cover are mapped to the third value greater than the second value.

[0007] In a preferred embodiment, the specific operation of performing nonlinear transformations on the energy availability parameter and the signal reliability parameter in the urgency fusion calculation module is as follows: A first nonlinear transformation is applied to the energy availability parameter. The first nonlinear transformation is a logarithmic function with the natural constant as the base. The argument of the logarithmic function is the sum of the product of the value and a preset adjustment factor multiplied by the energy availability parameter. The value of the logarithmic function is divided by the natural logarithm of the sum of the value and the adjustment factor. The result is used as the output of the first nonlinear transformation. A second nonlinear transformation is applied to the signal reliability parameter. This second nonlinear transformation is defined by a piecewise function, which has a preset signal reliability decision threshold and a preset steepness coefficient. When the signal reliability parameter is lower than the signal reliability decision threshold, the output of the second nonlinear transformation is set to zero. When the signal reliability parameter reaches or exceeds the signal reliability decision threshold, the output of the second nonlinear transformation is obtained through the following calculations: First, the first difference is obtained by subtracting the signal reliability decision threshold from the signal reliability parameter. Second, the second difference is obtained by subtracting the signal reliability decision threshold from the first difference. Then, the first difference is divided by the second difference to obtain a ratio. Next, the product of the negative steepness coefficient and the ratio is calculated, and the product is used as an exponent for exponential operation with the natural constant as the base. Finally, the result of the exponential operation is subtracted from the first difference, and the obtained value is the output of the second nonlinear transformation.

[0008] In a preferred embodiment, the specific process of generating a comprehensive urgency score is as follows: The output of the first nonlinear transformation is multiplied by the first balance coefficient to obtain the first weighted result; the output of the second nonlinear transformation is multiplied by the second balance coefficient to obtain the second weighted result; the sum of the first balance coefficient and the second balance coefficient is one. The first weighted result is added to the second weighted result to obtain a comprehensive environmental fitness factor; The environmental fitness factor is multiplied by the event priority parameter to obtain the final urgency score.

[0009] In a preferred embodiment, the decision execution and state control module maintains an internal state machine, the state set of the internal state machine including a dormant state, a ready state, a reporting attempt state, and a suspended state; the multiple preset thresholds stored in the non-volatile memory include at least an immediate reporting threshold, a regular reporting threshold, and a dormant threshold, wherein the value of the immediate reporting threshold is greater than the regular reporting threshold, and the value of the regular reporting threshold is greater than the dormant threshold. The specific process of driving the internal state machine to perform state transitions based on the comparison results is as follows: First, a dynamic decision factor is calculated, which is equal to the difference between a numerical value and a preset historical weight factor, plus the product of the historical weight factor and a recent communication success rate moving average. Then, the immediate reporting threshold is multiplied by the dynamic decision factor to obtain the immediate reporting effective threshold, and the regular reporting threshold is multiplied by the dynamic decision factor to obtain the regular reporting effective threshold, while the dormant threshold remains unchanged. Finally, the urgency score is compared with the dormancy threshold, the regular reporting activation threshold, and the immediate reporting activation threshold. Based on the comparison results and the current state of the internal state machine, the state transition target is determined. The decision execution and state control module generates a communication result each time it exits the reporting attempt state.

[0010] In a preferred embodiment, triggering the corresponding operation based on the comparison result specifically includes: When the urgency score is greater than or equal to the immediate reporting threshold, the control wireless communication unit immediately initiates a network connection and reports data, while the internal state machine transitions to the reporting test state. When the urgency score is less than the immediate reporting threshold but greater than or equal to the normal reporting threshold, the internal state machine transitions to a suspended state, starts a delay timer, and controls the device to enter low-power sleep mode. The delay duration of the delay timer is equal to a preset base delay unit duration multiplied by the square of a ratio. The ratio is calculated as follows: first, the larger value between the urgency score and a preset minimum normal number is taken to obtain a comparison base; then, the immediate reporting threshold is divided by the comparison base to obtain the ratio. When the urgency score is less than the regular reporting threshold and greater than or equal to the dormancy threshold, the internal state machine transitions to or remains in the ready state. When the urgency score is less than the sleep threshold and the internal state machine is not currently in a suspended state, the current time, urgency score, energy availability parameter, signal reliability parameter, and event flag are combined into a single state information and appended to the log storage area of ​​the non-volatile memory. Then, the control device enters a sleep state.

[0011] In a preferred embodiment, the specific process of statistical analysis of historical records in the parameter adaptive optimization module is as follows: A health weight is calculated for each historical record. The calculation process is as follows: First, the difference between the current time and the time the historical record was generated is calculated; second, the result of an exponential operation with a base of a natural constant is calculated, and the product of a preset time decay coefficient with a negative exponent is obtained by multiplying the time difference to obtain a time decay factor; then, the connection success or failure flag in the historical record is added to a preset positive constant to obtain a success factor; finally, the time decay factor is multiplied by the success factor to obtain the health weight of the historical record. The energy availability parameter is divided into several consecutive first intervals, and the signal reliability parameter is divided into several consecutive second intervals. Each first interval and each second interval are combined to form multiple two-dimensional state clusters. For each state cluster, the weighted average communication success rate and weighted average energy efficiency of that state cluster are calculated using the health weights of the historical records belonging to that state cluster.

[0012] In a preferred embodiment, the process of dynamically adjusting the weighting coefficients or adjusting the values ​​of multiple preset thresholds in the urgency score fusion calculation based on the analysis results specifically includes: Construct a system-level performance objective function, which is the weighted sum of the natural logarithms of the weighted average communication success rate and the weighted average energy efficiency of each state cluster; By analyzing historical records, the influence trend on the system-level performance objective function is evaluated when the first balance coefficient, the second balance coefficient, the immediate reporting threshold, the regular reporting threshold, or the dormant threshold take different values, and the gradient direction of the system-level performance objective function with respect to these parameters is estimated. Based on the gradient direction, the parameters are updated using the gradient descent method with momentum, and the updated parameter values ​​are written to non-volatile memory.

[0013] The beneficial effects of this invention are as follows: The state perception and quantification module acquires manhole cover status event flags, energy availability parameters, and signal reliability parameters in real time. The urgency fusion calculation module fuses these parameters into an urgency score. The decision execution and state control module makes intelligent decisions based on this score, triggering immediate reporting, suspension with delay, or silent recording operations to ensure priority reporting of critical alarms such as manhole cover damage, while saving energy and network resources. The parameter adaptive optimization module learns from historical communication results and dynamically adjusts internal weighting coefficients and decision thresholds to achieve continuous optimization. Ultimately, while ensuring timely reporting of critical events, the device's operating time is significantly extended, and an optimal dynamic balance between energy, network resources, and reporting urgency is achieved. Attached Figure Description

[0014] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a block diagram of the system structure of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0016] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0017] In the description of this application, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0018] Example 1 This embodiment provides, for example Figure 1-2 The system shown is a smart sensing and reporting system for manhole cover status based on cellular IoT, specifically including: a status sensing and quantification module, an urgency fusion calculation module, a decision execution and status control module, and a parameter adaptive optimization module connected in sequence; wherein; The status awareness and quantization module is activated by the microcontroller unit during a preset awareness period or when triggered by a touch event. It samples the voltage signal of the device's power battery through its analog-to-digital conversion channel, processed by a voltage divider circuit. It also queries the wireless communication unit for the current received signal strength via a serial communication interface and reads the latest determined event flag representing the manhole cover's status from non-volatile memory. The event flag is generated by the microcontroller unit after recognizing images of the manhole cover captured by the image acquisition module, and includes at least three states: normal, slightly abnormal, and severely damaged. Based on a preset mapping relationship, it converts the battery voltage signal into an energy availability parameter representing the device's remaining energy availability, converts the received signal strength into a signal reliability parameter representing the current wireless channel connection reliability, and maps the event flag to an event priority parameter representing the urgency of the manhole cover's status. Urgency fusion calculation module: After obtaining the energy availability parameter, signal reliability parameter and event priority parameter, it performs urgency score fusion calculation. It uses preset calculation rules to perform nonlinear transformation on the energy availability parameter and signal reliability parameter respectively and weighted combination, and then multiplies it with the event priority parameter to generate an urgency score that comprehensively reflects whether the current manhole cover status should be reported immediately. Decision execution and state control module: Compares the urgency score with multiple preset thresholds stored in non-volatile memory. Based on the comparison result, it drives an internal state machine to perform state transitions and triggers corresponding operations. The operations include: the wireless communication unit immediately initiates a network connection and reports an event flag containing the manhole cover status and related data, or suspends the reporting task and enters low-power sleep mode, or only records the current time and status information in non-volatile memory without starting the wireless communication unit; after each communication attempt is completed, a communication result is generated. Parameter adaptive optimization module: After each communication attempt is completed, the communication result is saved as a historical record, and the historical record is statistically analyzed periodically. Based on the analysis results, the weighting coefficient in the urgency score fusion calculation process is dynamically adjusted or the values ​​of multiple preset thresholds are adjusted.

[0019] In summary, this system is deployed within a monitoring device installed beneath the manhole cover. The microcontroller unit can be a low-power HC32L176KATA microcontroller; the image acquisition module can use a GC0310 camera chip with a wide-angle lens to periodically capture images of the inner wall of the manhole cover; the non-volatile memory can be an FM24C64A ferroelectric memory to store images, event flags, and system parameters; the wireless communication unit can be an ML307R cellular communication module supporting LTE Cat1; the battery can be an ER26500 lithium thionyl chloride battery, supplemented by an HPC1520 composite pulse capacitor to support the instantaneous high current during communication; touch event triggering can be implemented using a TTP233 capacitive touch chip; this system runs on this hardware platform to solve specific problems in manhole cover monitoring scenarios.

[0020] In this embodiment, it is specifically necessary to explain the specific operation of converting the battery voltage signal into an energy availability parameter in the state perception and quantification module as follows: The microcontroller unit performs digital filtering on the battery voltage signal obtained by the analog-to-digital conversion channel after it has been processed by the voltage divider circuit. The voltage divider circuit consists of a first voltage divider resistor, a second voltage divider resistor, and a fuse. The filtered battery voltage signal is used as input to look up the battery discharge characteristic curve in a pre-stored non-volatile memory to obtain the basic energy estimate; An aging compensation factor based on the cumulative working cycle count of the battery is introduced. The aging compensation factor is equal to the rated working cycle count divided by the sum of the rated working cycle count and the cumulative working cycle count. The basic energy estimate is multiplied by a compensation term consisting of the product of the value minus the preset attenuation coefficient and the value minus the aging compensation factor to obtain the energy availability parameter. The battery discharge characteristic curve lookup table was obtained through experimental calibration, establishing a correspondence between battery terminal voltage and remaining capacity percentage. An aging compensation factor, estimated based on the cumulative number of battery operating cycles, was introduced to simulate the degradation of battery performance with usage cycles. The rated number of operating cycles is the nominal lifespan parameter of the battery model, for example, it can be set to 10,000 cycles. The cumulative number of operating cycles is counted and stored by the microcontroller unit; each complete sensing, decision-making, and communication (or recording) cycle is counted as one cycle. The attenuation coefficient is a decimal between 0.05 and 0.2, for example, 0.1, preset by the system designer, used to control the intensity of aging compensation. The compensation term, "the product of the value minus the preset attenuation coefficient and the value minus the aging compensation factor," has the following effect: when the battery is new (aging compensation factor close to 1), the compensation term is approximately equal to 1, and the energy availability parameter is mainly determined by the basic energy estimate; as the battery ages (aging compensation factor decreases), the compensation term becomes less than 1, thus moderately reducing the calculated energy availability parameter value, proactively reflecting the capacity degradation trend, and guiding the system to adopt a more conservative communication strategy in the later stages of battery life. The specific steps to convert the received signal strength into a signal reliability parameter are as follows: The microcontroller unit obtains the received signal strength in decibels and milliwatts from the wireless communication unit; The conversion process uses a piecewise nonlinear function, which is defined with a minimum communication threshold, a reliability quality threshold, and an extremely high signal threshold. When the received signal strength is below the minimum communication threshold, the signal reliability parameter is set to zero; When the received signal strength is between the reliable quality threshold and the very high signal threshold, the signal reliability parameter is equal to the value minus the result of the exponent operation with the natural constant as the base. The exponent of the exponent operation is the negative form factor, and the product of the square of the difference between the received signal strength and the reliable quality threshold, where the form factor is a real number greater than zero. When the received signal strength is higher than the extremely high signal threshold, the signal reliability parameter is set to one. Specifically, the minimum communication threshold, reliability quality threshold, and extremely high signal threshold can be set according to the technical specifications of the wireless communication unit and the actual deployment environment. For example, the minimum communication threshold can be set to -110dBm, below which a connection is considered virtually impossible to establish; the reliability quality threshold can be set to -95dBm, representing the critical point for stable and reliable signal quality; the extremely high signal threshold can be set to -75dBm, beyond which the improvement in signal quality contributes very little to the improvement in connection reliability; the shape factor is a real number greater than zero, such as 0.05, used to control the steepness of the transition curve near the reliability quality threshold; this design makes the signal reliability parameter change sensitively in the critical signal range (such as near -95dBm), effectively distinguishing between the critical channel states of "usable" and "unusable"; The specific steps to map event flags to event priority parameters are as follows: The microcontroller unit reads the latest determined event flag written by the image recognition algorithm or touch trigger from a specified address in the non-volatile memory. The event flag is an enumerated type flag that represents the state of the manhole cover. Based on the preset mapping relationship, event flags with different enumeration values ​​are directly mapped to different fixed values ​​as event priority parameters; among them, event flags indicating normal state synchronization are mapped to the first value, event flags indicating minor abnormalities in the manhole cover are mapped to the second value greater than the first value, and event flags indicating severe damage to the manhole cover are mapped to the third value greater than the second value. "Minor anomaly" can refer to a state where the manhole cover has cracks or rust but has not yet completely failed, while "severe damage" can refer to a state where the manhole cover is broken or missing, which may immediately cause a safety accident; the first, second, and third values ​​can be set to 1, 3, and 10, respectively; this mapping relationship reflects the non-linear leap of the business urgency level, ensuring that high safety risk events (severe damage, priority 10) receive a much higher weight in decision-making than routine synchronization (priority 1) and general early warning (priority 3), thereby triggering immediate reporting in most cases.

[0021] In this embodiment, it is specifically necessary to explain the specific operations of the nonlinear transformation of the energy availability parameter and the signal reliability parameter in the urgency fusion calculation module as follows: A first nonlinear transformation is applied to the energy availability parameter. The first nonlinear transformation is a logarithmic function with the natural constant as the base. The argument of the logarithmic function is the sum of the product of the value and a preset adjustment factor greater than zero multiplied by the energy availability parameter. The value of the logarithmic function is divided by the natural logarithm of the sum of the value and the adjustment factor. The result is used as the output of the first nonlinear transformation. A second nonlinear transformation is applied to the signal reliability parameter. This second nonlinear transformation is defined by a piecewise function, which has a preset signal reliability decision threshold ranging from zero to one and a preset steepness coefficient greater than zero. When the signal reliability parameter is below the signal reliability decision threshold, the output of the second nonlinear transformation is set to zero. When the signal reliability parameter reaches or exceeds the signal reliability decision threshold, the output of the second nonlinear transformation is calculated as follows: First, calculate the first difference obtained by subtracting the signal reliability decision threshold from the signal reliability parameter; second, calculate the second difference obtained by subtracting the signal reliability decision threshold from the first difference; next, divide the first difference by the second difference to obtain a ratio; then, calculate the product of the negative steepness coefficient and the ratio, and use the product as the exponent to perform exponential operation with the natural constant as the base; finally, subtract the result of the exponential operation from the first difference, and the obtained value is the output of the second nonlinear transformation. The adjustment factor is a preset real number greater than zero, such as 9, used to control the shape of the first nonlinear transformation curve. The larger the adjustment factor, the higher the output growth rate when the energy availability parameter is low. This simulates that when energy is scarce, each unit of extra energy is highly valuable in increasing the "reporting willingness". This transformation introduces the diminishing marginal utility effect when energy is abundant. The signal reliability decision threshold is a preset value between 0 and 1, such as 0.3, representing the lowest signal reliability level that the system considers worthwhile to attempt communication. The steepness coefficient is a preset real number greater than zero, such as 5, which controls the rate of output growth after the signal reliability exceeds the threshold. The second nonlinear transformation forces the communication decision to be highly sensitive to the critical channel state. When the signal reliability parameter just exceeds the threshold, the output value grows most rapidly, prompting the device to keenly seize the brief window of good communication. The specific process for generating a comprehensive urgency score is as follows: The output of the first nonlinear transformation is multiplied by the first balance coefficient to obtain the first weighted result; the output of the second nonlinear transformation is multiplied by the second balance coefficient to obtain the second weighted result; both the first balance coefficient and the second balance coefficient are configurable positive real numbers, and the sum of the first balance coefficient and the second balance coefficient is equal to the value one; The first weighted result is added to the second weighted result to obtain a comprehensive environmental fitness factor; finally, the environmental fitness factor is multiplied by the event priority parameter to obtain the final urgency score. Both the first and second balance coefficients are configurable positive real numbers, and their sum is one. In the initial configuration, both can be set to 0.5, indicating that energy state and channel conditions have equal initial weight in the decision-making process. The environmental fitness factor is between 0 and 1, quantifying the "expected cost-benefit ratio" of performing communication under the current energy and signal conditions. The urgency score is obtained by multiplying the environmental fitness factor by the event priority parameter, which means that the urgency of the event itself (e.g., severe damage is 10) has a global amplifying effect on the entire decision-making process, ensuring that the priority of business logic is always the final decision-making factor, while the fusion evaluation of energy and signal provides a feasibility judgment based on the current physical constraints.

[0022] In this embodiment, it is specifically noted that the decision execution and state control module maintains an internal state machine. The state set of the internal state machine includes a dormant state, a ready state, a reporting attempt state, and a suspended state. The multiple preset thresholds stored in the non-volatile memory include at least an immediate reporting threshold, a regular reporting threshold, and a dormant threshold, wherein the value of the immediate reporting threshold is greater than the value of the regular reporting threshold, and the value of the regular reporting threshold is greater than the value of the dormant threshold. The specific process of driving an internal state machine to perform state transitions based on the comparison results is as follows: First, a dynamic decision factor is calculated. This dynamic decision factor equals a numerical value minus a preset historical weighting factor (ranging from 0 to 1), resulting in a third difference. This difference is then multiplied by the historical weighting factor and a recent communication success rate moving average. The recent communication success rate moving average is calculated and updated by the parameter adaptive optimization module based on the results of the most recent specified number of communication attempts. The historical weighting factor is a preset value between 0 and 1, for example, 0.3, used to adjust the influence of historical communication performance on the current decision. The dynamic decision factor incorporates historical performance feedback into real-time decision-making: when the recent communication success rate is high, the dynamic decision factor is greater than 1, leading to a higher effective threshold, stricter decision criteria, and suppression of unnecessary communication to conserve energy; when the recent communication success rate is low, the dynamic decision factor is less than 1, the effective threshold is lowered, decision criteria are relaxed, and attempts to communicate are encouraged to explore potentially improved network environments. Then, the immediate reporting threshold is multiplied by the dynamic decision factor to obtain the immediate reporting effective threshold, and the regular reporting threshold is multiplied by the dynamic decision factor to obtain the regular reporting effective threshold, while the dormant threshold remains unchanged. The immediate reporting threshold, regular reporting threshold, and dormant threshold are pre-configured constants, and the immediate reporting threshold, regular reporting threshold, and dormant threshold can be calibrated according to the application scenario's requirements for response speed and energy saving, for example, they can be set to 25.0, 8.0, and 2.0 respectively. Finally, the urgency score is compared with the hibernation threshold, the regular reporting activation threshold, and the immediate reporting activation threshold. Based on the comparison results and the current state of the internal state machine, the next state to which the state machine will transition is determined. After each communication attempt, the decision execution and status control module generates a communication result. This result includes the urgency score that triggered the communication attempt, energy availability parameters, signal reliability parameters, actual communication time, connection success or failure flags, and an energy consumption indicator estimated based on the operating status of the wireless communication unit. The generated communication result is provided to the parameter adaptive optimization module as the basic input data for its statistical analysis. The energy consumption indicator can be roughly estimated based on the product of the total power-on time of the wireless communication unit during the communication attempt and the typical operating current. The structured design of the communication result data packet ensures that the parameter adaptive optimization module can obtain the "context-action-result" closed-loop information of a complete communication attempt, which is the data foundation for effective statistical analysis and parameter optimization. The corresponding operation is triggered based on the comparison result, specifically: When the urgency score is greater than or equal to the immediate reporting threshold, the control wireless communication unit immediately initiates a network connection and reports data, while the internal state machine transitions to the reporting test state. When the urgency score is less than the immediate reporting threshold but greater than or equal to the regular reporting threshold, the internal state machine transitions to a suspended state, starts a delay timer, and controls the device to enter low-power sleep mode. The delay timer's delay duration is calculated as follows: a preset base delay unit is used as the baseline duration, multiplied by the square of a ratio. The ratio is calculated as follows: first, the larger value between the urgency score and a preset, greater than zero, minimal zero-prevention constant coefficient is taken to obtain a comparison base; then, the immediate reporting threshold is divided by the comparison base to obtain the ratio. The base delay unit can be set according to application requirements, such as 5 minutes. The minimum zero constant coefficient is used to prevent division by zero errors or calculation anomalies when the urgency score is zero or extremely low. It can be set to 0.1. The delay duration calculation formula (base delay unit * (immediate reporting effective threshold / max(urgency score, minimum zero constant coefficient))^2) creates a non-linear elastic backoff mechanism. Its effect is: the closer the urgency score is to the immediate reporting effective threshold, the shorter the calculated delay duration and the faster the response; the closer the urgency score is to the regular reporting effective threshold, the delay duration will increase sharply in a square relationship, and the waiting time will become very long. This design realizes fine-grained scheduling of medium-priority tasks, which not only avoids frequent state switching caused by small fluctuations in the score near the critical value (anti-jitter), but also allows suspended tasks with relatively higher urgency to get a re-evaluation opportunity earlier, while allowing less urgent tasks to wait quietly for longer, thereby smoothing the potential request traffic of the entire network. When the urgency score is less than the regular reporting threshold and greater than or equal to the dormancy threshold, the internal state machine transitions to or remains in the ready state. When the urgency score is less than the sleep threshold and the internal state machine is not currently in a suspended state, the current time information, urgency score, energy availability parameter, signal reliability parameter and event flag are combined into a state information and appended to a preset log storage area in the non-volatile memory, which is managed in a cyclic overwrite manner. After the writing is completed, the control device enters a sleep state and the wireless communication unit is not started during this process. The log storage area is managed using a circular overwrite method, meaning that when the storage space is full, new records will overwrite the oldest records, thus preserving the state history of the most recent period within the limited non-volatile memory space. This "silent recording" operation is the system's ultimate energy-saving strategy under extremely low urgency. Its effect is that when it is determined that the current state is not worth the high power consumption of wireless communication, the system saves a snapshot of the current state at the cost of negligible storage write power consumption. This preserves historical information for possible subsequent analysis or fault diagnosis, and ensures that the device can enter the deepest sleep state, maximizing the standby time. Only when it is triggered by an event or periodically woken up at some later time, and the calculated new urgency score is high enough, can the state information of these silent records be packaged and reported together with the latest state.

[0023] In this embodiment, it is specifically necessary to explain the process in the parameter adaptive optimization module of saving the communication result after each communication attempt as a historical record and periodically performing statistical analysis on the historical record as follows: Historical records are stored in a dedicated historical record storage area in non-volatile memory, managed in a circular queue manner. The historical record storage area stores the most recent specified number of historical records. A health weight is calculated for each historical record. The calculation process is as follows: First, calculate the difference between the current time and the time the historical record was generated. Second, calculate the time decay factor by multiplying the result of an exponential operation with a negative exponent by a preset time decay coefficient. Then, add the connection success or failure flag from the historical record to a preset positive constant to obtain the success factor. The positive constant ensures that the success factor is not zero even if communication fails. Finally, multiply the time decay factor by the success factor to obtain the health weight of that historical record. The time decay coefficient is a preset positive real number, such as 0.01 per day, which gives higher weight to recent records. The positive constant is a preset small positive number, such as 0. 1. To ensure that failure records also have non-zero weights, the most recent specified number can be set to 50. The effect of health weight calculation is to achieve differentiated evaluation of the value of historical records. The time decay factor gives higher weight to recent communication attempt records, which prompts the optimization process to pay more attention to the latest performance of the device in the current environment and enhances the system's ability to track slow environmental changes. The success factor ensures that records of successful communication dominate the evaluation and guides the optimization direction toward strategies that are more likely to succeed. Giving failure records a non-zero but smaller weight retains the possibility of learning from failures and prevents the optimization process from going astray due to accidental failures. This weighting mechanism simulates the human learning mode of "valuing recent successful experiences but not forgetting the lessons of failures". The entire possible range of energy availability parameters is divided into several continuous and ordered first intervals, and the entire possible range of signal reliability parameters is divided into several continuous and ordered second intervals. Each first interval and each second interval are combined to form multiple two-dimensional rectangular regions, and each rectangular region is defined as a state cluster. Thus, the continuous state space composed of energy availability parameters and signal reliability parameters is divided into multiple discrete and non-overlapping state clusters. For example, the range of energy availability parameter from zero to one is divided into three first intervals, and the range of signal reliability parameter from zero to one is divided into three second intervals. After combination, nine state clusters are obtained. The effect of discretizing the continuous state space into state clusters is significant. It transforms the problem that originally required complex evaluation in an infinite continuous space into performance statistical analysis of a finite number of discrete regions. This greatly reduces the computational complexity and storage space requirements, making it possible to achieve real-time parameter adaptive optimization on resource-constrained microcontroller units. Each state cluster represents a specific combination of energy and signal conditions. Performance evaluation of each cluster can clearly reveal the strengths and weaknesses of the current decision-making strategy under different operating conditions. For each state cluster, the weighted average communication success rate and weighted average energy efficiency of the state cluster are calculated using the health weights of the historical records belonging to that state cluster. The weighted average communication success rate is equal to the sum of the products of the health weights of all historical records in the state cluster and the corresponding connection success or failure flags, and then divided by the sum of the health weights of all historical records in the state cluster. The weighted average energy efficiency equals the sum of the health weights corresponding to the successfully connected records in all historical records within the state cluster, divided by the sum of the products of the health weights of all historical records in the state cluster and the corresponding energy consumption estimates. The weighted average communication success rate reflects the probability of the current strategy achieving reliable communication under the working conditions represented by a specific state cluster, while the weighted average energy efficiency quantifies the communication success reward that can be obtained per unit of energy consumption under the working conditions. Using health weights for calculation ensures that recent and successful records contribute more to these two performance indicators, making the performance evaluation more reflective of the effectiveness of the strategy in the current period. The process of dynamically adjusting the weighting coefficients or adjusting the values ​​of multiple preset thresholds in the urgency score fusion calculation based on the analysis results is as follows: The weighting coefficients include the first and second balance coefficients in the urgency score fusion calculation process; A system-level performance objective function is constructed. The calculation process of this system-level performance objective function is as follows: the total cluster weight of each state cluster is multiplied by the weighted average communication success rate of that state cluster by a preset third balancing coefficient, and the natural logarithm of the weighted average energy efficiency of that state cluster is multiplied by a preset fourth balancing coefficient. The product of the sum of the two is then accumulated for all state clusters to obtain the system-level performance objective function. Among them, the third balancing coefficient and the fourth balancing coefficient are preset real numbers that are greater than zero, used to adjust the relative importance of communication success rate and energy efficiency in the objective function. The third and fourth balance coefficients are used to adjust the relative importance of communication success rate and energy efficiency in the optimization objective. For example, they can be set to 1.0 and 0.5 respectively, indicating that the optimization focuses more on improving the communication success rate while taking energy efficiency into account. Taking the natural logarithm of the weighted average energy efficiency has the effect of transforming the absolute difference in energy efficiency into a perception of the relative difference. This is in line with the law of "diminishing marginal returns" in resource optimization, that is, when energy efficiency is already high, its further improvement will contribute less to the overall utility of the system. The design effect of this objective function is to provide a single, quantifiable optimization objective for the performance of the entire system. It balances the sometimes conflicting demands of "communication reliability" and "energy economy", so that subsequent parameter adjustments have a clear direction. By analyzing historical records, the impact trends on the weighted average communication success rate and weighted average energy efficiency of each state cluster are evaluated when the first balance coefficient, second balance coefficient, immediate reporting threshold, regular reporting threshold, or sleep threshold take different values. Furthermore, the gradient direction of the system-level performance objective function relative to the first balance coefficient, second balance coefficient, immediate reporting threshold, regular reporting threshold, or sleep threshold is estimated. Specifically, for the first balance coefficient, the gradient direction is estimated as follows: all historical records are divided according to the value of the energy availability parameter at the time the current communication attempt is triggered, using the median value of the energy availability parameter across all records as the boundary. The system is divided into high-energy and low-energy groups. The average performance of the high-energy and low-energy groups is calculated separately. The average performance is obtained by weighting the weighted average communication success rate and the weighted average energy efficiency of the state cluster containing the historical records within each group. By comparing the difference in the average performance of the high-energy and low-energy groups, the trend of the system-level performance objective function relative to the first balance coefficient is estimated. This trend is used as the gradient direction. The gradient directions of the second balance coefficient, the immediate reporting threshold, the regular reporting threshold, or the dormant threshold are estimated using a similar method based on grouping and comparing historical records based on relevant features. This gradient estimation method based on group comparison avoids complex mathematical differentiation operations, making it very suitable for implementation on microcontroller units with limited computing power. It infers the potential direction of parameter adjustment by analyzing the correlations that naturally emerge in historical data. For example, if the average performance of the high-energy group is significantly better than that of the low-energy group, it indicates that the current strategy performs better when energy is sufficient, suggesting that the first balance coefficient should be increased to give higher weight to the energy state in the decision-making process, thereby making fuller use of the high-energy conditions. Based on the gradient direction, the first balance coefficient, second balance coefficient, immediate reporting threshold, regular reporting threshold, or dormant threshold are updated; the parameter updates follow these rules: First, based on the previous parameter update direction and the currently estimated gradient direction, a weighted average is calculated using a preset momentum coefficient to obtain the current parameter update direction. Then, the current values ​​of the first balance coefficient, second balance coefficient, immediate reporting threshold, regular reporting threshold, or dormant threshold are added to a small adjustment amount determined by a preset learning rate and the sign function of the current parameter update direction to obtain the updated values ​​of the first balance coefficient, second balance coefficient, immediate reporting threshold, regular reporting threshold, or dormant threshold. Finally, it is ensured that the updated values ​​of the first balance coefficient, second balance coefficient, immediate reporting threshold, regular reporting threshold, or dormant threshold fall between preset minimum and maximum values. The learning rate can be set to a very small value, such as 0.01, to ensure that each update is small and avoids drastic policy fluctuations. The momentum coefficient can be set to 0.5. The effect of introducing the momentum coefficient is that the parameter update direction not only considers the current gradient estimate, but also retains a part of the historical update direction. This helps to smooth the optimization path, prevents oscillations when there is noise in the gradient direction estimate, and may help the optimization process cross narrow gaps, thus more stably tending towards the region with better performance. The parameter values ​​are limited to a preset range, ensuring the predictability and stability of the system behavior and preventing the optimization process from generating unreasonable parameter values. Write the updated values ​​of the first balance coefficient, the second balance coefficient, the immediate reporting threshold, the regular reporting threshold, or the sleep threshold into non-volatile memory for use in subsequent decisions; The effect of writing the updated parameters to non-volatile memory is that the learned optimization strategy is persistently saved. When the device experiences a power outage and restart, the optimization strategy is retained and continues to take effect, thereby achieving continuous and cumulative performance improvement of the device throughout its entire life cycle.

[0024] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0025] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0026] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0027] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0028] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0029] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0030] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A smart sensing and reporting system for manhole cover status based on cellular Internet of Things, characterized in that, Specifically, it includes: The state perception and quantification module, the urgency fusion calculation module, the decision execution and state control module, and the parameter adaptive optimization module are connected in sequence, wherein; State perception and quantization module: It is activated by the microcontroller unit at a preset perception period or when triggered by a touch event. It samples the voltage signal of the device power battery processed by the voltage divider circuit through its analog-to-digital conversion channel, queries the current received signal strength from the wireless communication unit through the serial communication interface, and reads the latest determined event flag representing the state of the manhole cover from the non-volatile memory. Event flags are generated by the microcontroller unit after recognizing the manhole cover images captured by the image acquisition module, and include at least three states: normal, slightly abnormal, and severely damaged. Based on a preset mapping relationship, the battery voltage signal is converted into an energy availability parameter that characterizes the remaining energy available in the device, the received signal strength is converted into a signal reliability parameter that characterizes the reliability of the current wireless channel connection, and the event flags are mapped into event priority parameters that characterize the urgency of the manhole cover status. Urgency fusion calculation module: After obtaining the energy availability parameter, signal reliability parameter and event priority parameter, it performs urgency score fusion calculation. It uses preset calculation rules to perform nonlinear transformation on the energy availability parameter and signal reliability parameter respectively and weighted combination, and then multiplies it with the event priority parameter to generate an urgency score that comprehensively reflects whether the current manhole cover status should be reported immediately. Decision execution and state control module: Compares the urgency score with multiple preset thresholds stored in non-volatile memory. Based on the comparison result, it drives an internal state machine to perform state transitions and triggers corresponding operations. The operations include: the wireless communication unit immediately initiates a network connection and reports an event flag containing the manhole cover status and related data, or suspends the reporting task and enters low-power sleep mode, or only records the current time and status information in non-volatile memory without starting the wireless communication unit; after each communication attempt is completed, a communication result is generated. Parameter adaptive optimization module: After each communication attempt is completed, the communication result is saved as a historical record, and the historical record is statistically analyzed periodically. Based on the analysis results, the weighting coefficient in the urgency score fusion calculation process is dynamically adjusted or the values ​​of multiple preset thresholds are adjusted.

2. The intelligent sensing and reporting system for manhole cover status based on cellular Internet of Things as described in claim 1, characterized in that: In the state sensing and quantification module, the specific operation of converting the battery voltage signal into an energy availability parameter is as follows: The microcontroller unit performs digital filtering on the battery voltage signal obtained by the analog-to-digital conversion channel after it has been processed by the voltage divider circuit; The filtered battery voltage signal is used as input to look up the battery discharge characteristic curve in a pre-stored non-volatile memory to obtain the basic energy estimate; An aging compensation factor based on the cumulative number of battery working cycles is introduced to simulate the degradation of battery performance with the number of uses. The calculation process is as follows: First, calculate the sum of the rated number of working cycles and the cumulative number of working cycles to obtain the calculation result. Second, divide the rated number of working cycles by the calculation result to obtain the aging compensation factor. A preset attenuation coefficient between 0 and 1 is introduced to adjust the intensity of aging compensation; The energy availability parameter is obtained by multiplying the base energy estimate by the compensation term. The calculation process of the compensation term is as follows: First, calculate the difference between the value and the aging compensation factor. Second, calculate the product of the preset attenuation coefficient and the difference. Finally, subtract the product from the value to get the compensation term.

3. The intelligent sensing and reporting system for manhole cover status based on cellular Internet of Things as described in claim 2, characterized in that: The specific operation of converting the received signal strength into a signal reliability parameter is as follows: The microcontroller unit obtains the received signal strength in decibels and milliwatts from the wireless communication unit; The conversion process uses a piecewise nonlinear function, which predefines three thresholds: the minimum communication threshold, the reliability quality threshold, and the extremely high signal threshold. When the received signal strength is below the minimum communication threshold, the signal reliability parameter is set to zero; When the received signal strength is between the reliable quality threshold and the very high signal threshold, the signal reliability parameter is equal to the value minus the result of the exponent operation with the natural constant as the base, the product of the negative exponent of the exponent operation and the square of the difference between the received signal strength and the reliable quality threshold. When the received signal strength is higher than the extremely high signal threshold, the signal reliability parameter is set to one.

4. The intelligent sensing and reporting system for manhole cover status based on cellular Internet of Things as described in claim 3, characterized in that: The specific operation of mapping event flags to event priority parameters is as follows: The microcontroller unit reads the latest determined event flag written by the image recognition algorithm or touch trigger from a specified address in the non-volatile memory. The event flag is an enumerated type flag that represents the state of the manhole cover. Based on the preset mapping relationship, event flags with different enumeration values ​​are directly mapped to different fixed values ​​as event priority parameters; among them, event flags indicating normal state synchronization are mapped to the first value, event flags indicating minor abnormalities in the manhole cover are mapped to the second value greater than the first value, and event flags indicating severe damage to the manhole cover are mapped to the third value greater than the second value.

5. The intelligent sensing and reporting system for manhole cover status based on cellular Internet of Things as described in claim 4, characterized in that: In the aforementioned urgency fusion calculation module, the specific operations for performing nonlinear transformations on the energy availability parameter and the signal reliability parameter are as follows: A first nonlinear transformation is applied to the energy availability parameter. The first nonlinear transformation is a logarithmic function with the natural constant as the base. The argument of the logarithmic function is the sum of the product of the value and a preset adjustment factor multiplied by the energy availability parameter. The value of the logarithmic function is divided by the natural logarithm of the sum of the value and the adjustment factor. The result is used as the output of the first nonlinear transformation. A second nonlinear transformation is applied to the signal reliability parameter. This second nonlinear transformation is defined by a piecewise function, which has a preset signal reliability decision threshold and a preset steepness coefficient. When the signal reliability parameter is below the signal reliability decision threshold, the output of the second nonlinear transformation is set to zero. When the signal reliability parameter reaches or exceeds the signal reliability decision threshold, the output of the second nonlinear transformation is obtained through the following calculations: First, the first difference is obtained by subtracting the signal reliability decision threshold from the signal reliability parameter. Second, the second difference is obtained by subtracting the signal reliability decision threshold from the first difference. Then, the first difference is divided by the second difference to obtain a ratio. Next, the product of the negative steepness coefficient and this ratio is calculated, and this product is used as an exponent for exponential operation with the natural constant as the base. Finally, the result of the exponential operation is subtracted from the first difference, and the obtained value is the output of the second nonlinear transformation.

6. The intelligent sensing and reporting system for manhole cover status based on cellular Internet of Things as described in claim 5, characterized in that: The specific process for generating a comprehensive urgency score is as follows: The output of the first nonlinear transformation is multiplied by the first balance coefficient to obtain the first weighted result; the output of the second nonlinear transformation is multiplied by the second balance coefficient to obtain the second weighted result; the sum of the first balance coefficient and the second balance coefficient is one. The first weighted result is added to the second weighted result to obtain a comprehensive environmental fitness factor; The environmental fitness factor is multiplied by the event priority parameter to obtain the final urgency score.

7. A manhole cover status intelligent sensing and reporting system based on cellular Internet of Things as described in claim 6, characterized in that: The decision execution and state control module maintains an internal state machine. The state set of the internal state machine includes a dormant state, a ready state, a reporting attempt state, and a suspended state. The multiple preset thresholds stored in the non-volatile memory include at least an immediate reporting threshold, a regular reporting threshold, and a dormant threshold, wherein the value of the immediate reporting threshold is greater than the value of the regular reporting threshold, and the value of the regular reporting threshold is greater than the value of the dormant threshold. The specific process of driving the internal state machine to perform state transitions based on the comparison results is as follows: First, a dynamic decision factor is calculated. The dynamic decision factor is equal to the difference between the value and a preset historical weight factor, plus the product of the historical weight factor and a recent communication success rate moving average. Then, the immediate reporting threshold is multiplied by the dynamic decision factor to obtain the immediate reporting effective threshold, and the regular reporting threshold is multiplied by the dynamic decision factor to obtain the regular reporting effective threshold, while the dormant threshold remains unchanged. Finally, the urgency score is compared with the dormancy threshold, the regular reporting activation threshold, and the immediate reporting activation threshold. Based on the comparison results and the current state of the internal state machine, the state transition target is determined. The decision execution and state control module generates a communication result each time it exits the reporting attempt state.

8. The intelligent sensing and reporting system for manhole cover status based on cellular Internet of Things as described in claim 7, characterized in that: The operation triggered based on the comparison result is as follows: When the urgency score is greater than or equal to the immediate reporting threshold, the control wireless communication unit immediately initiates a network connection and reports data, while the internal state machine transitions to the reporting test state. When the urgency score is less than the immediate reporting threshold but greater than or equal to the regular reporting threshold, the internal state machine transitions to a suspended state, starts a delay timer, and controls the device to enter low-power sleep mode. The delay duration of the delay timer is equal to a preset base delay unit duration multiplied by the square of a ratio. The ratio is calculated as follows: first, the larger value between the urgency score and a preset minimum normal number is taken to obtain a comparison base; then, the immediate reporting threshold is divided by the comparison base to obtain the ratio. When the urgency score is less than the regular reporting threshold and greater than or equal to the dormancy threshold, the internal state machine transitions to or remains in the ready state. When the urgency score is less than the sleep threshold and the internal state machine is not currently in a suspended state, the current time, urgency score, energy availability parameter, signal reliability parameter, and event flag are combined into a single state information and appended to the log storage area of ​​the non-volatile memory. Then, the control device enters a sleep state.

9. A manhole cover status intelligent sensing and reporting system based on cellular Internet of Things as described in claim 8, characterized in that: The specific process of statistical analysis of historical records in the parameter adaptive optimization module is as follows: A health weight is calculated for each historical record. The calculation process is as follows: First, the difference between the current time and the time when the historical record was generated is calculated; second, the result of the exponentiation operation with the natural constant as the base is calculated, and the product of the time difference and a preset time decay coefficient with a negative exponent is obtained to get the time decay factor; then, the successful or failed connection flag in the historical record is added to a preset positive constant to get the success factor; finally, the time decay factor is multiplied by the success factor to get the health weight of the historical record. The energy availability parameter is divided into several consecutive first intervals, and the signal reliability parameter is divided into several consecutive second intervals. Each first interval and each second interval are combined to form multiple two-dimensional state clusters. For each state cluster, the weighted average communication success rate and weighted average energy efficiency of that state cluster are calculated using the health weights of the historical records belonging to that state cluster.

10. A manhole cover status intelligent sensing and reporting system based on cellular Internet of Things as described in claim 9, characterized in that: The process of dynamically adjusting the weighting coefficients or adjusting the values ​​of multiple preset thresholds in the urgency score fusion calculation based on the analysis results is as follows: Construct a system-level performance objective function, which is the weighted sum of the natural logarithms of the weighted average communication success rate and the weighted average energy efficiency of each state cluster; By analyzing historical records, we evaluate the impact trend on the system-level performance objective function when the first balance coefficient, second balance coefficient, immediate reporting threshold, regular reporting threshold, or dormancy threshold take different values, and estimate the gradient direction of the system-level performance objective function relative to these parameters. Based on the gradient direction, the parameters are updated using the gradient descent method with momentum, and the updated parameter values ​​are written to non-volatile memory.

Images