Intelligent light controller operation data analysis method and system

Abstract

The present application relates to the technical field of light controller operation data analysis, and particularly relates to a kind of intelligent light controller operation data analysis method and system, the method includes the following steps: identifying deviation event;According to external control instruction determines the area occupancy state and the expected execution state of luminaire implied by external control instruction;According to local perception data, generate local occupancy state detection result, and compare the area occupancy state implied by external control instruction with the area occupancy state represented by local occupancy state detection result, to obtain consistent or conflicting comparison result;According to operation state data, determine the actual operation state of luminaire;Evaluate the matching degree between actual operation state and expected execution state;Determine whether deviation event is caused by inaccurate external control instruction or luminaire internal component failure.The present application can improve the accuracy of fault diagnosis, thereby improving operation efficiency and system reliability.

Description

Technical Field

[0001] This invention relates to the technical field of lighting controller operation data analysis, specifically to a method and system for intelligent lighting controller operation data analysis. Background Technology

[0002] In modern large-scale buildings and parks, intelligent lighting systems are widely used to optimize energy use and provide a comfortable lighting environment. These systems typically collect a large amount of operational data and use this data for fault prediction and energy efficiency management. However, with the deepening of the smart park concept, intelligent lighting systems often need to be deeply integrated with other building management systems, especially space utilization monitoring systems, to achieve more macro-level control and resource optimization. In this integrated environment, intelligent lighting systems are given the responsibility of prioritizing responses to external system commands. However, in actual operation, the sensors relied upon by external space utilization monitoring systems, such as passive infrared sensors, are easily interfered with by environmental factors (such as airflow and temperature fluctuations generated by HVAC systems), resulting in intermittent or regional inaccuracies in the output of personnel occupancy data. Summary of the Invention

[0003] The purpose of this invention is to address the aforementioned shortcomings by proposing a method and system for analyzing the operating data of an intelligent lighting controller.

[0004] The present invention adopts the following technical solution: A method for analyzing operational data of an intelligent lighting controller, the method comprising the following steps: Acquire the operating status data of the lighting fixtures, and identify deviation events when the operating status data deviates from the preset range; In response to a deviation event, acquire the external control commands received by the luminaire at the time the deviation event occurs, as well as the local sensing data of the area where the luminaire is located at the time the deviation event occurs; Determine the area occupancy status and expected execution status of the lighting fixtures implied by the external control commands based on the external control commands; The local occupancy status detection result is generated based on the local sensing data, and the area occupancy status implied by the external control command is compared with the area occupancy status represented by the local occupancy status detection result to obtain consistent or conflicting comparison results. Determine the actual operating status of the lighting fixtures based on the operating status data; Assess the degree of match between the actual operating status and the expected execution status; Based on the comparison results and the degree of matching, it can be determined whether the deviation event is caused by inaccurate external control commands or by a malfunction of internal components of the lighting fixture.

[0005] This technical solution can effectively distinguish between "false anomalies" and "true anomalies" that occur during the operation of intelligent lighting systems, avoid false alarms, improve the accuracy of fault diagnosis, and thus improve operation and maintenance efficiency and system reliability.

[0006] This invention also discloses an intelligent lighting controller operation data analysis system, which applies the above-mentioned intelligent lighting controller operation data analysis method. The system includes: The acquisition module acquires the operating status data of the lighting fixtures and identifies deviation events when the operating status data deviates from a preset range. The response module, in response to a deviation event, acquires the external control commands received by the luminaire at the time the deviation event occurs, as well as the local sensing data of the area where the luminaire is located at the time the deviation event occurs; The expected state determination module determines the area occupancy status and the expected execution status of the lighting fixtures implied by the external control commands based on the external control commands. The comparison module generates local occupancy status detection results based on local sensing data, and compares the area occupancy status implied by external control commands with the area occupancy status represented by the local occupancy status detection results to obtain consistent or conflicting comparison results. The actual status determination module determines the actual operating status of the lighting fixtures based on the operating status data. The evaluation module assesses the degree of matching between the actual operating state and the expected execution state. The discrimination module determines, based on the comparison results and the degree of matching, whether the deviation event is caused by inaccurate external control commands or by a malfunction of internal components of the lighting fixture.

[0007] This technical solution provides a system for analyzing the operational data of intelligent lighting controllers. Through modular design, it clearly defines functions such as data acquisition, command response, status determination, comparison, evaluation, and discrimination, making the system structure clear, easy to implement and maintain, and thus effectively supporting the fault diagnosis and operation and maintenance management of intelligent lighting systems.

[0008] This invention significantly improves the accuracy of fault diagnosis and the efficiency of operation and maintenance of intelligent lighting systems, and has significant progressive and practical value.

[0009] To further understand the features and technical content of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are for reference and illustration only and are not intended to limit the present invention. Attached Figure Description

[0010] Figure 1 This is a flowchart of the intelligent lighting controller operation data analysis method of the present invention; Figure 2This is a schematic diagram of the intelligent lighting controller operation data analysis system of the present invention. Detailed Implementation

[0011] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can understand the advantages and effects of the present invention from the content disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the spirit of the present invention. Furthermore, the accompanying drawings of the present invention are for simple illustrative purposes only and are not depictions of actual dimensions; this is stated in advance. The following embodiments will further describe the relevant technical content of the present invention in detail, but the disclosed content is not intended to limit the scope of protection of the present invention.

[0012] This embodiment provides a method and system for analyzing the operational data of an intelligent lighting controller, combined with... Figure 1 and Figure 2 As shown.

[0013] refer to Figure 1 A method for analyzing operational data of an intelligent lighting controller, comprising the following steps: Acquire the operating status data of the lighting fixtures, and identify deviation events when the operating status data deviates from the preset range; In response to a deviation event, acquire the external control commands received by the luminaire at the time the deviation event occurs, as well as the local sensing data of the area where the luminaire is located at the time the deviation event occurs; Determine the area occupancy status and expected execution status of the lighting fixtures implied by the external control commands based on the external control commands; The local occupancy status detection result is generated based on the local sensing data, and the area occupancy status implied by the external control command is compared with the area occupancy status represented by the local occupancy status detection result to obtain consistent or conflicting comparison results. Determine the actual operating status of the lighting fixtures based on the operating status data; Assess the degree of match between the actual operating status and the expected execution status; Based on the comparison results and the degree of matching, it can be determined whether the deviation event is caused by inaccurate external control commands or by a malfunction of internal components of the lighting fixture.

[0014] This invention aims to provide a method for analyzing the operational data of an intelligent lighting controller, effectively distinguishing between "false anomalies" caused by inaccurate external control commands and "true anomalies" caused by malfunctions in internal components of the lighting fixture. By comprehensively analyzing the operating status data of the lighting fixture, external control commands, and local sensing data, this method can more accurately diagnose the root cause of problems, thereby improving the operational efficiency and reliability of intelligent lighting systems.

[0015] "Lamp operating status data" refers to various data generated by the intelligent lamps during operation, such as power consumption, brightness, on / off status, and color temperature. This data reflects the actual working condition of the lamps. "Deviation event" refers to a situation where the lamp's operating status data significantly differs from the preset normal operating range, such as abnormally high or low power consumption or unstable brightness output. "External control commands" refer to commands sent to the intelligent lighting controller by external systems (such as space utilization monitoring systems). These commands typically include instructions for switching lamps on and off, adjusting brightness, etc., and may implicitly include judgments about area occupancy status. "Local sensing data" refers to data collected by sensors on the lamp itself or in its surrounding area, such as passive infrared sensor data, ambient light sensor data, and temperature sensor data. This data reflects local environmental information and human activity around the lamps. "Area occupancy status" refers to whether there is human occupancy in a specific area, usually inferred from external control commands or local sensing data. "Expected execution status" refers to the theoretically expected operating state of the lamp after receiving external control commands. "Actual operating status" refers to the actual operating state of the lamp at the moment a deviation event occurs. "Comparison result" refers to the conclusion drawn after comparing the area occupancy status implied by the external control command with the area occupancy status represented by the local occupancy status detection result; this may result in agreement or conflict. "Matching degree" refers to the degree of conformity between the actual operating state of the lighting fixture and the expected execution state.

[0016] This method is typically deployed on a smart lighting controller or a connected edge computing device, or it can be implemented on a cloud server. The implementation environment includes various sensors (such as passive infrared sensors, ambient light sensors, temperature sensors, etc.), a smart lighting controller, and communication interfaces with external systems (such as building management systems, space utilization monitoring systems).

[0017] The core of the intelligent lighting controller operation data analysis method proposed in this invention lies in achieving accurate attribution of deviation events through multi-dimensional data analysis.

[0018] First, when acquiring the operating status data of the lighting fixtures, and identifying deviation events when the data deviates from a preset range, various methods can be employed. For example, one can continuously monitor parameters such as real-time power consumption, brightness output, and switching frequency of the lighting fixtures, comparing them with historical baseline data or preset thresholds. When any parameter exceeds the preset normal fluctuation range, it is identified as a deviation event. Another approach is to utilize machine learning models to detect anomalies in the lighting fixtures' operating status data. By training the model to learn normal operating patterns, when new data points significantly deviate from the normal pattern, they are identified as deviation events.

[0019] Secondly, in the step of responding to a deviation event and acquiring the external control commands received by the luminaire at the time of the deviation event, as well as the local sensing data of the area where the luminaire is located at the time of the deviation event, it can be implemented as follows: Once a deviation event is identified, the system immediately queries the external control command log at the time of the event to obtain the specific command content received by the luminaire at that time. Simultaneously, the system collects local sensing data corresponding to that moment from the sensors built into the luminaire or from the local sensor network deployed in the area, such as passive infrared sensor data, ambient light data, and temperature data.

[0020] Furthermore, in determining the area occupancy status and the expected execution state of the lighting fixtures implied by the external control commands, the process can be implemented as follows: External control commands typically involve direct operations on the lighting fixtures, such as "turn on" or "adjust brightness to 50%." These commands often implicitly contain the external system's judgment on the area occupancy status. For example, if the command is "turn on the lighting fixtures," it may imply that the external system believes the area is occupied. The system will parse these commands, extract their implicit area occupancy status (e.g., "occupied" or "unoccupied"), and determine the expected execution state that the lighting fixtures should achieve under normal circumstances (e.g., "lighting fixtures on, brightness 50%)" based on the command content.

[0021] Next, in the step of generating a local occupancy status detection result based on local sensing data and comparing the area occupancy status implied by the external control command with the area occupancy status represented by the local occupancy status detection result to obtain a consistent or conflicting comparison result, it can be implemented as follows: Local sensing data, such as data from a passive infrared sensor, can be used to independently determine the occupancy status of an area. For example, when a passive infrared sensor continuously detects human movement signals over a period of time, it can generate a local occupancy status detection result indicating "occupied". Then, this local detection result is compared with the area occupancy status implied by the external control command. If the two determinations are consistent (e.g., both indicate "occupied"), the comparison result is "consistent"; if the two determinations are inconsistent (e.g., the external command indicates "occupied", while the local detection result indicates "unoccupied"), the comparison result is "conflicting".

[0022] Then, in the step of determining the actual operating status of the luminaire based on the operating status data, it can be implemented as follows: The system analyzes the luminaire's operating status data at the time of the deviation event. For example, if the deviation event is an abnormal increase in power consumption, then the actual operating status may be determined as "excessive power consumption." If the deviation event is a lower-than-expected brightness output, then the actual operating status may be determined as "insufficient brightness."

[0023] Subsequently, in the step of assessing the degree of matching between the actual operating state and the expected execution state, this can be achieved by comparing the actual operating state of the luminaire with the previously determined expected execution state. For example, if the expected execution state is "luminaire on, brightness 50%", while the actual operating state is "luminaire on, brightness 30%", then the degree of matching is low. The degree of matching can be expressed by quantitative indicators, such as the percentage difference in brightness, the absolute value of the power consumption deviation, etc.

[0024] Finally, in the step of determining whether the deviation event is caused by inaccurate external control commands or by a malfunction of internal components of the luminaire based on the comparison results and the degree of matching, it can be implemented as follows: If the comparison result is "conflicting" (i.e., the external command and local sensing data are inconsistent in their judgment of the area occupancy status), and the degree of matching between the actual operating state and the expected execution state is low, it can be preliminarily determined that the deviation event is caused by inaccurate external control commands. For example, the external system incorrectly judges that the area is occupied and sends an on command, but the area is actually empty, resulting in abnormal power consumption after the luminaire is turned on. Conversely, if the comparison result is "consistent" (i.e., the external command and local sensing data are consistent in their judgment of the area occupancy status), but the degree of matching between the actual operating state and the expected execution state is low, it can be preliminarily determined that the deviation event is caused by a malfunction of internal components of the luminaire. For example, both the external system and the local sensing determine that the area is occupied and send an on command, but the brightness is still insufficient after the luminaire is turned on, which may indicate a malfunction in the driver or LED module inside the luminaire.

[0025] The steps of generating local occupancy status detection results based on local sensing data, and comparing the area occupancy status implied by external control commands with the area occupancy status represented by the local occupancy status detection results to obtain consistent or conflicting comparison results include: The local sensing data is time-stamp aligned and denoised to obtain the processed local sensing data. The local sensing data is fused and processed to generate a local occupancy status judgment result; The confidence level of the local occupancy status judgment result is evaluated to obtain the confidence level evaluation result. The confidence level evaluation considers the consistency between local sensing data, the integrity and real-time performance of local sensing data, the interference intensity of local environmental disturbances on local sensing sensors, and the adaptive calibration information of local sensing sensors. Based on the confidence assessment results, output the local occupancy status probability value and uncertainty value, which represent the local occupancy status judgment results, and generate the local occupancy status detection results based on the local occupancy status probability value and uncertainty value; The local occupancy status is determined based on the local occupancy status detection results, and the local occupancy status represented by the local occupancy status detection results is compared with the local occupancy status implied by the external control commands to obtain consistent or conflicting comparison results. The local sensing data includes local passive infrared sensor data, internal temperature fluctuation data of the lamps, and surrounding wireless signal characteristic data.

[0026] Specifically, locally sensed data refers to raw data acquired from sensors in the area where the lighting fixtures are located, such as data from local passive infrared sensors, internal temperature fluctuation data of the lighting fixtures, and surrounding wireless signal characteristics. This data reflects the actual occupancy status within the area. Timestamp alignment and denoising of locally sensed data involves synchronizing local sensed data from different sensor sources in time, ensuring that all data points correspond to the same moment, and removing random noise, outliers, or interference signals to improve the accuracy and reliability of the data. For example, algorithms such as moving averages and Kalman filtering can be used for denoising.

[0027] Furthermore, generating a local occupancy status judgment result by fusing the processed local sensing data refers to comprehensively analyzing and processing various local sensing data after alignment and denoising to arrive at a preliminary judgment on whether there is personnel occupancy within the area. This may involve multi-sensor data fusion algorithms, such as Bayesian networks, support vector machines, or deep learning models, to improve the accuracy of the judgment by comprehensively considering the characteristics of different types of data.

[0028] The confidence level assessment of local occupancy status judgment results refers to the quantitative evaluation of the reliability of the generated local occupancy status judgment results. Confidence level assessment is a crucial step, comprehensively considering multiple influencing factors, including: the consistency between local sensing data, referring to whether data from different sensors corroborate each other in indicating the occupancy status of the area; for example, if an infrared sensor detects movement while the wireless signal strength also changes, the consistency is high; the completeness and real-time performance of local sensing data, referring to whether the data stream is continuous and without gaps, and whether the data can promptly reflect the current environmental state; the interference intensity of local environmental disturbances on local sensing sensors, referring to the degree of influence of non-occupancy factors in the environment (such as airflow, temperature changes, electromagnetic interference, etc.) on sensor readings; and the adaptive calibration information of local sensing sensors, referring to the calibration or correction information performed by the sensors themselves during operation, which can reflect the health status of the sensors and the accuracy of the data. By comprehensively considering these factors, a quantitative confidence level assessment result can be obtained.

[0029] Therefore, based on the confidence assessment results, outputting the local occupancy status probability value and uncertainty value, which characterize the local occupancy status judgment result, means outputting the local occupancy status judgment result in the form of probability values ​​and uncertainty values. The probability value represents the likelihood that the area is occupied, while the uncertainty value reflects the degree of ambiguity or potential error in the judgment. For example, a high probability value accompanied by a low uncertainty value indicates a high degree of confidence in the occupancy status judgment. Generating the local occupancy status detection result based on the local occupancy status probability value and uncertainty value means combining the local occupancy status probability value and uncertainty value to form a final, more reliable local occupancy status detection result.

[0030] Specifically, determining the area occupancy status represented by the local occupancy status detection results means clarifying whether the area is in an "occupied" or "idle" state based on the generated local occupancy status detection results. Subsequently, comparing the area occupancy status represented by the local occupancy status detection results with the area occupancy status implied in external control commands involves comparing the actual occupancy status obtained from local sensing data with the preset or implied occupancy status in the external control commands to obtain consistent or conflicting comparison results. The comparison result may be consistent (e.g., the command requests the light to turn on, and the local system detects someone) or conflicting (e.g., the command requests the light to turn off, and the local system detects someone).

[0031] The present invention preprocesses raw local sensing data to ensure data quality and reliability. Furthermore, by fusing data from multiple sensors, area occupancy information can be obtained from different dimensions, thereby improving the accuracy of occupancy status judgment. Based on this, a confidence assessment mechanism is introduced, comprehensively considering data consistency, completeness, real-time performance, environmental interference, and sensor self-calibration information to quantify the reliability of the occupancy judgment results. This ensures that subsequent occupancy status detection results include not only occupancy probability but also uncertainty, thus more comprehensively reflecting the actual situation. Finally, comparing this refined and evaluated local occupancy status detection result with the area occupancy status implied by external control commands can more accurately identify whether there is consistency or conflict between the two, providing a solid data foundation for subsequently determining the root cause of deviation events (inaccurate external control commands or internal component failure of the lighting fixture).

[0032] The above technical solutions significantly improve the accuracy and robustness of local occupancy detection. Specifically, by fusing and refining multi-source local sensing data, combined with a confidence assessment mechanism, the risk of false alarms or missed alarms from a single sensor can be effectively reduced, and the reliability of the judgment results can be quantified. Therefore, when comparing local occupancy status with external control commands, more accurate and reliable consistent or conflicting comparison results can be obtained. This provides a more accurate basis for determining the cause of deviation events in the intelligent lighting controller's operational data analysis methods, avoiding misjudgments caused by inaccurate occupancy status assessments, and ultimately improving the efficiency and accuracy of fault diagnosis for the entire system.

[0033] The present invention further proposes a step for generating a local occupancy status judgment result from the fused local sensing data, including: Analyze the trigger frequency and signal strength changes of local passive infrared sensor data within a preset first time window to determine whether local passive infrared sensor data generates an indication of occupancy. The average value and standard deviation of the internal temperature fluctuation data of the lamps within a preset second time window, as well as the average intensity and fluctuation amplitude of the surrounding wireless signal characteristic data within a preset second time window, are analyzed to determine whether the internal temperature fluctuation data of the lamps and the surrounding wireless signal characteristic data together form a stable pattern indication that is consistent with the physical characteristics of stationary personnel. The system performs a fusion judgment based on the occupied status indication and the stable mode indication to generate a local occupancy status judgment result.

[0034] Specifically, local passive infrared sensor data is typically used to detect human movement within an area. By monitoring changes in trigger frequency and signal strength within a preset first time window, the presence of moving individuals can be effectively determined, thus generating an indication of occupancy. For example, when a passive infrared sensor triggers multiple times within a short period and the signal strength fluctuates significantly, it usually indicates the presence of active individuals in the area.

[0035] Among these, data on internal temperature fluctuations of the luminaire and the characteristics of surrounding wireless signals can provide clues about the presence of stationary individuals. Analyzing the average and standard deviation of the internal temperature fluctuation data within a preset second time window can identify subtle but continuous temperature changes caused by human body heat. Simultaneously, analyzing the average intensity and fluctuation amplitude of the surrounding wireless signal characteristics data can detect the absorption, reflection, or scattering effects of wireless signal propagation by the human body. These effects exhibit specific stable patterns in the presence of stationary individuals. When both sets of data jointly show a stable pattern consistent with the physical characteristics of a stationary individual, it can be determined that a stable pattern indicating the presence of such a person is present.

[0036] In practical applications, occupancy indicators and stable pattern indicators can be considered as two different dimensions or levels of confidence in evidence of occupancy. By fusing these two indicators—for example, using logical AND, logical OR, weighted average, or machine learning-based classification algorithms—the true occupancy status within a region can be comprehensively assessed, resulting in a more accurate and robust local occupancy assessment. The first and second time windows can be flexibly configured according to the actual application scenario and sensor characteristics. For example, the first time window can be shorter to quickly respond to movement, while the second time window can be longer to capture stable stationary patterns.

[0037] This invention effectively solves the problem of inaccurate judgment in complex occupancy scenarios by layering and collaboratively analyzing different types of local sensing data. Specifically, local passive infrared sensor data excels at capturing human movement, but has limited ability to detect stationary individuals. Data on internal temperature fluctuations of lighting fixtures and surrounding wireless signal characteristics can compensate for this deficiency, as they can sensitively reflect subtle environmental changes caused by the presence of stationary individuals. By first extracting occupancy indicators from both movement and stillness dimensions, and then fusing these indicators, this solution can more comprehensively and accurately depict the true occupancy status within the area. This step-by-step fusion strategy enables the system to make more reliable judgments when facing various complex situations such as movement, stillness, or brief absence of personnel, thereby improving the accuracy and robustness of local occupancy status judgment results.

[0038] In some preferred embodiments, a specific example is given below. Assume a smart lighting controller is operating within an office area. When an employee enters the area, local passive infrared sensors detect multiple triggers and significant signal strength changes within a preset first time window (e.g., 5 seconds), generating an "occupied" indication. Subsequently, the employee remains stationary at their desk. At this time, the local passive infrared sensors may no longer trigger, but the system continues to analyze internal temperature fluctuations in the lighting fixtures and surrounding wireless signal characteristics. Within a preset second time window (e.g., 30 seconds), the average internal temperature fluctuation data may be slightly higher than the ambient background temperature, with a low standard deviation, while the average strength and fluctuation amplitude of the surrounding wireless signal characteristics exhibit a stable pattern consistent with human presence. The system generates a "stable pattern indication" accordingly. Finally, the system merges the "occupied" indication (which may have become invalid or weakened) with the "stable pattern indication" for a final judgment. For example, if the confidence level of the "stable pattern indication" is high, the system will still determine that the area is occupied even if the "occupied" indication has weakened. This integrated judgment mechanism ensures that the system can accurately identify the presence of people even if they remain stationary for a long time, avoiding misjudgments caused by people being stationary, thereby guaranteeing the accuracy of lighting control.

[0039] This invention further proposes a method for evaluating the confidence level of local occupancy status determination results. The steps for obtaining the confidence level evaluation result include: Identify and quantify the consistency among local sensing data, the integrity and real-time performance of local sensing data, the interference intensity of local environmental disturbances on local sensing sensors, and the impact of adaptive calibration information of local sensing sensors on the reliability of local occupancy status judgment results. Based on the degree of influence, the support value corresponding to each factor is determined. The factors include the consistency between local sensing data, the completeness and real-time performance of local sensing data, the interference intensity of local environmental disturbances on local sensing sensors, and the adaptive calibration information of local sensing sensors. The support value is a numerical value used to characterize the degree of reliability of the corresponding factor in the local occupancy status judgment result. According to the preset weight allocation rules, the support values ​​of each factor are weighted and fused to calculate the comprehensive confidence value. The corresponding weights are adjusted according to the historical performance of each factor in different scenarios and the current environmental conditions. Output the overall confidence score as the confidence assessment result.

[0040] Specifically, identifying and quantifying the consistency among local sensing data, the completeness and real-time performance of local sensing data, the interference intensity of local environmental disturbances on local sensing sensors, and the impact of adaptive calibration information of local sensing sensors on the reliability of local occupancy status judgment results refers to in-depth analysis and quantification of multiple key factors affecting the reliability of local occupancy status judgment results. For example, the consistency among local sensing data can be quantified by calculating the degree of consistency of different sensor data sources (such as local passive infrared sensor data, internal temperature fluctuation data of lamps, and surrounding wireless signal characteristic data) in indicating occupancy status, for example, by using correlation coefficients or voting mechanisms. The completeness and real-time performance of local sensing data can be assessed by monitoring the continuity, missing rate, and data latency of the data stream. The interference intensity of local environmental disturbances on local sensing sensors can be quantified by analyzing the impact of environmental noise (such as airflow, sudden temperature changes, and unexpected wireless signals) on sensor readings. The adaptive calibration information of local sensing sensors can reflect the calibration effect and stability of sensors under different environmental conditions. The quantification of these impact levels aims to provide an objective basis for subsequent confidence assessment.

[0041] The determination of support values ​​for each factor based on its degree of influence means assigning a numerical value to each influencing factor based on the aforementioned quantification results. This value characterizes the degree to which the factor supports the reliability of the local occupancy status judgment. For example, if the consistency among local sensing data is high, its corresponding support value will be high; if the intensity of local environmental disturbance is large, its corresponding support value will be low. These support values ​​can be normalized to a specific range (e.g., 0 to 1) to facilitate subsequent fusion calculations.

[0042] In practical applications, the support values ​​of each factor are weighted and fused according to a preset weighting rule to calculate the overall confidence score. This step aims to comprehensively consider the contribution of each factor to the confidence score. The preset weighting rule can be determined based on expert experience, historical data analysis, or machine learning models. For example, in some scenarios, the consistency of locally perceived data may be more critical, thus its weight will be higher; while in other scenarios, the impact of real-time performance or environmental interference may be more significant, and its weight will be adjusted accordingly. It is worth noting that the corresponding weights can be dynamically adjusted based on the historical performance of each factor in different scenarios and current environmental conditions to adapt to environmental changes and improve the accuracy and robustness of the confidence score assessment.

[0043] Therefore, a comprehensive confidence score is output as the confidence assessment result. This comprehensive confidence score is a single, quantitative indicator that comprehensively reflects the reliability of the local occupancy status judgment result. This result can be used for subsequent decision-making, such as triggering further verification mechanisms or adjusting lighting control strategies when the confidence score is low.

[0044] This invention overcomes the qualitative or non-systematic problems that may exist in traditional confidence assessment by identifying and quantifying multiple key factors affecting the reliability of local occupancy status judgment results. Specifically, by quantifying the consistency between local sensing data, the integrity and real-time nature of the data, the interference intensity of local environmental disturbances on sensors, and the adaptive calibration information of sensors, this solution can obtain the specific degree of influence of each factor on the reliability of the judgment result. Furthermore, these degrees of influence are converted into corresponding support values, providing standardized input for subsequent comprehensive evaluation. Finally, these support values ​​are weighted and fused through preset weight allocation rules, and the weights are dynamically adjusted according to historical performance in different scenarios and current environmental conditions, so that the comprehensive confidence value can more accurately and comprehensively reflect the true reliability of the local occupancy status judgment result. This systematic quantification and dynamic weighted fusion mechanism effectively avoids the limitations of single-factor evaluation, improves the robustness and adaptability of confidence assessment, and thus provides a more reliable basis for subsequent deviation event discrimination.

[0045] Through the above technical solution, this invention provides a more accurate and reliable method for assessing the confidence level of local occupancy status judgment results. Compared to assessment methods that only consider various factors, this solution significantly improves the objectivity and accuracy of confidence level assessment by quantifying the influence of each factor, determining the support value, and dynamically weighting and fusing the results. Therefore, during the analysis of intelligent lighting controller operation data, once the confidence level of the local occupancy status judgment result is accurately assessed, the system can more accurately determine whether external control commands are accurate or whether there is a fault in the internal components of the lighting fixture, thereby effectively reducing misjudgments and improving the efficiency and accuracy of fault diagnosis. Furthermore, the dynamic weight adjustment mechanism allows this assessment method to better adapt to complex and changing environments, further enhancing the robustness and practicality of the system.

[0046] In some preferred embodiments, it is assumed that the intelligent lighting controller is located in an open-plan office area. When the system needs to evaluate the confidence level of the local occupancy status judgment, it performs the following steps: First, the system identifies and quantifies various influencing factors. For example, the consistency between local sensing data can be quantified by analyzing the voting results of local passive infrared sensor data, internal temperature fluctuation data of the lighting fixtures, and surrounding wireless signal characteristic data on the "occupied" or "unoccupied" status indication over the past minute. If all three indicate "occupied," the consistency score is high. The integrity and real-time performance of local sensing data can be quantified by monitoring the packet loss rate and average transmission delay. The interference intensity of local environmental disturbances on local sensing sensors can be quantified by analyzing the impact of airflow fluctuations at the air conditioning vents in the area on passive infrared sensors, or the impact of changes in nearby Wi-Fi signal strength on wireless signal characteristic data. The adaptive calibration information of the local sensing sensors can be obtained based on the time of the sensor's most recent calibration and the error range in the calibration report. Second, based on these quantification results, the system determines the support value corresponding to each factor. For example, if data consistency reaches 90% or higher, its support value might be 0.95; if data integrity is good and real-time performance is high, its support value might be 0.90; if environmental disturbance intensity is low, its support value might be 0.85; if sensor calibration status is good, its support value might be 0.92. Finally, the system weights and fuses these support values ​​according to a preset weighting rule. For example, in an office area, data consistency might be assigned a weight of 0.4, data integrity and real-time performance 0.3, environmental disturbance 0.2, and adaptive calibration information 0.1. These weights can be adjusted based on historical data, such as performance in different time periods (working hours, non-working hours) or under different weather conditions. A comprehensive confidence score is calculated through weighted summation, for example, (0.95...). 0.4)+(0.90 0.3)+(0.85 0.2)+(0.92 0.1) = 0.38 + 0.27 + 0.17 + 0.092 = 0.912. This comprehensive confidence level of 0.912 is used as the final confidence level assessment result, indicating that the reliability of the current local occupancy status judgment result is relatively high.

[0047] This invention further proposes a step of weighted fusion of the support values ​​corresponding to each factor according to a preset weight allocation rule, including: Continuously monitor the environmental disturbance characteristics of the area where the lighting fixtures are located, and extract features from the environmental abnormal signals when they are identified as not conforming to the preset known interference patterns. The extracted features are compared with a pre-defined library of known interference features, and features that cannot be matched are marked as novel environmental interference. Analyze the impact of novel environmental interference on local sensing data output and generate a quantified novel interference impact factor; The support values ​​of the additional factors are determined based on the new interference factors, and the additional factors are included in each factor. The weights of the additional factors are determined according to the preset weight allocation rules, so as to perform weighted fusion of the support values ​​of each factor, including the additional factors.

[0048] Specifically, continuous monitoring of environmental disturbance characteristics in the area where the lighting fixtures are located refers to the real-time acquisition of environmental data streams through various sensors deployed in or around the lighting fixtures (e.g., acoustic sensors, electromagnetic field sensors, air quality sensors, etc.) to capture any abnormal fluctuations or patterns that may affect local sensing data (such as local passive infrared sensor data, internal temperature fluctuation data of the lighting fixtures, and surrounding wireless signal characteristic data). When an abnormal environmental signal that does not conform to a preset known interference pattern is identified, for example, when a pattern recognition algorithm or anomaly detection model finds that the characteristics of the currently collected environmental data are significantly different from or do not match pre-stored, known interference patterns (such as air conditioner start-up, people walking, door and window opening and closing, etc.), then an abnormal environmental signal is considered to have been detected.

[0049] Furthermore, feature extraction of abnormal environmental signals refers to using signal processing techniques (such as Fourier transform, wavelet analysis, and statistical feature extraction) to extract key information that characterizes the essential attributes of the abnormal signals, such as frequency components, energy distribution, duration, and intensity change rate. The extracted features are then compared with a pre-set library of known interference features, which stores typical feature patterns of various known environmental interferences (such as fan noise, induction cooker radiation, and large equipment startup). Through similarity matching and classifier discrimination, it is determined whether the features of the current abnormal signal match any known interference feature in the library. Features that cannot be matched are marked as novel environmental interference, meaning that this interference is something the system has not encountered or identified before and is therefore unknown.

[0050] Analyzing the impact of novel environmental interference on local sensing data output involves quantifying the degree of deviation, noise, or distortion caused by this interference to local sensing data, such as passive infrared sensor data, internal temperature fluctuation data of lighting fixtures, and surrounding wireless signal characteristic data, through experimental analysis, machine learning model prediction, or expert system evaluation. This generates a quantified novel interference impact factor, a numerical value characterizing the intensity of the negative impact of novel interference on the reliability of local sensing data. For example, the impact factor can be a value between 0 and 1, with a higher value indicating a more severe impact.

[0051] In practical applications, the support value of the additional factors is determined based on the influence factor of the novel interference. For example, a lower support value can be set when the influence factor is high, and vice versa, to reflect the degree to which the novel interference weakens the reliability of the local occupancy status judgment results. The additional factors are included in all factors, meaning that the novel environmental interference is considered an independent factor affecting the confidence assessment, on par with factors such as the consistency between local sensing data, the completeness of local sensing data, and the real-time nature of the local sensing data. The weights of the additional factors are determined according to a preset weighting rule. These weights can be dynamically adjusted based on the potential harm of the novel interference, its historical frequency of occurrence, and its impact characteristics on different types of local sensing data. Finally, the support values ​​of all factors, including the additional factors, are weighted and fused to calculate a more comprehensive and accurate overall confidence value.

[0052] The solution of this invention overcomes the aforementioned limitations by introducing a mechanism for identifying, quantifying, and dynamically incorporating novel environmental interferences into the confidence assessment. Traditional solutions, lacking the ability to perceive and quantify the impact of unknown interferences, may lead to distorted confidence assessment results. This solution, through continuous monitoring of environmental disturbance characteristics, can promptly detect and extract abnormal signal features that do not conform to known patterns, marking them as novel environmental interferences. It is precisely this quantification of the impact of novel interferences that enables the system to generate an additional factor support value reflecting their negative effects. By incorporating this additional factor and its corresponding weight into the original weighted fusion calculation, the comprehensive confidence value can more comprehensively reflect the impact of all known and unknown interferences on the reliability of the local occupancy status judgment results, thereby avoiding assessment bias caused by the failure to consider novel interferences.

[0053] In some preferred embodiments, a specific example is given below. Assume a smart office area where a lighting controller is operating and continuously monitoring local sensing data to determine area occupancy. One day, a new high-frequency wireless charging device is installed near the area, and its electromagnetic radiation causes persistent, previously unidentified interference to the surrounding wireless signal characteristics of the lighting fixture. First, the system continuously monitors the environmental disturbance characteristics of the area where the lighting fixture is located. When the high-frequency wireless charging device starts operating, the lighting fixture's electromagnetic field sensor (as part of the environmental disturbance characteristic monitoring) detects an abnormal electromagnetic signal pattern. This signal pattern is compared with a pre-set library of known interference characteristics (e.g., including Wi-Fi signals, Bluetooth signals, microwave oven radiation, etc.) and fails to match, thus being identified as a new type of environmental interference. Next, the system analyzes the impact of this new environmental interference on the local sensing data output. Through comparative analysis of historical and real-time data, it is found that the high-frequency electromagnetic radiation causes a significant decrease in the signal-to-noise ratio of the surrounding wireless signal characteristics, exhibiting periodic fluctuations, thereby affecting the accuracy of judging the vital signs of stationary individuals based on wireless signal characteristics. The system generates a quantified new interference impact factor, for example, set to 0.7 (indicating a significant impact). Then, based on this new interference impact factor, a support value corresponding to an additional factor is determined, for example, set to 0.3 (a low support value reflecting the negative impact of the interference). This additional factor is incorporated into the factors evaluating the confidence level of the local occupancy status judgment result, and a dynamically adjusted weight is assigned to this new interference additional factor according to a preset weighting rule; for example, a higher weight is given based on the intensity and duration of its impact factor. Finally, when calculating the overall confidence level value, the support values ​​of all factors, including the new interference additional factor, are weighted and fused. In this way, even in the face of previously unknown environmental interference, the system can dynamically adjust the confidence level assessment, ensuring that the final confidence level value accurately reflects the reliability of the local occupancy status judgment result, avoiding misjudgments caused by new interference, and thus ensuring the accuracy of the intelligent lighting controller's operational data analysis.

[0054] After outputting the overall confidence score as the confidence assessment result, the above method also includes the following steps: Monitor the overall confidence level and compare it with a preset low confidence threshold; Monitor the actual occupancy status of the area where the lighting fixtures are located; When the overall confidence level value is continuously lower than the low confidence level threshold within the preset monitoring period, and the actual occupancy status of the area where the lamp is located remains consistent within the preset monitoring period, the self-diagnosis process of confidence assessment is triggered. In the self-diagnostic process of confidence assessment, analyze the historical trends of each factor; Identify the key factors that cause the overall confidence score to remain below the low confidence threshold for an extended period of time based on the historical trends of each factor. Based on key factors, the parameters for determining the support values ​​of each factor and the preset weighting rules are adjusted. The support values ​​of each factor are re-determined based on the adjusted parameters, and the re-determined support values ​​of each factor are weighted and fused according to the adjusted preset weighting rules to recalculate the overall confidence score. The recalculated overall confidence score is then used as the updated confidence score assessment result. The parameters for determining the support values ​​of each factor are used to determine the degree of influence of each factor on the reliability of the local occupancy status judgment result.

[0055] Specifically, in the above scheme, the system is configured to continuously monitor the output comprehensive confidence score. This comprehensive confidence score is a quantitative indicator characterizing the reliability of the local occupancy status judgment result. Simultaneously, this score is compared in real time with a preset low confidence threshold to determine whether the current confidence level is low. The low confidence threshold can be set according to the needs of the actual application scenario and the system's reliability requirements; for example, it can be set to 0.6 or 0.7.

[0056] Simultaneously, the system independently monitors the actual occupancy status of the area where the lighting fixtures are located. Monitoring actual occupancy can be achieved in various ways, such as cross-validation using higher-precision independent sensors (e.g., millimeter-wave radar data, thermal imaging sensor data, or anonymized video analysis data), or through manual verification. The aim is to provide an independent and relatively reliable benchmark for verifying the accuracy of the confidence assessment results.

[0057] When the overall confidence level is detected to be consistently below the low confidence threshold for a preset monitoring period (e.g., 1 hour, 2 hours, or longer), and the actual occupancy status of the area where the light fixture is located remains consistent during this period (e.g., continuously occupied or continuously unoccupied), the system will be triggered to enter the self-diagnostic process for confidence assessment. This triggering condition is designed to ensure that the self-diagnostic process is only initiated when there may be systemic problems with the confidence assessment results, avoiding frequent triggering due to momentary fluctuations or accidental events.

[0058] In the self-diagnostic process of confidence assessment, the system conducts an in-depth analysis of the historical trends of various factors affecting the confidence assessment, including the consistency between local sensing data, the completeness and real-time performance of local sensing data, the interference intensity of local environmental disturbances on local sensing sensors, and the adaptive calibration information of local sensing sensors. For example, it can analyze the average value, fluctuation range, and abnormal peak values ​​of these factors over a period of time. Through the analysis of historical trends, the system can identify key factors that cause the overall confidence score to remain consistently low. For example, if the intensity of local environmental disturbances is found to be continuously increasing, it may indicate that this factor is the main reason for the decline in confidence score.

[0059] Once key factors are identified, the system adjusts the parameters for determining the support values ​​of each factor and the preset weighting rules based on these factors. The parameters quantify the impact of each factor on the reliability of the local occupancy status judgment; these can be, for example, the coefficients or thresholds of a function. The weighting rules determine the relative importance of each factor in the weighted fusion calculation of the overall confidence score. By adjusting these parameters and rules, the system can more accurately reflect the actual contribution of each factor to the confidence score under the current environmental and sensor conditions.

[0060] Based on the adjusted parameters, the system recalculates the support values ​​for each factor. Then, according to the adjusted preset weighting rules, these recalculated support values ​​are weighted and fused to recalculate the overall confidence score. This recalculated overall confidence score will be used as the updated confidence assessment result for subsequent deviation event detection.

[0061] This invention effectively addresses the problem of persistently inaccurate confidence assessment results due to environmental changes or sensor performance drift in the basic scheme by introducing a self-diagnostic process for confidence assessment. Specifically, when the system detects that the overall confidence value remains consistently below a preset threshold for a period of time, while the actual occupancy status remains stable, this indicates that the current confidence assessment model may be biased. At this point, by initiating the self-diagnostic process, the system can deeply analyze the historical data of various factors affecting the confidence assessment, thereby accurately identifying the key factors causing the decline in confidence. For example, if the adaptive calibration information of a sensor is found to be in an abnormal state for a long period, or if the intensity of local environmental disturbances consistently exceeds the normal range, these may be identified as key factors. Once the key factors are identified, the system can selectively adjust the parameters used to calculate the support value and the weighting rules for each factor during weighted fusion. This dynamic adjustment mechanism enables the confidence assessment model to adapt to constantly changing environmental conditions and sensor states, thereby outputting more accurate and reliable confidence assessment results. Therefore, when the system determines whether a deviation event is caused by inaccurate external control commands or by a malfunction of internal components of the lighting fixture, it can make a judgment based on more reliable confidence information, significantly reducing the risk of misjudgment.

[0062] Through the above technical solution, the present invention can achieve adaptive optimization and self-correction of the confidence assessment link in the intelligent lighting controller operation data analysis method.

[0063] In some preferred embodiments, a specific example is given below. Assume a smart office area where a lighting controller continuously operates, generating a local occupancy status judgment based on local sensing data (such as local passive infrared sensor data, internal temperature fluctuation data of the lighting fixtures, and surrounding wireless signal characteristic data), and further assessing its confidence level. Initially, the confidence level assessment result (overall confidence value) is typically high, indicating that the judgment result is reliable. However, after several months of operation, the office area's air conditioning system is upgraded, causing changes in local airflow and temperature fluctuation patterns. Simultaneously, several new wireless devices are added in the vicinity, making the wireless signal environment more complex. After these changes, the system detects that the overall confidence value begins to consistently fall below a preset low confidence threshold (e.g., 0.6), and during these low confidence periods, the actual occupancy status monitored by independent millimeter-wave radar data, thermal imaging sensor data, and anonymized video analysis data systems remains stable (e.g., someone is continuously working in the area). At this point, the self-diagnostic process of this invention is triggered. The system begins to analyze the historical trends of various factors. Analysis revealed a significant upward trend in historical data regarding the interference intensity of local environmental disturbances on local sensing sensors, with a marked increase in its fluctuation range. Conversely, other factors (such as the consistency between local sensing data and the completeness and real-time performance of local sensing data) showed relatively smaller changes. Based on this, the system identified "the interference intensity of local environmental disturbances on local sensing sensors" as the key factor causing the persistently low overall confidence score. According to this key factor, the system automatically adjusted the parameters determining the support value for this factor. For example, it increased a threshold previously used to calculate the support value, reducing its tolerance to environmental disturbances and thus more sensitively reflecting their impact. Simultaneously, the system adjusted the preset weighting rules. For instance, it reduced the weight of "the interference intensity of local environmental disturbances on local sensing sensors" in the weighted fusion calculation, or increased the weight of relatively stable factors such as "the consistency between local sensing data." Based on these adjusted parameters and rules, the system recalculated the support values ​​of each factor and performed weighted fusion to obtain an updated overall confidence score. After adjustments, the new overall confidence level returned to a high level (e.g., above 0.8), indicating that the confidence assessment model has successfully adapted to the new environmental conditions. Therefore, the system can continue to accurately identify the causes of deviations and avoid misjudgments caused by environmental changes.

[0064] The present invention further proposes a step for continuously monitoring the actual occupancy status of the area where the lighting fixture is located, including: When the environmental characteristics are identified as matching the preset complex deployment environment pattern, the multi-source heterogeneous sensor data acquisition mode is activated. In the multi-source heterogeneous sensor data acquisition mode, multi-source data is acquired; Perform time synchronization and spatial registration of multi-source data; Denoising processing is performed on multi-source data after time synchronization and spatial registration; When intermittent missing or partial occlusion of data from any type of sensor is detected, data from other types of sensors are used to complete the information of that type of sensor data. When a conflict is detected between two or more types of sensor data, the conflicting data is resolved according to the preset sensor priority and data confidence rules. Time series analysis is used to smooth the completed and eliminated data to generate continuous actual occupancy status data. Assess the integrity of continuous actual occupancy status data; The actual occupancy status of the area where the lighting fixtures are located is determined based on the continuous actual occupancy status data after assessment.

[0065] Specifically, when the system identifies that the current environmental characteristics match a preset complex deployment environment pattern, such as in areas with frequent personnel movement, physical obstructions, or high environmental noise, the system will automatically activate the multi-source heterogeneous sensor data acquisition mode. In this mode, the system acquires multi-source data from various types of sensors, such as millimeter-wave radar data, thermal imaging sensor data, and anonymized video analysis data. To ensure effective fusion of this heterogeneous data, the acquired multi-source data needs to be time-synchronized and spatially registered to eliminate time delays and spatial differences between different sensors, mapping all data to a unified spatiotemporal coordinate system. Subsequently, the time-synchronized and spatially registered multi-source data undergoes denoising processing to filter out environmental interference or noise generated by the sensors themselves, improving data purity.

[0066] Furthermore, during data processing, when the system detects intermittent missing or partially obstructed data from any type of sensor—for example, a sensor temporarily malfunctioning or being blocked by an object—the system intelligently uses data from other types of sensors to complete the missing or obstructed data, maintaining data continuity and integrity. Simultaneously, when conflicts are detected between two or more types of sensor data—for example, different sensors inconsistently determining the occupancy status of the same area—the system resolves the conflicting data according to preset sensor priorities and data confidence rules, ensuring the accuracy of the final judgment. For example, it can prioritize sensor data with higher confidence or make decisions based on empirical rules specific to the scenario.

[0067] Therefore, by using time series analysis, the completed and simplified data is smoothed to generate continuous actual occupancy status data, avoiding the impact of instantaneous fluctuations on the judgment results. Based on this, the system evaluates the completeness of the continuous actual occupancy status data to ensure the quality and reliability of the generated data. Finally, based on the evaluated continuous actual occupancy status data, the system can accurately determine the actual occupancy status of the area where the lighting fixtures are located.

[0068] This invention effectively addresses the problem of inaccurate actual occupancy status monitoring that may result from single or limited sensor data in complex environments by introducing an adaptive multi-source heterogeneous sensor data acquisition and processing mechanism. Specifically, when the environment is identified as a complex pattern, the system can proactively activate multi-source data acquisition, leveraging the complementary advantages of different types of sensors (such as millimeter-wave radar, thermal imaging, and anonymized video analysis) to obtain more comprehensive and robust environmental information. Time synchronization and spatial registration ensure effective fusion of heterogeneous data; noise reduction processing improves data purity. More importantly, this solution addresses the challenge of intermittent missing or partially obscured sensor data through an information completion mechanism, ensuring data continuity; while a conflict data resolution mechanism resolves potential discrepancies between different sensors, improving data consistency. Finally, time series analysis smooths the data, and the integrity of continuous actual occupancy status data is assessed, ensuring that the final output actual occupancy status data is highly reliable and accurate. This series of processing steps works together to enable the system to obtain high-quality actual occupancy status information in various complex scenarios, providing a solid data foundation for subsequent confidence assessment and self-diagnosis processes.

[0069] Through the above technical solution, the present invention significantly improves the accuracy, robustness and reliability of the intelligent lighting controller in monitoring the actual occupancy status of the area where the lighting fixtures are located in complex deployment environments.

[0070] In some preferred embodiments, a specific example is given below. Suppose a large, open-plan office area where a smart lighting controller needs to accurately sense the actual occupancy status of people to optimize lighting. This area may have complex environmental characteristics such as office partitions, mobile workstations, and frequent personnel movement. When the system identifies that the environmental characteristics of the office area match a preset complex deployment environment pattern, such as during peak weekday hours, the system will activate a multi-source heterogeneous sensor data acquisition mode. At this time, the system will simultaneously acquire multi-source data from millimeter-wave radar sensors deployed on the ceiling, thermal imaging sensors, and anonymized video analysis modules. To ensure that these data can work collaboratively, the system first synchronizes all sensor data in time to ensure that they reflect the environmental state at the same moment, and performs spatial registration to unify the data from different sensor perspectives onto the office area's floor plan. Subsequently, these synchronized and registered data undergo noise reduction processing to filter out interference caused by non-human factors such as air conditioning airflow and changes in external light. In actual operation, the following situations may occur: For example, when an employee remains stationary at their workstation for an extended period, the millimeter-wave radar may show reduced response due to lack of movement, but thermal imaging sensors and anonymized video analysis data can still clearly indicate their presence. In this case, the system will use thermal imaging and video analysis data to supplement the missing information from the millimeter-wave radar. Similarly, when multiple employees are active in a certain area simultaneously, the millimeter-wave radar may detect multiple targets, while the thermal imaging sensor may experience overlapping heat sources due to dense crowds, and anonymized video analysis may provide more accurate information on the number of people and their locations. If different sensors conflict in their judgments of the number of people or their locations, the system will resolve the conflicting data according to preset priority rules (e.g., prioritizing anonymized video analysis for number of people and prioritizing millimeter-wave radar for motion detection). After completion and resolution, the data will be smoothed through time-series analysis to generate a continuous and stable stream of actual occupancy status data, avoiding momentary misjudgments. Finally, the system will evaluate the integrity of this continuous data, such as checking for long periods of data gaps or abnormal fluctuations, to ensure the reliability of the data and accurately determine the actual occupancy status of the office area, providing high-quality input for the operation data analysis of the intelligent lighting controller.

[0071] In some embodiments of the present invention described above, when environmental characteristics are identified as matching a preset complex deployment environment pattern, a multi-source heterogeneous sensor data acquisition mode is activated, and multi-source data is acquired. Specifically, the aforementioned multi-source data may include millimeter-wave radar data, thermal imaging sensor data, and anonymized video analysis data.

[0072] Millimeter-wave radar data refers to data acquired using millimeter-wave radar sensors regarding the distance, speed, and angle of a target object. Millimeter-wave radar has the ability to penetrate non-metallic obstacles and is less affected by environmental factors such as light and smoke, providing high-precision motion and presence detection. Thermal imaging sensor data refers to data capturing the surface temperature distribution of objects within an area. Thermal imaging sensors can detect the heat emitted by the human body, effectively identifying the presence of people even in complete darkness or low-light environments, and offering good privacy protection. Anonymized video analytics data refers to data extracted from video streams after processing, free of personally identifiable information, used to analyze the activity patterns and occupancy status of people within an area. Anonymization effectively protects user privacy while leveraging the powerful capabilities of video analytics to provide refined personnel counting, location tracking, and behavior recognition information.

[0073] This invention introduces millimeter-wave radar data, thermal imaging sensor data, and anonymized video analysis data as multi-source data to improve the accuracy and robustness of monitoring the actual occupancy status of lighting fixtures in complex deployment environments. Millimeter-wave radar data provides precise motion and presence information, particularly suitable for detecting stationary or slightly moving personnel, compensating for potential blind spots in traditional sensors. Thermal imaging sensor data identifies personnel by sensing heat, unaffected by lighting conditions, enhancing detection capabilities in dark or changing lighting environments. Anonymized video analysis data, while protecting privacy, provides richer visual information, aiding in identifying complex human behavior patterns and precise location. The combination of these heterogeneous data sources enables comprehensive, multi-dimensional perception of actual occupancy status under a multi-source heterogeneous sensor data acquisition mode, effectively addressing issues such as false alarms, missed alarms, or incomplete data that may occur with single sensors in complex environments. For example, when data from one sensor fails due to partial obstruction, data from other sensors can still provide valid occupancy information, ensuring the continuity and reliability of data acquisition.

[0074] Through the above technical solution, the present invention can significantly enhance the ability to monitor the actual occupancy status of the area where the lighting fixtures are located in complex deployment environments.

[0075] The present invention further proposes a step for evaluating the integrity of continuous actual occupancy status data, including: Monitoring and characterizing the occupancy changes within areas where the boundary of the area where the lighting fixture is located is less than a preset distance threshold; Based on the changes in occupancy response to the boundary of the area where the luminaire is located, the fluctuation frequency of the area boundary is determined. The change in occupancy response refers to the changes in the occupancy status, occupancy probability and / or occupancy intensity corresponding to the multi-source data when the target object crosses the boundary of the area where the luminaire is located. When the frequency of regional boundary fluctuations exceeds the preset frequency threshold, the virtual boundary of the area where the lamp is located is redefined based on the effective coverage and corresponding response intensity of millimeter-wave radar data, thermal imaging sensor data, and anonymized video analysis data, and the data integrity assessment weight is increased within the corresponding time period of virtual boundary adjustment. When the frequency of fluctuations in the region boundary is lower than or equal to the preset frequency threshold, the current virtual boundary is maintained, and the default data integrity assessment weight is maintained. The monitoring personnel's stay time in the area where the lights are located is identified as a rapid entry / exit event when the stay time is less than a preset stay time threshold, and the data integrity assessment weight of the corresponding time period of the rapid entry / exit event is increased. When the dwell time is higher than or equal to the preset dwell time threshold, the default data integrity assessment weight is maintained; Monitor instantaneous occlusion events of any data stream among millimeter-wave radar data, thermal imaging sensor data, and anonymized video analysis data. When it is identified that the response intensity of any data stream decreases by more than a preset decrease threshold or is interrupted within a preset third time window, and the other two data streams are occupied in the area where the indicator lights are located, use the other two data streams to complete the occluded data and increase the data integrity assessment weight within the corresponding time period of the data completion process. When no transient occlusion event is detected, the default data integrity assessment weight is maintained; Based on the results of virtual boundary redetering, rapid entry / exit event identification, and data completion, the data integrity assessment parameters are adjusted. Then, based on the adjusted data integrity assessment parameters and the corresponding data integrity assessment weights, the integrity of continuous actual occupancy status data is assessed. The data integrity assessment parameters are used to characterize the degree of impact of virtual boundary changes, rapid entry / exit events, and data completion on the integrity of continuous actual occupancy status data.

[0076] Specifically, the multi-source data monitoring the occupancy changes in the area where the boundary of the area where the luminaire is located is less than a preset distance threshold refers to the multi-source data monitoring the occupancy changes in the area where the luminaire is located is located and the distance between the luminaire and the boundary is less than a preset distance threshold. This can be understood as monitoring the millimeter-wave radar data, thermal imaging sensor data, and anonymized video analysis data corresponding to the occupancy changes in the area where the luminaire is located is located and the distance between the luminaire and the boundary is less than a preset distance threshold.

[0077] When assessing the integrity of continuous actual occupancy data, continuous monitoring of occupancy changes near the boundary of the area where the luminaire is located is first required. The area boundary fluctuation frequency refers to the frequency at which the occupancy status, probability, and / or intensity reflected by multi-source data (such as millimeter-wave radar data, thermal imaging sensor data, and anonymized video analysis data) changes when a target object crosses the boundary of the area where the luminaire is located. When this fluctuation frequency is high, it indicates that the area boundary may be dynamically changing or that people are frequently entering and exiting. In this case, the virtual boundary of the area where the luminaire is located needs to be dynamically redefined based on the effective coverage and response strength of each sensor, and the data integrity assessment weight of the corresponding time period should be increased accordingly to more accurately reflect the reliability of the data during that period. Conversely, if the fluctuation frequency is low, the current virtual boundary and the default assessment weight are maintained.

[0078] Furthermore, to more accurately assess data integrity, it is also necessary to monitor the time people spend in the area where the lights are located. When the time people spend in the area is less than a preset threshold, this is usually identified as a rapid entry / exit event, such as a person briefly passing through or quickly retrieving an item. For such events, due to their transient nature, data collection may be incomplete. Therefore, it is necessary to increase the weight of data integrity assessment for the time period corresponding to the rapid entry / exit event to ensure that sufficient attention is paid to the data integrity of these brief events.

[0079] Furthermore, this invention also considers the possibility of transient occlusion events in multi-source data streams. A transient occlusion event refers to a situation where, within a preset third time window, the response intensity of any type of sensor data (such as millimeter-wave radar data, thermal imaging sensor data, or anonymized video analysis data) decreases by more than a preset decrease threshold or is interrupted, while the other two types of data still indicate occupancy within the area. In this case, to compensate for the information loss of the occluded data, the other two types of unoccluded data are used to complete the information of the occluded data, and the data integrity assessment weight within the corresponding time period of the data completion process is increased to reflect the improved reliability of the data after intelligent completion.

[0080] Finally, based on the results of the virtual boundary redetering, rapid entry / exit event identification, and data completion, the data integrity assessment parameters are dynamically adjusted. These parameters characterize the impact of virtual boundary changes, rapid entry / exit events, and data completion on the integrity of continuous actual occupancy status data. By adjusting these parameters and their corresponding data integrity assessment weights, a more accurate and robust assessment of the integrity of continuous actual occupancy status data can be achieved.

[0081] The present invention addresses the shortcomings of traditional methods in assessing the integrity of actual occupancy status data in complex dynamic environments by introducing a dynamic monitoring and response mechanism for fluctuations in area boundaries, rapid entry and exit of personnel, and instantaneous sensor occlusion.

[0082] Through the above technical solution, the present invention can significantly improve the accuracy and robustness of continuous actual occupancy status data integrity assessment.

[0083] In some preferred embodiments, consider a smart office area where the lighting controller continuously monitors occupancy status using millimeter-wave radar, thermal imaging sensors, and anonymized video analytics. When an employee quickly crosses the office doorway and stays in the area for less than 5 seconds, the system detects a brief but noticeable change in occupancy response near the area boundary in the millimeter-wave radar and thermal imaging sensor data, and the employee's stay time is less than a preset stay time threshold (e.g., 10 seconds). At this point, the system identifies this as a rapid entry / exit event and accordingly increases the data integrity assessment weight for the corresponding time period.

[0084] refer to Figure 2 The present invention further proposes an intelligent lighting controller operation data analysis system and an intelligent lighting controller operation data analysis method. The system includes: The acquisition module acquires the operating status data of the lighting fixtures and identifies deviation events when the operating status data deviates from a preset range. The response module, in response to a deviation event, acquires the external control commands received by the luminaire at the time the deviation event occurs, as well as the local sensing data of the area where the luminaire is located at the time the deviation event occurs; The expected state determination module determines the area occupancy status and the expected execution status of the lighting fixtures implied by the external control commands based on the external control commands. The comparison module generates local occupancy status detection results based on local sensing data, and compares the area occupancy status implied by external control commands with the area occupancy status represented by the local occupancy status detection results to obtain consistent or conflicting comparison results. The actual status determination module determines the actual operating status of the lighting fixtures based on the operating status data. The evaluation module assesses the degree of matching between the actual operating state and the expected execution state. The discrimination module determines, based on the comparison results and the degree of matching, whether the deviation event is caused by inaccurate external control commands or by a malfunction of internal components of the lighting fixture.

[0085] Specifically, the acquisition module is configured to acquire the operating status data of the lighting fixtures and identify deviation events when the operating status data deviates from a preset range. This module is responsible for continuously monitoring various operating parameters of the lighting fixtures, such as power consumption, brightness, and color temperature. Once these parameters are detected to exceed the normal fluctuation range, the identification of deviation events is triggered.

[0086] The response module is configured to respond to deviation events by acquiring external control commands received by the luminaire at the time of the deviation event, as well as local sensing data of the area where the luminaire is located at the time of the deviation event. After a deviation event occurs, the module rapidly collects contextual information related to the event, including control commands issued by the user or system and environmental sensor data.

[0087] The expected state determination module is configured to determine the area occupancy status and the expected execution state of the luminaire implied by the external control commands based on the external control commands. This module parses the control commands and infers the expected area status of the system at the time the commands are issued (e.g., whether someone is occupying the area) and the operating state that the luminaire should achieve (e.g., brightness level).

[0088] The comparison module is configured to generate local occupancy status detection results based on local sensing data, and compare the area occupancy status implied by external control commands with the area occupancy status represented by the local occupancy status detection results to obtain consistent or conflicting comparison results. This module independently determines the area occupancy status using local sensor data and cross-validates it with the expected external commands.

[0089] The actual status determination module is configured to determine the actual operating status of the luminaire based on operating status data. This module accurately reflects the current actual working status of the luminaire based on its real-time operating data.

[0090] The evaluation module is configured to assess the degree of match between actual operating conditions and expected performance. This module quantifies the difference between the actual performance of the lighting fixtures and the expected targets.

[0091] The discrimination module is configured to determine, based on the comparison results and the degree of matching, whether the deviation event is caused by inaccurate external control commands or by a malfunction in an internal component of the lighting fixture. This module is the core decision-making part of the system, comprehensively analyzing the comparison results and the degree of matching to provide a root cause diagnosis of the deviation event.

[0092] The present invention visualizes each logical step in the intelligent lighting controller's data analysis method as an independent system module, thereby achieving automation, parallelization, and efficient execution of the method.

[0093] The content disclosed above is only a preferred and feasible embodiment of the present invention, and is not intended to limit the scope of protection of the present invention. Therefore, all equivalent technical changes made based on the content of the present invention specification and drawings are included within the scope of protection of the present invention. Furthermore, the elements therein can be updated as technology develops.

Claims

1. A method for analyzing operational data of an intelligent lighting controller, characterized in that, The method includes the following steps: Acquire the operating status data of the lighting fixtures, and identify deviation events when the operating status data deviates from the preset range; In response to a deviation event, acquire the external control commands received by the luminaire at the time the deviation event occurs, as well as the local sensing data of the area where the luminaire is located at the time the deviation event occurs; Determine the area occupancy status and expected execution status of the lighting fixtures implied by the external control commands based on the external control commands; The local occupancy status detection result is generated based on the local sensing data, and the area occupancy status implied by the external control command is compared with the area occupancy status represented by the local occupancy status detection result to obtain consistent or conflicting comparison results. Determine the actual operating status of the lighting fixtures based on the operating status data; Assess the degree of match between the actual operating status and the expected execution status; Based on the comparison results and the degree of matching, it can be determined whether the deviation event is caused by inaccurate external control commands or by a malfunction of internal components of the lighting fixture.

2. The intelligent lighting controller operation data analysis method as described in claim 1, characterized in that, The steps of generating local occupancy status detection results based on local sensing data, and comparing the area occupancy status implied by external control commands with the area occupancy status represented by the local occupancy status detection results to obtain consistent or conflicting comparison results include: The local sensing data is time-stamp aligned and denoised to obtain the processed local sensing data. The local sensing data is fused and processed to generate a local occupancy status judgment result; The confidence level of the local occupancy status judgment result is evaluated to obtain the confidence level evaluation result. The confidence level evaluation considers the consistency between local sensing data, the integrity and real-time performance of local sensing data, the interference intensity of local environmental disturbances on local sensing sensors, and the adaptive calibration information of local sensing sensors. Based on the confidence assessment results, output the local occupancy status probability value and uncertainty value, which represent the local occupancy status judgment results, and generate the local occupancy status detection results based on the local occupancy status probability value and uncertainty value; The local occupancy status is determined based on the local occupancy status detection results, and the local occupancy status represented by the local occupancy status detection results is compared with the local occupancy status implied by the external control commands to obtain consistent or conflicting comparison results. The local sensing data includes local passive infrared sensor data, internal temperature fluctuation data of the lamps, and surrounding wireless signal characteristic data.

3. The intelligent lighting controller operation data analysis method as described in claim 2, characterized in that, The steps for generating a local occupancy status judgment result from the fused local sensing data include: Analyze the trigger frequency and signal strength changes of local passive infrared sensor data within a preset first time window to determine whether local passive infrared sensor data generates an indication of occupancy. The average value and standard deviation of the internal temperature fluctuation data of the lamps within a preset second time window, as well as the average intensity and fluctuation amplitude of the surrounding wireless signal characteristic data within a preset second time window, are analyzed to determine whether the internal temperature fluctuation data of the lamps and the surrounding wireless signal characteristic data together form a stable pattern indication that is consistent with the physical characteristics of stationary personnel. The system performs a fusion judgment based on the occupied status indication and the stable mode indication to generate a local occupancy status judgment result.

4. The intelligent lighting controller operation data analysis method as described in claim 2, characterized in that, The steps to assess the confidence level of the local occupancy status determination result and obtain the confidence level assessment result include: Identify and quantify the consistency among local sensing data, the integrity and real-time performance of local sensing data, the interference intensity of local environmental disturbances on local sensing sensors, and the impact of adaptive calibration information of local sensing sensors on the reliability of local occupancy status judgment results. Based on the degree of influence, the support value corresponding to each factor is determined. The factors include the consistency between local sensing data, the completeness and real-time performance of local sensing data, the interference intensity of local environmental disturbances on local sensing sensors, and the adaptive calibration information of local sensing sensors. The support value is a numerical value used to characterize the degree of reliability of the corresponding factor in the local occupancy status judgment result. According to the preset weight allocation rules, the support values ​​of each factor are weighted and fused to calculate the comprehensive confidence value. The corresponding weights are adjusted according to the historical performance of each factor in different scenarios and the current environmental conditions. Output the overall confidence score as the confidence assessment result.

5. The intelligent lighting controller operation data analysis method as described in claim 4, characterized in that, The steps for weighted fusion of the support values ​​corresponding to each factor according to the preset weight allocation rules include: Continuously monitor the environmental disturbance characteristics of the area where the lighting fixtures are located, and extract features from the environmental abnormal signals when they are identified as not conforming to the preset known interference patterns. The extracted features are compared with a pre-defined library of known interference features, and features that cannot be matched are marked as novel environmental interference. Analyze the impact of novel environmental interference on local sensing data output and generate a quantified novel interference impact factor; The support values ​​of the additional factors are determined based on the new interference factors, and the additional factors are included in each factor. The weights of the additional factors are determined according to the preset weight allocation rules, so as to perform weighted fusion of the support values ​​of each factor, including the additional factors.

6. The intelligent lighting controller operation data analysis method as described in claim 4, characterized in that, After outputting the overall confidence score as the confidence assessment result, the method further includes the following steps: Monitor the overall confidence level and compare it with a preset low confidence threshold; Monitor the actual occupancy status of the area where the lighting fixtures are located; When the overall confidence level value is continuously lower than the low confidence level threshold within the preset monitoring period, and the actual occupancy status of the area where the lamp is located remains consistent within the preset monitoring period, the self-diagnosis process of confidence assessment is triggered. In the self-diagnostic process of confidence assessment, analyze the historical trends of each factor; Identify the key factors that cause the overall confidence score to remain below the low confidence threshold for an extended period of time based on the historical trends of each factor. Based on key factors, the parameters for determining the support values ​​of each factor and the preset weighting rules are adjusted. The support values ​​of each factor are re-determined based on the adjusted parameters, and the re-determined support values ​​of each factor are weighted and fused according to the adjusted preset weighting rules to recalculate the overall confidence score. The recalculated overall confidence score is then used as the updated confidence score assessment result. The parameters for determining the support values ​​of each factor are used to determine the degree of influence of each factor on the reliability of the local occupancy status judgment result.

7. The intelligent lighting controller operation data analysis method as described in claim 6, characterized in that, The steps for continuously monitoring the actual occupancy status of the area where the lighting fixtures are located include: When environmental characteristics are identified as matching the preset complex deployment environment pattern, the multi-source heterogeneous sensor data acquisition mode is activated. In the multi-source heterogeneous sensor data acquisition mode, multi-source data is acquired; Perform time synchronization and spatial registration of multi-source data; Denoising processing is performed on multi-source data after time synchronization and spatial registration; When intermittent missing or partial occlusion of data from any type of sensor is detected, data from other types of sensors are used to complete the information of that type of sensor data. When a conflict is detected between two or more types of sensor data, the conflicting data is resolved according to the preset sensor priority and data confidence rules. Time series analysis is used to smooth the data after completion and resolution to generate continuous actual occupancy status data; Assess the integrity of continuous actual occupancy status data; The actual occupancy status of the area where the lighting fixtures are located is determined based on the continuous actual occupancy status data after assessment.

8. The intelligent lighting controller operation data analysis method as described in claim 7, characterized in that, The multi-source data includes millimeter-wave radar data, thermal imaging sensor data, and anonymized video analysis data.

9. The intelligent lighting controller operation data analysis method as described in claim 8, characterized in that, The steps for assessing the integrity of continuous physical occupancy status data include: Monitoring and characterizing the occupancy changes within areas where the boundary of the area where the lighting fixture is located is less than a preset distance threshold; Based on the changes in occupancy response to the boundary of the area where the luminaire is located, the fluctuation frequency of the area boundary is determined. The change in occupancy response refers to the changes in the occupancy status, occupancy probability and / or occupancy intensity corresponding to the multi-source data when the target object crosses the boundary of the area where the luminaire is located. When the frequency of regional boundary fluctuations exceeds the preset frequency threshold, the virtual boundary of the area where the lamp is located is redefined based on the effective coverage and corresponding response intensity of millimeter-wave radar data, thermal imaging sensor data, and anonymized video analysis data, and the data integrity assessment weight is increased within the corresponding time period of virtual boundary adjustment. When the frequency of fluctuations in the region boundary is lower than or equal to the preset frequency threshold, the current virtual boundary is maintained, and the default data integrity assessment weight is maintained. The monitoring personnel's stay time in the area where the lights are located is identified as a rapid entry / exit event when the stay time is less than a preset stay time threshold, and the data integrity assessment weight of the corresponding time period of the rapid entry / exit event is increased. When the dwell time is higher than or equal to the preset dwell time threshold, the default data integrity assessment weight is maintained; Monitor instantaneous occlusion events of any data stream among millimeter-wave radar data, thermal imaging sensor data, and anonymized video analysis data. When it is identified that the response intensity of any data stream decreases by more than a preset decrease threshold or is interrupted within a preset third time window, and the other two data streams are occupied in the area where the indicator lights are located, use the other two data streams to complete the occluded data and increase the data integrity assessment weight within the corresponding time period of the data completion process. When no transient occlusion event is detected, the default data integrity assessment weight is maintained; Based on the results of virtual boundary redetering, rapid entry / exit event identification, and data completion, the data integrity assessment parameters are adjusted. Then, based on the adjusted data integrity assessment parameters and the corresponding data integrity assessment weights, the integrity of continuous actual occupancy status data is assessed. The data integrity assessment parameters are used to characterize the degree of impact of virtual boundary changes, rapid entry / exit events, and data completion on the integrity of continuous actual occupancy status data.

10. A smart lighting controller operation data analysis system, employing the smart lighting controller operation data analysis method described in claim 1, characterized in that, The system includes: The acquisition module acquires the operating status data of the lighting fixtures and identifies deviation events when the operating status data deviates from a preset range. The response module, in response to a deviation event, acquires the external control commands received by the luminaire at the time the deviation event occurs, as well as the local sensing data of the area where the luminaire is located at the time the deviation event occurs; The expected state determination module determines the area occupancy status and the expected execution status of the lighting fixtures implied by the external control commands based on the external control commands. The comparison module generates local occupancy status detection results based on local sensing data, and compares the area occupancy status implied by external control commands with the area occupancy status represented by the local occupancy status detection results to obtain consistent or conflicting comparison results. The actual status determination module determines the actual operating status of the lighting fixtures based on the operating status data. The evaluation module assesses the degree of matching between the actual operating state and the expected execution state. The discrimination module determines, based on the comparison results and the degree of matching, whether the deviation event is caused by inaccurate external control commands or by a malfunction of internal components of the lighting fixture.

Images