Long-term rental house monitoring management system and method

Abstract

The application relates to the technical field of house renting management, and discloses a long-term house renting monitoring management system and method, which comprises the following steps: dividing monitoring units in the same physical space into equipment groups, recording state jump time stamps, automatically depositing behavior logic templates based on historical time stamp sequences, and generating risk early warnings through logical self-consistency comparison between real-time sequences and templates. The application enables the system to understand the cooperative behavior logic of the equipment, realizes cognitive transition from physical quantity monitoring to behavior logic verification by constructing behavior templates through reuse of basic switch signals, and simultaneously utilizes state credibility evaluation and a distributed priority arbitration mechanism to naturally form an anti-interference data quality firewall and system-level resource coordination capability under a centerless architecture, thereby significantly improving the economy and reliability of intelligent reconstruction of old houses.

Description

TECHNICAL FIELD

[0001] The present application relates to a long-term rental housing monitoring management system and method, belonging to the technical field of rental housing management data processing. BACKGROUND

[0002] In the field of long-term rental housing management, for vulnerable groups such as nursing homes and elderly care, the existing technology generally adopts a single-point sensor threshold alarm method, that is, by deploying temperature and humidity, smoke, current and other independent sensors inside the house, monitoring whether a specific physical quantity exceeds the preset threshold and triggering an alarm. This scheme is considered as a standard management means, and its core assumption is that the risk is caused only by the over-standard of a single physical quantity. However, when the system faces the chronic risk caused by the coupling of cross-device behavior logic, this method has some fundamental defects.

[0003] For example, when a tenant turns on a high-power heating device for a long time but closes the doors and windows tightly, the traditional system remains silent because the values of each sensor do not exceed the threshold, and it cannot identify the risk of hypoxia caused by the increase of carbon dioxide concentration or the fire hazard caused by the continuous use of heating equipment due to the combination of continuous heating and zero ventilation in a closed space. Specifically: 1. The system can only perceive the instantaneous state of the physical quantity and cannot understand the behavior coordination relationship between the device group, such as the correlation between the opening of the heater and the closing of the doors and windows, resulting in complete oversight of the risk of behavior combination. For example, in a closed space, people's breathing will continuously consume oxygen and produce carbon dioxide, and the long-term accumulation will cause the indoor oxygen concentration to decrease and the carbon dioxide concentration to increase, thereby causing the risk of hypoxia or air quality deterioration; 2. The system relies on the original physical quantity value to judge the risk and ignores the behavior logic information contained in the timing of device state jump, resulting in a lag in the system's response to chronic and cross-domain risks.

[0004] The industry has tried to alleviate the problem by increasing the density of sensors or optimizing the threshold settings, but such improvements are still limited by: the increase in the number of sensors to raise the deployment cost, which is difficult to adapt to the renovation of old houses; threshold optimization requires frequent manual calibration, which cannot adapt to dynamic tenant behavior; unable to break through the shackles of isolated data point analysis, unable to deconstruct the device coordination logic from a system level. The above bottlenecks are essentially due to the systematic neglect of the behavior coordination state of the device group, which has become a common problem that restricts the improvement of the efficiency of intelligent rental housing management. Therefore, how to build a management system that does not rely on physical quantity thresholds and realizes the active perception of implicit risks through the self-consistency of device behavior logic, and then solve the problem of accurate identification of cross-device coordination failure, has become a technical problem to be solved by the present application. SUMMARY

[0005] The present application provides a long-term rental housing monitoring management system, which mainly aims to solve the problem of the inability of the prior art to actively identify the risk of device group behavior logic disruption.

[0006] To achieve the above object, the application provides a long-term house renting monitoring and management system, comprising:

[0007] A device group definition module is configured to logically define at least two monitoring units deployed in the same physical space and having inherent correlation in behavior as a device group, wherein the physical space comprises a kitchen or a bathroom in an apartment, and the monitoring units comprise intelligent electrical sockets, intelligent door magnetic sensors or intelligent water flow switches.

[0008] A state jump record module is configured to record the unique identification code of the monitoring unit and the accurate time stamp when the state jump occurs when the running state of the monitoring unit jumps from one state to another state.

[0009] A behavior logic template sedimentation module is configured to automatically learn and generate one or more high-frequency behavior logic templates representing the regular life habits of tenants based on the accurate time stamp sequence recorded by the device group within a preset initial learning period through an unsupervised learning algorithm, and the high-frequency behavior logic template represents the stable timing correlation between the state jumps of each monitoring unit in the device group; wherein the preset initial learning period is a time period that can be pre-configured by the housing manager according to the actual application scenario, for example, it can be set to the first two weeks after the tenant moves in, or the first 14 days of system operation, to ensure that sufficient data for analyzing the stable life habits of users can be collected.

[0010] A logic self-consistency verification module is configured to continuously compare the accurate time stamp sequence output by the state jump record module in real time with the high-frequency behavior logic template representing the regular life habits of tenants, and when there is a behavior logic deviation between the real-time time stamp sequence and the high-frequency behavior logic template, a risk warning signal related to the potential safety risk or abnormal life of the tenant is generated; the determination of the behavior logic deviation is based on the following logic: when the absolute time difference between the actual state jump time of a certain monitoring unit and the expected state jump time indicated by the high-frequency behavior logic template exceeds the preset logic tolerance threshold of the template, it is determined that there is a behavior logic deviation inconsistent with the regular life habits learned by the tenant.

[0011] Preferably, the running state recorded by the state jump record module is two binary states of on and off; and the high-frequency behavior logic template generated by the behavior logic template sedimentation module comprises typical timing between the state jumps of different monitoring units in the group and the time interval range corresponding to each typical timing.

[0012] Preferably, the system further comprises a state credibility evaluation module configured to: based on the accurate time stamps recorded by the state jump recording module, dynamically calculate and update the historical average stable running time length of each monitoring unit in the device group; and before the comparison by the logical self-consistency verification module, when the state jump recording module records a state jump of a certain monitoring unit, the state credibility evaluation module calculates the actual time interval between this state jump and the previous state jump of the monitoring unit; if the actual time interval is less than ten percent of the historical average stable running time length of the monitoring unit, the state credibility evaluation module enters a delay confirmation mode; only when the state of the monitoring unit has not returned to the state before the jump after the delay confirmation mode lasts for a preset time length, the state credibility evaluation module recognizes the validity of this state jump, and submits the time stamp of this state jump to the behavior logic template sedimentation module for template generation during the initial learning period; after the initial learning period ends, the time stamp of this state jump is submitted to the logical self-consistency verification module for real-time comparison.

[0013] Preferably, the system further comprises a priority label configuration module configured to assign different priority numerical labels to at least two monitoring units consuming the common resource belonging to at least two different device groups.

[0014] Preferably, the system further comprises a resource competition logic arbitration module configured to: before the state of a monitoring unit configured with a lower priority numerical label is about to be switched from a non-running state to a running state, based on the latest accurate time stamps of other monitoring units recorded by the state jump recording module, determine whether there is a monitoring unit configured with a higher priority numerical label that is currently in a running state; if so, the resource competition logic arbitration module delays the state switching of the lower priority unit until the higher priority monitoring unit returns to a non-running state.

[0015] Preferably, the common resource includes the total electricity load of an apartment or the total water pressure of an apartment.

[0016] Preferably, the system further comprises a pre-warning signal sending module configured to send the risk pre-warning signal generated by the logical self-consistency verification module to a remote management server through a wireless communication network.

[0017] Preferably, the behavior logic template sedimentation module, during the initial learning period, counts the occurrence frequency of the state jump sequence of each monitoring unit in the device group, and identifies the sequence with an occurrence frequency higher than eighty percent as a high-frequency behavior logic template.

[0018] The application discloses a long-term housing risk early warning method, characterized in that the method comprises the following steps: step a, logically defining at least two monitoring units arranged in the same physical space as a device group; step b, recording the unique identification code of the monitoring unit and the accurate time stamp when the state of the monitoring unit jumps from one state to another state; step c, automatically generating one or more high-frequency behavior logic templates based on the accurate time stamp sequence recorded by the device group in a preset initial learning period, wherein the high-frequency behavior logic template represents the stable timing correlation between the state jumps of the monitoring units in the device group; and step d, continuously comparing the real-time output accurate time stamp sequence with the high-frequency behavior logic template, and generating a risk warning signal when there is a behavior logic deviation between the real-time time stamp sequence and the high-frequency behavior logic template; the behavior logic deviation is determined based on the following logic: when the absolute time difference between the actual state jump time of a certain monitoring unit and the expected state jump time indicated by the high-frequency behavior logic template exceeds the preset logical tolerance threshold of the template, it is determined that there is a behavior logic deviation.

[0019] Compared with the prior art, the application has the following beneficial effects:

[0020] 1. For the weak groups, such as nursing homes, and the scenes of taking care of the elderly, the time stamp sequence of the state jump of the device is converted into a behavior logic template, the system obtains the autonomous insight ability of the device cooperation law, when the real-time behavior sequence deviates from the timing logic preset by the template, for example, the exhaust fan is not started within the reasonable time window after the water heater is turned off, the system automatically identifies the logical vacancy of such key steps, the cognitive upgrading from physical quantity monitoring to behavior logic verification enables the system to capture the cross-device cooperation failure risk that cannot be perceived by the traditional threshold mechanism, and provides higher-dimensional information support for management decision-making; the binary state signal of the existing device, such as the on-off record of the intelligent socket, is reused to construct the behavior logic template, and high-precision sensors do not need to be deployed, when the water heater continuously runs and the door and window sensor does not respond, the system can determine the ventilation logic fracture only by comparing the time correlation of the state jumps of the two, and the logical reconstruction of low-dimensional data enables the old house reconstruction to be free from the dependence on special hardware, and significantly reduces the technical threshold and marginal cost of large-scale landing.

[0021] 2、State credibility assessment module analyzes the distribution characteristics of the device's historical stable operation time length, for example, triggering a delay confirmation for abnormal jumps shorter than 10% of the historical average. This naturally filters out transient signal interference. The resource competition arbitration module uses pre-set priority labels and global timestamp visibility to make low-priority devices (such as bathroom water heaters) automatically delay startup when detecting high-priority devices (such as kitchen ovens) running. These two mechanisms form a closed loop with the core behavior logic verification, ensuring input data quality, expanding system-level collaboration capabilities through priority arbitration, and providing decision-making basis through behavior template analysis. The three are coupled through timestamp data flow to form a self-consistent robust system that can resist signal disturbances and resource conflicts without external compensation mechanisms.

[0022] 3、Low-priority devices avoid resource conflicts by analyzing global timestamp bulletin boards, such as instant water heaters that pause startup when total water pressure is insufficient. This process does not rely on centralized controllers or real-time communication, but rather on distributed interpretation of public information and local execution of static rules, maintaining the simplicity of edge computing architecture while naturally emerging system-level resource coordination capabilities. The early warning signal sending module maps inconsistent behavior logic events (such as continuous water use without ventilation) into structured alarms, directly pointing to specific device group collaboration failure scenarios. This allows remote management servers to issue disposal instructions based on precise device control actions, such as forcibly shutting down the water heater and starting the exhaust fan, forming a closed-loop management path of risk identification-decision output-execution feedback.

[0023] 4、In the context of nursing homes and elderly care, potential safety risks and resource usage conflicts are identified by analyzing the timing logic of device operation rather than relying on isolated physical quantity thresholds. This is because the life patterns of this specific population are usually highly regular, providing an ideal data foundation for the system to automatically generate accurate and stable high-frequency behavior logic templates based on unsupervised learning algorithms. Therefore, whether it is a behavior sequence interruption caused by acute events such as falls, or chronic risks caused by declining cognitive abilities and forgetting to turn off appliances, can be accurately identified as deviations from existing safety templates through logic self-consistency verification modules, generating risk warning signals. More importantly, this method only relies on binary state data from smart electrical outlets, door magnets, and other monitoring units, achieving non-intrusive passive monitoring while protecting user privacy and dignity, addressing the limitations of traditional invasive monitoring solutions, and providing an important passive safety net for people who need special care. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 State jump record and behavior logic verification flowchart in the long-term rental monitoring and management system of the present application;

[0025] Figure 2 The learning curve graph of the behavior logic template stability and matching accuracy of the long-term rental house monitoring management system of the present application;

[0026] Figure 3 The modular structure schematic diagram of the long-term rental house monitoring management system of the present application;

[0027] Figure 4 The template index and recognition quantity change graph in the learning cycle of the present application.

[0028] The object implementation, function features and advantages of the present application will be further explained with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0029] It should be understood that the specific embodiments described herein merely exemplify the present application and are not intended to limit the present application.

[0030] The embodiment of the application provides a long-term housing monitoring management system and method, for the weak group, such as the scene of nursing home, solitary old people, aims at analyzing the timing logic of equipment operation, rather than relying on isolated physical quantity threshold, to actively identify potential safety risks and resource use conflicts, the monitoring unit deployed in the apartment can include the intelligent electric socket for supplying power for the toilet water heater, the intelligent door magnetic sensor installed on the toilet door, and another intelligent electric socket configured for the large-power electric appliance such as the electric oven in the kitchen, these monitoring units jointly constitute a set that needs to be cooperatively monitored in physical space and functional logic, the initialization operation of the system starts from the functional implementation of the device group definition module, one specific implementation mode of the module is a software configuration interface for the house manager, the interface can be embodied as an application program installed on a smart phone or a personal computer, the manager can define at least two monitoring units that are deployed in the same physical space and have inherent correlation in behavior as a device group in logic through the interface, for example, the toilet intelligent electric socket and the intelligent door magnetic sensor are defined as a bathroom behavior group by the manager, the inherent logic is that the shower behavior will inevitably cause the orderly change of the states of the two devices, the essence of the definition process is to create an association record in the configuration database in the background of the system, the record binds a unique group identification code and a set of unique identification codes of the monitoring units, after the group definition is completed, the state jump record module starts to continuously work on each monitoring unit, the module is usually a program fixed in the hardware firmware of the monitoring unit, when the running state of the monitoring unit changes, such as the intelligent electric socket jumps from the closed state to the opened state, the module immediately captures the event, it records two core information: the unique identification code of the monitoring unit, the identification code is usually the MAC address or a globally unique factory serial number of the monitoring unit;and the precise time stamp when the state jump occurs, to ensure the uniformity and comparability of cross-device time stamps, all monitoring units periodically synchronize with the same time server through the network time protocol, so that the precise time stamp can achieve millisecond-level synchronization accuracy, in the embodiment, the running state is limited to two binary states of basic and reliable on and off states, which significantly reduces the requirement for sensor hardware complexity, then the system enters the core intelligent learning stage, which is executed by the behavior logic template sedimentation module, in a preset initial learning period, for example, the first two weeks after the tenant moves in, all state jump records in the specified device group are collected and analyzed, the ultimate goal is to automatically mine high-frequency behavior logic templates that can represent the stable living habits of tenants from these raw time series data streams through unsupervised learning algorithms, specifically, the module first aggregates the collected time stamp sequences by device group, for the aforementioned bathroom behavior group, the system may accumulate a large number of sequence fragments such as [time T1, door magnet, off]>[time T2, socket, on]>[time T3, socket, off]>[time T4, door magnet, on], the algorithm will perform deep pattern matching and frequency statistics on these sequences, a feasible unsupervised learning algorithm is the sequence pattern analysis technology based on time series data mining, which can identify sequence patterns that repeatedly appear and have stable time intervals between steps, for example, the algorithm will find that after the door closing event, the socket opening event will probably occur within 5 to 60 seconds, and after the socket opening event, the socket closing event will probably occur within 15 to 30 minutes, when the occurrence frequency of a certain time sequence is higher than a preset threshold in all recorded sequences in the initial learning period, here set to eighty percent, the sequence is identified and solidified as a high-frequency behavior logic template, the setting of the eighty percent threshold is essentially a trade-off between universality and specificity in the template generation process, a too high threshold, for example, ninety-five percent, may not form an effective template due to occasional normal behavior variation of tenants, resulting in insufficient coverage; on the contrary, a too low threshold, for example, fifty percent, may incorrectly absorb some accidental, irregular behaviors as templates, thereby causing excessive false alarms in the subsequent verification stage; therefore, eighty percent is considered a robust engineering choice that can ensure that the template represents the core behavior rules while effectively filtering out random noise.

[0031] The high-frequency behavior logic template formed by the final precipitation has a data structure that not only contains the typical time sequence of device state transitions, such as door magnet. closed, socket. opened, socket. closed, door magnet. opened, but more importantly, it also accurately records the stable time interval range between each two consecutive state transitions in the sequence, for example, 5 seconds-60 seconds, 15 minutes-30 minutes, 2 seconds-45 seconds. This complete structure containing time sequence and interval constitutes the fundamental basis for subsequent logic self-consistency verification. After the initial learning period ends, the system seamlessly switches to continuous monitoring mode, with the logic self-consistency verification module taking on the real-time analysis task. This module receives the accurate timestamp sequence output by the state transition recording module in real time and continuously performs pattern matching with all generated high-frequency behavior logic templates. When a behavior sequence that matches the initial part of the template starts, for example, the bathroom behavior group detects that the door magnet is closed, the verification module will predict the expected state transition time window of the next expected state transition, i.e., the opening of the intelligent socket, based on the corresponding time interval range recorded in the template. If the absolute time difference between the actual state transition time of a monitoring unit and the expected state transition time indicated by the high-frequency behavior logic template based on the previous event exceeds the preset logic tolerance threshold of the template, the system determines that there is a behavior logic deviation. The logic tolerance threshold here is an additional buffer band attached to the time interval range inherent in the template, and its purpose is to accommodate reasonable immediacy fluctuations that do not affect the essence of the behavior logic. For example, if the template expects the water heater to be closed within 15 to 30 minutes, a 2-minute logic tolerance threshold means that as long as the water heater is closed within the wider window of 13 to 32 minutes, it is considered logically consistent. However, if the water heater remains on for 40 minutes and then closes, its closing time has fallen significantly outside this window, and the logic self-consistency is determined to be broken. At this time, the verification module will immediately generate a structured risk warning signal, which is sent to the remote management server through the warning signal sending module via the wireless communication network deployed in the apartment, for timely research and disposal by the management personnel. To improve the overall quality of input data and prevent false positives due to circuit transient disturbances or user's unintentional rapid operations, the system also integrates a state credibility assessment module. This module performs a pre-posed effectiveness inquiry before submitting any state transition data to the core logic analysis engine. It dynamically calculates and continuously updates the historical average stable running time for each monitoring unit. When the state transition recording module captures a new state transition, the state credibility assessment module calculates the actual time interval between this transition and the previous state transition of the monitoring unit. If this actual time interval is significantly less than the device's regular behavior pattern, for example, less than 10% of its historical average stable running time, the module determines that this transition is likely a signal noise or invalid operation. At this time, it will enter a short delay confirmation mode,Only when the state of the monitoring unit does not return to the original state before the state jump after a preset delay confirmation duration, such as 5 seconds, the module finally recognizes the service validity of this state jump and submits its timestamp to the behavior logic template deposition module or the logic self-consistency verification module. This mechanism builds a barrier at the front end of the data stream, effectively filtering invalid data glitches caused by physical reasons or human error touch. In addition, to deal with the systemic risks that may be caused by multiple devices competing for limited public resources, the system also includes a set of distributed resource coordination mechanisms. First, the priority tag configuration module allows the administrator to configure different priority digital tags for monitoring units that consume public resources. For example, in the scenario where the total power load of an apartment is limited, the smart socket associated with the kitchen electric oven that guarantees basic cooking needs can be configured as a high priority tag, while the smart socket of the water heater used for bathing is configured as a lower priority tag. Here, the public resource can be the total power load of the apartment, or the total water pressure of the apartment. Then, the resource competition logic arbitration module makes autonomous decisions based on these tags. The function of this module is embedded in the firmware of each related monitoring unit. Before a monitoring unit configured with a lower priority digital tag, such as a water heater, switches from a non-running state to a running state, its built-in arbitration logic will actively query the running state of other devices in the current network. This process is achieved by listening to the state jump records broadcast globally, each record containing an accurate timestamp and device identity information. If it determines that there is a monitoring unit configured with a higher priority digital tag, such as an electric oven, that is currently in a running state, the resource competition logic arbitration module will actively delay the state switch request of the lower priority unit. It will continue to monitor the network state until it detects the broadcast of the higher priority monitoring unit returning to the non-running state, and then allow its startup process to continue. This decentralized arbitration mechanism, through self-awareness of information between devices and local rule execution, skillfully avoids problems such as circuit overload or water pressure drop caused by simultaneous startup of high-power appliances, achieving system-level resource coordination and safe operation without increasing the hardware cost of centralized controllers.

[0032] Embodiment 1: In the scenario of nursing home, the internally deployed monitoring unit includes a smart electrical socket connected to a high-power electric heater and a smart door magnetic sensor installed on the bathroom door. During the initial learning period, the system automatically identifies and solidifies the event sequence of the electric heater being turned on shortly after the bathroom door is closed and being turned off after a relatively fixed long period of time as a high-frequency behavior logic template through the behavior logic template sedimentation module. This template not only defines the sequence of device actions, but more importantly, sets a stable time interval range for the electric heater's on duration based on historical data. The effectiveness of this learning process is based on the continuous operation of the preposed state credibility assessment module: when the instantaneous fluctuation of the power grid causes the smart electrical socket of the electric heater to produce continuous on-off jumps within milliseconds, the module calculates that the actual time interval is much smaller than 10% of the historical average stable running time of the device. Then it enters the delay confirmation mode. After confirming that the state is not sustained and effective, such signal noise is filtered. This purification mechanism ensures the high fidelity of the data stream input to the behavior logic template sedimentation module, so that the final generated behavior logic template can accurately represent the user's true and stable habits.

[0033] In the subsequent daily operation, the logical self-consistency verification module takes the aforementioned high-fidelity behavior logic template as the benchmark for continuous comparison. When it detects that the actual on-time of the electric heater has exceeded the sum of the upper limit of the stable time interval embedded in the template and the logical allowable deviation threshold, the system generates a risk warning signal. The fundamental basis of this warning is not whether the power of the electric heater or the environmental temperature exceeds some isolated physical threshold, but rather the duration of its operation, which is a behavioral characteristic that has deviated logically from the behavior template that has been learned as a safe way. This achieves active identification of potential fire risks that are slowly accumulated due to abnormal behavior, which cannot be perceived by traditional threshold alarm mechanisms. Further, an intelligent electrical outlet associated with an electric oven is also deployed in the kitchen of the apartment. The outlet is assigned a higher priority digital tag via the priority tag configuration module. When the tenant is using the electric oven and enters the bathroom and triggers the behavior logic related to showering, the resource competition logic arbitration mechanism inside the system is activated, and the arbitration logic in the bathroom water heater intelligent socket determines that the electric oven with higher priority is in operation. Before starting the water heater, it will analyze the state jump record broadcast globally to determine that the electric oven is in operation. Based on this, the arbitration logic will definitely delay the start of the water heater until it monitors that the state of the electric oven returns to non-operation. This process demonstrates an inherent conflict resolution capability: within a single architecture, the system makes local decisions based on static priority rules in a distributed manner, so that the system-level requirement of ensuring global power safety has priority over meeting local, non-immediate comfort requirements, thereby resolving potential resource competition conflicts among multiple high-power devices without relying on any central controller intervention.

[0034] Example 2: To verify the actual effectiveness of the system of the present application in identifying chronic risks caused by cross-device behavior logic coupling, the following test was performed, which aimed to accurately compare the response differences between the system of the present application and traditional single-point physical quantity threshold alarm systems in dealing with the same continuous heating, continuous closed risk scenario. The test platform was built in a software simulation environment, which internally integrated a standard single-room thermodynamic and aerodynamic model. The following key virtual components were deployed in the platform: a virtual electric heater with a rated power of 1500W, whose start-stop state was recorded and controlled by a simulated intelligent electrical socket; a virtual window, whose opening and closing state was recorded by a simulated intelligent door magnetic sensor; a traditional carbon dioxide concentration sensor as a control group, whose alarm threshold was set to 1500ppm according to indoor environmental safety standards; and a monitoring and management system that integrated all the functional modules of the present application deployed on a virtual edge computing node. The key parameters in the test were set according to the determined engineering logic: first, the total simulation time was set to 120 minutes, which was a balance between the integrity of the test process and the effective use of computing resources, ensuring that the time was sufficient to cover a typical behavior sequence completely and provide a sufficient time window for observing the cumulative effects of chronic risks; second, the system core logic tolerance threshold, which was essentially a trade-off between the sensitivity of the system response and the tolerance to normal behavior fluctuations, was set to avoid false alarms due to small changes in user habits, while ensuring timely detection of significant abnormalities. The threshold value was associated with the average duration of the corresponding step in the learned behavior template, and in this test it was set to ten percent of the heating duration in the template. The test process was divided into two stages. The first stage was behavior logic template sedimentation, in which the platform performed 10 simulations of normal heating behavior sequences, each sequence including window closing, electric heater turning on, continuous running for 28 to 32 minutes, electric heater turning off, and then window opening for ventilation. The state jump recording module in the system of the present application accurately recorded the time stamps of each jump. Based on these sequence data, the behavior logic template sedimentation module output a high-frequency behavior logic template after statistically analyzing sequences with a frequency higher than 80%. This template clearly defined the normal operation duration of the electric heater as 28 minutes and 32 minutes. The second stage was risk scenario verification, in which the platform started an abnormal heating sequence simulation: the window was closed, the electric heater was turned on and continuously ran, but after the expected 32-minute time point, both the window and the electric heater remained in their original states until the 120-minute simulation ended. In this process, the test platform simultaneously recorded the state outputs of the system of the present application and the traditional CO2 sensor. The core data comparison at different key time points is shown in the following table:

[0035] Table 1: CO2 concentration and risk warning signal table in long-term rental monitoring and management system.

[0036]

[0037] Data shows that, at the 36th minute of the simulation, although the reading of the traditional CO2 sensor increased significantly, it still did not reach 1500. The alarm threshold of ppm was maintained, so the system remained in a normal state. However, the system of this invention generated a risk warning signal at this moment. The direct cause of this differentiated response is the operation of the logic self-consistency verification module in the system of this invention: this module compares the actual state transition time of the electric heater with the high-frequency behavioral logic template and determines that the absolute time difference has exceeded the preset logic tolerance threshold. Therefore, there is a behavioral logic deviation. Until the later stage of the simulation, the reading of the traditional sensor is always near the threshold and fails to effectively alarm, while the warning state of the system of this invention remains stable. The results of this experiment show that the technical solution proposed by this invention can effectively identify potential safety risks caused by the coordinated failure of the behavioral logic of multiple devices, which are slow to develop and cannot be captured by a single physical quantity threshold. The experimental conclusion forms a tight logical closed loop with the core technical concept of this invention: by converting the timestamp sequence of device state transitions into a behavioral logic template, the system realizes the transformation from physical quantity monitoring to behavioral logic verification. This enables it to make accurate and reliable warnings based solely on the deviation between the behavioral sequence and the safety mode in the early stage of risk accumulation, thereby fundamentally solving the inherent technical defects of the existing technology when facing such chronic and cross-domain risks.

[0038] Example 3: This example combines Figures 1 to 4 This document describes the implementation of a long-term rental housing monitoring and management system and method, such as... Figure 1 As shown in the diagram, scenarios 1 and 2 describe the system's operation flow under normal operation and abnormal conditions (outside the template range), respectively. In scenario 1, when the smart socket's state changes from on to off, the state transition recording module records the transition time. The system further performs logical verification of the query template, and the returned template time is 15-30 minutes. At this time, the behavior logic template library generates high-frequency behavior logic templates that correspond to the device group's state transitions and continuously compares them to ensure consistency with the expected time window, thereby verifying its logical self-consistency. If it meets expectations, the system continues to operate normally. In scenario 1, the warning sending module will not trigger any alarms. In scenario 2, the state change of the smart socket occurs at an abnormal time point, exceeding the expected time window of the behavior logic template. Specifically, the state change record of the smart socket occurs at 14:00:00, while according to the behavior logic template, the expected change time should be 14:28-14:32. The actual change time is 14:36. Therefore, the system triggers the behavior logic verification module to verify and generates a risk warning signal. This signal is sent to the remote management server through the warning sending module so that the management personnel can respond in a timely manner.

[0039] As Figure 2 shown, the graph shows the trend of template matching accuracy and template stability index during the learning period by recording the matching accuracy and stability index of the model. The horizontal axis of the graph is the learning period (days), from the 1st day to the 14th day, and the vertical axis is the percentage (%), representing the percentage of model matching accuracy and template stability index. Curve 1 represents the template matching accuracy, which gradually increases from the initial about 50% to nearly 90% matching accuracy as the learning period extends. Curve 2 represents the template stability index, which is low at the beginning of learning, but gradually improves over time and tends to be stable in the later period of learning, reaching a high stability. This shows that through continuous learning, the system can continuously optimize and improve the matching effect of the behavior logic template, thereby improving the intelligent level of the long-term rental monitoring and management system.

[0040] As Figure 3 shown, the graph shows four typical monitoring units, namely: intelligent electrical socket for water heater control, intelligent door magnetic installed on the bathroom door, intelligent electrical socket for electric oven, and intelligent water flow switch for water pipe monitoring. The system first divides the relevant monitoring units logically deployed in the same physical space into a device group through the device group definition module. The state jump record module is used to record the unique identification code and the accurate timestamp of state jump when the running state of the monitoring unit changes. The state credibility evaluation module dynamically calculates the historical stable running time of the monitoring unit based on the above timestamp, which is used to filter abnormal data jumps before real-time comparison. For monitoring units that consume common resources (such as water, electricity, etc.), the system configures different priority digital tags for them through the priority tag configuration module to facilitate subsequent resource coordination. The above modules collectively support the operation of the core modules, which specifically include: behavior logic template sedimentation module (unsupervised learning), used to automatically generate high-frequency behavior logic templates based on state jump time series; logic self-consistency verification module (real-time comparison), used to perform logical consistency verification between real-time state sequence and template to identify potential risks; resource competition logic arbitration module (distributed decision), used to dynamically determine resource use conflicts and control device operation based on priority tags. The behavior deviations and logic anomalies identified by all core modules in the system are finally transmitted to the remote management server in the form of structured warning signals through the warning signal sending module (wireless communication network) for disposal analysis by the backend. The solid line in the graph represents the data flow, and the dashed line represents the configuration relationship. The black square box indicates the core modules of the system.

[0041] As Figure 4As shown, the horizontal axis indicates the consecutive dates within the system learning period from day 1 to day 14, the left side of the vertical axis indicates the numerical range of the template matching accuracy (%) and the template stability index (%) in percentage (%), and the right side of the vertical axis indicates the trend of the number of recognition templates in the number of diagrams; in the figure, the solid line dot represents the growth trajectory of the template matching accuracy (%), the dashed line diamond represents the change of the template stability index (%), and the dotted triangular curve represents the cumulative growth of the number of recognition templates.

[0042] In the scene of nursing homes and care for the elderly, a long-term rental apartment scene is taken as an example, which includes a bathroom intelligent lighting lamp, a smart electrical socket linked with a water heater, and an independent intelligent exhaust fan switch. In this scene, a complete and safety-compliant showering behavior has a behavior logic chain: lighting on, water heater start, water heater off, exhaust fan start, exhaust fan off, and lighting off. The core task of the system in the initial learning period is to perform a three-stage unsupervised learning process aimed at building a high-frequency behavior logic template. The first stage is session segmentation, the system takes the state jump of the bathroom intelligent lighting lamp turning on as the starting marker of the behavior session and the state jump of the bathroom intelligent lighting lamp turning off as the ending marker, and the accurate time stamp sequence of all the state jumps of the monitoring units occurring during this period is segmented from the continuous data stream into independent session segments. The second stage is sequence vectorization and clustering, the system uses an edit distance-based sequence similarity measurement algorithm to calculate the similarity scores between the session segments, thereby converting the time series pattern matching problem into a vector distance calculation problem in high-dimensional space. Then, a density-based clustering algorithm DBSCAN suitable for time series data is executed, which automatically aggregates session sequences with high enough similarity into the same cluster. The system then verifies and confirms that the cluster with the highest frequency has a proportion of session sequences within it that is higher than 80% of the total number of session sequences in the initial learning period, and that cluster is confirmed as the basic data source for building a high-frequency behavior logic template. Those isolated sequences that cannot be classified into this high-frequency cluster are identified as low-frequency or irregular behaviors. The third stage is template structuring and solidification. Based on the identified high-frequency behavior cluster, the system constructs a probabilistic directed acyclic graph as the final high-frequency behavior logic template. Each node represents a unique device state jump event, and the directed edge connecting two nodes represents a universally observed behavior step with a stable time sequence association. Each edge from node to node is assigned a set of accurate statistical properties, including the conditional probability , the average time interval , and the time interval standard deviation This structured template, which constitutes the set, not only defines the behavioral steps, but also precisely defines the expected time relationships and regularities between the steps in statistical language.

[0043] After completing the learning phase and entering the continuous monitoring phase, the logic self-consistency verification module uses the previously generated probabilistic directed acyclic graph as a benchmark to verify the real-time state transition sequence. When an event matching the template's starting node occurs, the system makes a probabilistic prediction of the next expected event's occurrence time based on the corresponding edges in the graph. At this point, the determination of behavioral logic deviation no longer relies on a globally fixed logical tolerance threshold, but is accomplished through a dynamic, statistically significant verification logic. If we consider preceding events... actual occurrence time up to the current event actual occurrence time The actual time interval that has elapsed The average time interval between it and the records in the template The deviation exceeded a threshold defined by the standard deviation. If the dynamic boundary is determined, a behavioral logic deviation is identified. This determination logic is described by the following inequality: In this inequality, It is a dimensionless sensitivity coefficient, a definite value used to define the tolerable range of fluctuation equivalent to several standard deviations. This is an engineering parameter that can be pre-configured by the system administrator according to the required management precision. This value can be either 2 or 3, which correspond to approximately 95% and 99.7% confidence intervals in statistics, respectively. When the water heater is detected to be off, the exhaust fan does not operate as expected by the template. Once activated within the time window, the system generates a structured risk warning signal pointing to a lack of ventilation logic based on the statistical fact that it deviates from the learned high-credibility behavior pattern. This method of binding parameter settings with the inherent statistical distribution characteristics of the data improves the sensitivity and robustness of risk identification.

[0044] Embodiment 5: In the scenario of care for the elderly in nursing homes, this embodiment takes a rental apartment containing both high-security risk areas and regular living areas as the background, in which the monitoring units associated with the kitchen gas valve intelligent switch and the carbon monoxide sensor are defined as the kitchen safety group, and the monitoring units associated with the living room electric kettle intelligent socket and the intelligent window curtain controller are defined as the living room convenience group. Since there are differences in the management goals of the two groups, the former requires the highest identification sensitivity to any potential risks, and the latter focuses on avoiding nuisance false alarms due to normal fluctuations in user habits, therefore, a systematic setting procedure is adopted for the setting of multiple core parameters in the system, rather than using globally uniform fixed values.

[0045] First, the specific value of the frequency threshold used to identify high-frequency behaviors in the behavior logic template sedimentation module is determined as follows: after the initial learning period ends, the system independently performs a threshold optimization procedure for each device group. Taking the living room convenience group as an example, the system manager first sets a relatively high initial frequency threshold, for example, 95%, and then the system analyzes all conversation sequences during the learning period and presents non-high-frequency sequence patterns excluded due to the current threshold. Based on the manager's understanding of the expected functions and operation modes of the device group, the manager determines whether the excluded sequences contain key behaviors that should be identified as templates. If the current threshold is too high, resulting in insufficient template coverage, the manager reduces the frequency threshold of the group by a preset step, for example, 5%, and repeats the analysis and judgment process. This iteration continues until the manager confirms that the current threshold can ensure that all key and regular behavior patterns are included in the template while effectively avoiding extremely low-frequency accidental operations. Through this procedure, different device groups can obtain customized thresholds that meet their respective management objectives. Second, the judgment standard used to filter signal noise in the state credibility assessment module, i.e., the percentage of the actual time interval of state jumps to the historical average stable running time, is also determined based on empirical optimization logic. The system provides analysis tools to generate a distribution histogram of all state jump intervals for a specific monitoring unit during the learning period. The manager can observe the main distribution of time intervals formed by normal operations and the transient jump distribution that does not represent stable state transitions of devices caused by circuit transient disturbances or invalid physical operations on a very short time scale. The goal of this procedure is to set a proper percentage that, when converted into an absolute time threshold, can completely cover and filter the transient jump distribution area.Similarly, the preset duration that the module needs to last after entering the delay confirmation mode is set according to the signal transmission and stability characteristics of the network in which the physical device is located, and the duration is set to be slightly greater than the maximum signal jitter or bounce time observed by the system in the deployment environment to ensure that the system only responds to the fact that the device enters a stable new state, rather than misjudging transient processes; Furthermore, after the initial learning period is completed and the system automatically generates a set of candidate high-frequency behavior logic templates, a mandatory template review and confirmation link involving the manager is set, and the system will not automatically apply any learning results directly to risk verification. As an alternative, the system will present each candidate template to the manager in a human-readable natural language format through the interface of the remote management server, for example, a candidate template will be described as kitchen safety group template one: detection of gas valve opening, small stable increase in carbon monoxide sensor reading within the next thirty seconds to three minutes, this behavior pattern appears with a frequency of ninety-two percent in the learning period, the manager must review each candidate template and make an explicit approval or rejection operation, only the approved template will be moved to the formal behavior logic template library of the system for subsequent real-time logic self-consistency verification, if the manager determines that a learned template itself represents a behavior that does not conform to safety specifications, it can be discarded through the rejection operation, thereby avoiding the system from solidifying the dangerous behavior sequence as a benchmark for generating risk warnings.

[0046] In the specific implementation of using a probabilistic directed acyclic graph as a template, the conditional probability attribute contained in the directed edge connecting two event nodes in the graph is calculated as follows: in all the data in the learning period, the actual number of times that the transition from the previous event node to the current event node is successfully counted, and then the number is divided by the total number of times that the previous event node occurs. This quantitative probability value, together with the average time interval of the transition and its standard deviation, jointly constitutes a complete mathematical characterization of the stability of the behavior step, providing a rigorous data basis for subsequent dynamic risk judgment logic based on statistical significance.

[0047] It is obvious to those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application.

[0048] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting, although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A long-term rental housing monitoring and management system, characterized in that, The system includes: The device group definition module is configured to logically define at least two monitoring units deployed in the same physical space and having an inherent relationship in behavior as a device group, wherein the physical space includes a kitchen or bathroom in an apartment, and the monitoring units include smart sockets, smart door magnetic sensors or smart water flow switches; The state transition recording module is configured to record the unique identification code of the monitoring unit and the precise timestamp of the state transition when the operating state of the monitoring unit transitions from one state to another. The behavior logic template accumulation module is configured to automatically learn and generate one or more high-frequency behavior logic templates that represent the regular living habits of tenants based on the precise timestamp sequence recorded by the device group within the preset initial learning period through an unsupervised learning algorithm. The high-frequency behavior logic templates represent the stable temporal correlation between the state jumps of each monitoring unit in the device group. The logic self-consistency verification module is configured to continuously compare the precise timestamp sequence output in real time by the state transition recording module with the high-frequency behavioral logic template representing the tenant's regular living habits. When there is a behavioral logic deviation between the real-time timestamp sequence and the high-frequency behavioral logic template, a risk warning signal related to the tenant's potential safety risks or abnormal living conditions is generated. The determination of the behavioral logic deviation is based on the following logic: when the absolute time difference between the actual state transition time of a certain monitoring unit and the expected state transition time indicated by the high-frequency behavioral logic template exceeds the preset logical tolerance deviation threshold of the template, it is determined that there is a behavioral logic deviation that is inconsistent with the regular living habits that the tenant has learned.

2. The long-term rental housing monitoring and management system according to claim 1, characterized in that, The state transition recording module records two binary states: on and off. The behavior logic template accumulation module generates high-frequency behavior logic templates, including typical timing sequences between state transitions of different monitoring units within the group and the time interval range corresponding to each typical timing sequence.

3. The long-term rental housing monitoring and management system according to claim 1, characterized in that, The system also includes a status reliability assessment module, configured to dynamically calculate and update the historical average stable runtime of each monitoring unit in the device group based on the precise timestamps recorded by the status transition recording module. Before the logic self-consistency verification module performs the comparison, when the state transition recording module records a state transition of a certain monitoring unit, the state credibility assessment module calculates the actual time interval between the current state transition and the previous state transition of the monitoring unit. If the actual time interval is less than 10% of the historical average stable operating time of the monitoring unit, the state credibility assessment module enters the delayed confirmation mode. Only when the state of the monitoring unit has not recovered to the state before the transition after the delayed confirmation mode has been in effect for a preset duration, does the state credibility assessment module recognize the validity of the state transition, and submits the timestamp of the state transition to the behavior logic template accumulation module for template generation within the initial learning cycle. After the initial learning cycle ends, the timestamp of the state transition is submitted to the logic self-consistency verification module for real-time comparison.

4. The long-term rental housing monitoring and management system according to claim 1, characterized in that, The system also includes a priority label configuration module, which configures at least two monitoring units belonging to at least two different device groups and consuming public resources to be configured with different priority digital labels.

5. The long-term rental housing monitoring and management system according to claim 4, characterized in that, The system also includes a resource contention logic arbitration module, configured to: before the state of a monitoring unit configured with a low-priority digital tag is about to switch from a non-running state to a running state, determine whether there is a monitoring unit configured with a high-priority digital tag that is currently in a running state based on the latest accurate timestamp of other monitoring units recorded by the state transition recording module; if so, the resource contention logic arbitration module delays the state switch of the low-priority unit until the high-priority monitoring unit returns to a non-running state.

6. The long-term rental housing monitoring and management system according to claim 5, characterized in that, Public resources include the apartment’s total electricity load or the apartment’s total water pressure.

7. The long-term rental housing monitoring and management system according to claim 1, characterized in that, The system also includes a warning signal sending module, configured to send the risk warning signal generated by the logic self-consistency verification module to the remote management server via a wireless communication network.

8. The long-term rental housing monitoring and management system according to claim 1, characterized in that, During the initial learning cycle, the behavior logic template accumulation module counts the frequency of state transition sequences of each monitoring unit in the device group and identifies sequences with a frequency higher than 80% as high-frequency behavior logic templates.

9. A method for early warning of risks in long-term rental housing, characterized in that, The method includes the following steps: Step a: Logically define at least two monitoring units deployed in the same physical space as a device group; Step b: When the operating state of the monitoring unit changes from one state to another, record the unique identification code of the monitoring unit and the precise timestamp when the state change occurs. Step c: Based on the precise timestamp sequence recorded by the device group within the preset initial learning period, one or more high-frequency behavior logic templates are automatically generated through an unsupervised learning algorithm. The high-frequency behavior logic templates represent the stable temporal correlation between the state transitions of each monitoring unit in the device group. Step d involves continuously comparing the real-time output precise timestamp sequence with the high-frequency behavior logic template, and generating a risk warning signal when there is a behavior logic deviation between the real-time timestamp sequence and the high-frequency behavior logic template. The determination of the behavior logic deviation is based on the following logic: when the absolute time difference between the actual state transition time of a monitoring unit and the expected state transition time indicated by the high-frequency behavior logic template exceeds the preset logic tolerance threshold of the template, it is determined that there is a behavior logic deviation.

Images