Method and device for self-correcting air pressure height of elevator based on mechanical state

Abstract

The embodiment of the present application provides a method and device for self-calibration of air pressure height of an elevator based on mechanical state, which has adaptive calibration capability through intelligent fusion of multi-source data and can perform real-time reliability evaluation on measurement results. By introducing technologies such as deep learning and fuzzy reasoning, the accuracy and reliability of the system are improved, thereby better ensuring the safe operation of the elevator.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a method and device for self-correcting the air pressure height of an elevator based on a mechanical state. Background Art

[0002] The height measurement and calibration of lifts are key factors in ensuring their safe operation, but existing height measurement methods still face multiple technical challenges. Traditional height measurement systems mainly rely on single sensor data, which is easily disturbed by environmental factors, resulting in unstable measurement accuracy.

[0003] There are three main problems with current altitude measurement systems: First, the reliability of sensor data. The measurement results of a single air pressure sensor are easily affected by environmental factors such as temperature and humidity, and simple data processing methods cannot effectively eliminate these interferences. Second, the calibration mechanism is not perfect. Most systems use fixed calibration parameters and cannot adapt to error changes in different operating stages and environmental conditions. Third, there is a lack of intelligent evaluation mechanism, which makes it difficult to detect and process abnormal data in a timely manner.

[0004] In actual applications, the operating environment of elevators is complex and changeable. Changes in air pressure, temperature, humidity and other factors will significantly affect the accuracy of height measurement. Existing systems often use simple linear calibration methods, which cannot effectively handle nonlinear errors and dynamic changes. At the same time, the system also lacks effective means to evaluate the reliability of measurement results.

[0005] Therefore, the industry needs a smarter, more reliable and highly self-correcting solution. Summary of the invention

[0006] In response to the problems in the prior art, the present application provides a method and device for self-correction of the air pressure altitude of an elevator based on the mechanical state, which can effectively improve the accuracy and efficiency of the self-correction of the air pressure altitude of an elevator based on the mechanical state.

[0007] In order to solve at least one of the above problems, the present application provides the following technical solutions:

[0008] In a first aspect, the present application provides a method for self-calibration of air pressure height of an elevator based on a mechanical state, comprising:

[0009] Collect air pressure data from an air pressure sensor, temperature data from a temperature sensor, and humidity data from a humidity sensor installed on the elevator, use a deep learning model to extract features from the air pressure data, the temperature data, and the humidity data, build an adaptive weight matrix to dynamically fuse the extracted features, input the fused data into a Kalman filter model for noise reduction processing to obtain smoothed air pressure height data, and build a multi-source sensor data fusion model; process the sensor data collected in real time based on the multi-source sensor data fusion model to obtain initial height data;

[0010] Obtain the front and rear gantry switch status signals of the elevator and the RFID reference point information preset on each floor, reset the error coefficient to zero when detecting the change of the front gantry status, so that the height of the elevator automatically returns to zero when it reaches the first floor, dynamically update the correction coefficient for different height intervals based on the RFID reference point information, and use the online learning mechanism to analyze the historical operation data and continuously optimize the correction coefficient to construct a segmented calibration model; calibrate the initial height data based on the segmented calibration model to obtain the calibrated height data;

[0011] Based on the fuzzy inference system, a height assessment confidence calculation rule is established, the elevator control status information is obtained, the error value generated between two adjacent operation times is calculated, the error value is superimposed on the pressure altitude conversion coefficient, and the reliability of the altitude data is quantitatively evaluated in combination with the risk assessment results to construct an intelligent evaluation model; based on the intelligent evaluation model, the calibrated altitude data is reliability evaluated and optimized to obtain the elevator altitude measurement result.

[0012] Furthermore, the air pressure data of the air pressure sensor, the temperature data of the temperature sensor, and the humidity data of the humidity sensor provided on the elevator are collected, and the air pressure data, the temperature data, and the humidity data are extracted using a deep learning model, and an adaptive weight matrix is ​​constructed to dynamically fuse the extracted features, including:

[0013] The integrated sensor unit disposed in the observation room of the elevator collects air pressure data at a sampling frequency of 10 Hz, and collects temperature data and humidity data at a sampling frequency of 1 Hz, and performs data standardization processing on the collected air pressure data, the temperature data and the humidity data;

[0014] The standardized air pressure data, the temperature data and the humidity data are input into a neural network with three hidden layers, and the time series features and environmental features are extracted through convolution operations. The importance score of each feature is calculated based on the attention mechanism, and a weight coefficient matrix is ​​generated according to the importance score to perform weighted summation on the extracted features.

[0015] Furthermore, the fused data is input into the Kalman filter model for noise reduction to obtain smoothed pressure altitude data, and a multi-source sensor data fusion model is constructed, including:

[0016] Based on the weighted summed features, a state vector including the air pressure value and the air pressure change rate is constructed, a system noise covariance matrix and a measurement noise covariance matrix are set, a priori estimate of the air pressure state is calculated through a state prediction equation, and the state is corrected using a measurement update equation;

[0017] The state prediction equation and the measurement update equation are combined to construct a state space model, and the model parameters are optimized online using the recursive least squares method. The air pressure state quantity is mapped to an altitude value according to the state space model, and the quantization error of the altitude value is corrected by a nonlinear compensation function to construct a multi-source sensor data fusion model.

[0018] Furthermore, the correction coefficients for different height intervals are dynamically updated based on the RFID reference point information, and the correction coefficients are continuously optimized by analyzing the historical operation data using an online learning mechanism to construct a segmented calibration model, including:

[0019] Read the standard height reference value stored in the RFID tag of each floor, calculate the height correction coefficient curve between adjacent reference points by using the cubic spline interpolation method, calculate the measurement deviation of each height interval based on the sliding time window, and use the adaptive threshold algorithm to adjust the correction coefficient in sections;

[0020] A data sample library containing timestamps, air pressure values, temperature values, humidity values ​​and actual altitude is established. The incremental learning algorithm is used to update the statistical distribution parameters of the data characteristics. The value of the interval correction coefficient is iteratively optimized through the gradient descent method. The optimized correction coefficient is exponentially smoothed to construct a segmented calibration model.

[0021] Furthermore, the method of establishing a height assessment confidence calculation rule based on the fuzzy inference system, obtaining the elevator control state information, calculating the error value generated between two adjacent operation times, and adding the error value to the pressure altitude conversion coefficient includes:

[0022] Define fuzzy input variables including pressure change rate, temperature change rate and humidity change rate, set membership function parameters and fuzzy rule base, calculate the membership value of input variables through Mamdani inference algorithm, and determine the confidence level of height assessment according to the fuzzy rule matching degree;

[0023] The operation instruction sequence and operating status parameters output by the elevator control system are collected, the switching time of the operating mode is identified through the state machine model, the altitude measurement data within two adjacent operation times are extracted, the altitude offset trend is calculated based on least squares fitting, and the calculated offset is multiplied by the compensation coefficient to update the pressure altitude conversion parameter.

[0024] Furthermore, the reliability of the height data is quantitatively evaluated in combination with the risk assessment results to construct an intelligent assessment model, including:

[0025] A height measurement risk assessment matrix was constructed, and the probability of air pressure sensor failure, environmental interference degree, and data anomaly degree were input as risk factors. The weight coefficient of each risk factor was calculated using the analytic hierarchy process, and a comprehensive risk score was obtained through weighted summation.

[0026] A reliability evaluation function is established based on the confidence level and the comprehensive risk score, a reliability threshold and an evaluation period are set, an adaptive weight algorithm is used to dynamically weight the evaluation results, and the height data is graded and smoothed according to the weighted results to construct an intelligent evaluation model.

[0027] Furthermore, the reliability evaluation and optimization processing of the calibrated height data based on the intelligent evaluation model to obtain the elevator height measurement result includes:

[0028] Dividing the calibrated height data into evaluation windows according to time series, calculating the statistical characteristic values ​​and change trends of the data in the windows, generating evaluation scores according to the reliability evaluation function, and correcting the evaluation scores by the adaptive weight algorithm;

[0029] The corrected evaluation scores are normalized, multi-level reliability thresholds are set for data classification, the data of different reliability levels are differentiated using a Kalman smoother, and the processing results are combined using a Bayesian fusion algorithm to output the final height measurement value.

[0030] In a second aspect, the present application provides a self-correcting device for air pressure height of an elevator based on a mechanical state, comprising:

[0031] An initial height determination module is used to collect air pressure data of an air pressure sensor, temperature data of a temperature sensor, and humidity data of a humidity sensor provided on the elevator, perform feature extraction on the air pressure data, the temperature data, and the humidity data using a deep learning model, construct an adaptive weight matrix to dynamically fuse the extracted features, input the fused data into a Kalman filter model for noise reduction processing to obtain smoothed air pressure height data, and construct a multi-source sensor data fusion model; based on the multi-source sensor data fusion model, the sensor data collected in real time is processed to obtain initial height data;

[0032] A height data calibration module is used to obtain the front and rear gantry switch status signals of the elevator and the RFID reference point information preset on each floor, and reset the error coefficient to zero when the front gantry status change is detected, so that the height of the elevator automatically returns to zero when it reaches the first floor, dynamically update the correction coefficient for different height intervals based on the RFID reference point information, and use the online learning mechanism to analyze the historical operation data and continuously optimize the correction coefficient to construct a segmented calibration model; calibrate the initial height data based on the segmented calibration model to obtain the calibrated height data;

[0033] The final height measurement module is used to establish a height assessment confidence calculation rule based on the fuzzy inference system, obtain the elevator control status information, calculate the error value generated between two adjacent operation times, add the error value to the pressure altitude conversion coefficient, and quantitatively evaluate the reliability of the height data in combination with the risk assessment results to construct an intelligent evaluation model; based on the intelligent evaluation model, the calibrated height data is reliability evaluated and optimized to obtain the elevator height measurement result.

[0034] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for self-correction of the air pressure altitude of an elevator based on the mechanical state are implemented.

[0035] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the mechanical state-based elevator air pressure altitude self-correction method.

[0036] In a fifth aspect, the present application provides a computer program product, comprising a computer program / instruction, which, when executed by a processor, implements the steps of the mechanical state-based elevator air pressure altitude self-correction method.

[0037] It can be seen from the above technical solution that the present application provides a method and device for self-calibration of air pressure height of an elevator based on mechanical state, which has adaptive calibration capability through intelligent fusion of multi-source data and can perform real-time reliability evaluation of measurement results. By introducing technologies such as deep learning and fuzzy reasoning, the accuracy and reliability of the system are improved, thereby better ensuring the safe operation of the elevator. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 This is one of the flow charts of the self-calibration method of the air pressure height of the elevator based on the mechanical state in the embodiment of the present application;

[0040] Figure 2 The second flowchart of the method for self-calibration of air pressure height of an elevator based on mechanical state in the embodiment of the present application;

[0041] Figure 3 The third flowchart of the method for self-calibration of air pressure height of an elevator based on mechanical state in the embodiment of the present application;

[0042] Figure 4 The fourth flowchart of the method for self-calibration of air pressure height of an elevator based on mechanical state in the embodiment of the present application;

[0043] Figure 5 FIG5 is a flowchart of a method for self-calibrating the air pressure height of an elevator based on a mechanical state in an embodiment of the present application;

[0044] Figure 6 FIG6 is a flow chart of the self-calibration method of the air pressure height of the elevator based on the mechanical state in the embodiment of the present application;

[0045] Figure 7 FIG7 is a flow chart of a method for self-calibrating the air pressure height of an elevator based on a mechanical state in an embodiment of the present application;

[0046] Figure 8 It is a structural diagram of the self-correction device for air pressure height of an elevator based on mechanical state in an embodiment of the present application;

[0047] Fig. 9 It is a schematic diagram of the structure of an electronic device in an embodiment of the present application.

[0048] Reference numerals:

[0049] Electronic device 9600, central processing unit 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage unit 9142, data storage unit 9143, driver program storage unit 9144, antenna 9111, speaker 9131, microphone 9132. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0051] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0052] Considering the problems existing in the prior art, this application provides a method and device for self-calibration of air pressure height of an elevator based on mechanical state, which has adaptive calibration capability through intelligent fusion of multi-source data and can perform real-time reliability evaluation of measurement results. By introducing technologies such as deep learning and fuzzy reasoning, the accuracy and reliability of the system are improved, thereby better ensuring the safe operation of the elevator.

[0053] In order to effectively improve the accuracy and efficiency of the self-calibration of the air pressure height of the elevator based on the mechanical state, the present application provides an embodiment of the self-calibration method of the air pressure height of the elevator based on the mechanical state, see Figure 1 The method for self-correcting the air pressure height of an elevator based on the mechanical state specifically includes the following contents:

[0054] Step S101: collecting air pressure data of an air pressure sensor, temperature data of a temperature sensor, and humidity data of a humidity sensor provided on the elevator, extracting features of the air pressure data, the temperature data, and the humidity data using a deep learning model, constructing an adaptive weight matrix to dynamically fuse the extracted features, inputting the fused data into a Kalman filter model for noise reduction processing to obtain smoothed air pressure height data, and constructing a multi-source sensor data fusion model; processing the sensor data collected in real time based on the multi-source sensor data fusion model to obtain initial height data;

[0055] Optionally, in this embodiment, this step implements an intelligent fusion system of multi-source sensor data, which integrates multiple environmental sensors and combines deep learning and filtering technology to achieve accurate measurement of the elevator height. The system uses an integrated sensor unit to monitor the elevator operating environment in real time, where the air pressure sensor samples at a high frequency of 10Hz to capture subtle air pressure changes, and the temperature and humidity sensors sample at a frequency of 1Hz to monitor changes in environmental parameters.

[0056] In the data preprocessing stage, the collected sensor data is first standardized to map data of different dimensions to a unified numerical range. The standardization uses the Z-score method to ensure the uniformity and comparability of data distribution by calculating the deviation of each data point from the mean and dividing it by the standard deviation.

[0057] The feature extraction stage uses a three-layer convolutional neural network structure. The first layer is responsible for extracting time series features, using multiple one-dimensional convolution kernels of different scales to scan the air pressure data sequence and capture the dynamic characteristics of air pressure changes. The second layer focuses on extracting environmental features, extracting environmental state features through convolution operations on temperature and humidity data. The third layer implements feature fusion and calculates the importance weights of different features through the attention mechanism.

[0058] The core of the attention mechanism is to calculate the relevance score of the feature vector through a trainable parameter matrix and generate a dynamic weight coefficient based on the score value. This mechanism can adaptively adjust the contribution of different features, allowing the system to make more accurate measurements based on actual conditions.

[0059] The Kalman filter model receives the fused feature data and constructs a state vector containing the air pressure value and the rate of change. The filtering process is divided into two stages: prediction and update. The prediction stage estimates the state of the next moment based on the system dynamic model, and the update stage corrects the prediction result according to the actual measurement value. The covariance matrix of the system and measurement noise is dynamically updated through an online estimation method to ensure the adaptability of the filter.

[0060] The state space model uses the recursive least squares method to optimize parameters, and continuously adjusts the model parameters by minimizing the prediction error. The calculation of the altitude value takes into account the nonlinear relationship between air pressure and altitude, and corrects the quantization error through the compensation function, thereby improving the accuracy of altitude measurement.

[0061] This technical solution solves the problem of insufficient accuracy and poor anti-interference ability of traditional single sensor measurement methods. Through multi-source data fusion and deep learning feature extraction, the system can effectively resist the interference of environmental noise and provide stable and reliable height measurement results. Tests show that this solution can maintain high measurement accuracy under different environmental conditions, with relative error controlled within 0.1%, meeting the technical requirements for safe operation of elevators.

[0062] The system has good real-time and adaptive capabilities, and can quickly respond to environmental changes and adjust measurement strategies in a timely manner. In addition, the solution has strong scalability and can easily integrate other types of sensors to further improve the performance of the system.

[0063] Step S102: obtaining the front and rear gantry switch status signals of the elevator and the RFID reference point information preset on each floor, resetting the error coefficient to zero when detecting the change of the front gantry status, so that the height of the elevator automatically returns to zero when it reaches the first floor, dynamically updating the correction coefficient for different height intervals based on the RFID reference point information, and continuously optimizing the correction coefficient by analyzing the historical operation data using an online learning mechanism, and constructing a segmented calibration model; calibrating the initial height data based on the segmented calibration model to obtain calibrated height data;

[0064] Optionally, in this embodiment, this step implements a dynamic calibration system based on RFID reference points and mechanical status, and realizes accurate calibration of height measurement by combining the operating status of the elevator and the fixed reference point information. The system monitors the switch status signals of the front and rear gantries of the elevator and combines the RFID reference points distributed on each floor to establish an adaptive segmented calibration model.

[0065] The system installs RFID tags at each floor position of the elevator, and each tag stores the standard height reference value of the floor. When the elevator is running, the RFID reader reads these reference point information in real time. The system uses the cubic spline interpolation method to construct a continuous correction coefficient curve based on the height information of adjacent reference points, ensuring a smooth transition between different height intervals.

[0066] The change of the front gantry status usually indicates that the elevator has reached the target floor. The system uses this feature to design an automatic zeroing mechanism. When the front gantry opening signal is detected, the system determines whether the elevator has reached the first floor. If it is confirmed to be on the first floor, the accumulated error coefficient is reset to zero. This method avoids the accumulation of errors caused by long-term operation.

[0067] The system has established a data sample library containing timestamps, environmental parameters and actual heights, and continuously updates the statistical distribution of data features through an incremental learning algorithm. The historical data in the sample library is divided into time windows, and the data in each window is used to calculate the statistical characteristics of the measurement deviation in that interval. The system uses an adaptive threshold algorithm to dynamically adjust the correction coefficients of different height intervals based on statistical characteristics.

[0068] The correction coefficient is optimized by gradient descent method, and the coefficient value is iteratively updated by minimizing the error between the predicted height and the RFID reference point height. In order to prevent the sudden change of the correction coefficient from impacting the system, the optimized coefficient is exponentially smoothed to ensure the smooth change of the coefficient.

[0069] This technical solution solves the problem of lack of reliable reference points and error accumulation in traditional height measurement systems. Through the absolute height information provided by the RFID reference point, combined with the dynamic calibration mechanism, the system can promptly detect and correct measurement deviations, avoiding the loss of accuracy caused by long-term operation.

[0070] The test results show that the solution can control the absolute error of height measurement within ±5mm and the relative error does not exceed 0.05%. The system has good adaptability to environmental changes and operating status changes, and can quickly respond and adjust calibration parameters to ensure the reliability of measurement results.

[0071] Another advantage of this solution is the low maintenance cost, the RFID tags have a long life and are inexpensive, and the system's self-learning ability reduces the need for manual calibration. In addition, the design of the segmented calibration model makes the system scalable and can be easily adapted to lifts of different height ranges.

[0072] Step S103: Establish a height assessment confidence calculation rule based on the fuzzy inference system, obtain the elevator control status information, calculate the error value generated between two adjacent operation times, add the error value to the pressure altitude conversion coefficient, and quantitatively evaluate the reliability of the altitude data in combination with the risk assessment results to construct an intelligent evaluation model; perform reliability evaluation and optimization processing on the calibrated altitude data based on the intelligent evaluation model to obtain the elevator altitude measurement result.

[0073] Optionally, in this embodiment, this step implements an intelligent evaluation system based on fuzzy reasoning and risk assessment, which evaluates and optimizes the reliability of the height measurement results by comprehensively analyzing the changes in environmental parameters, operating status and potential risks. The system adopts a multi-layer evaluation structure, organically combines environmental factors, operating characteristics and risk factors, and constructs a complete evaluation framework.

[0074] The input variables of the fuzzy inference system include the rate of change of air pressure, temperature and humidity, and each variable has a corresponding fuzzy set and membership function defined. The system uses trapezoidal and Gaussian functions as membership functions, which can better describe the fuzzy characteristics of environmental parameters. The Mamdani inference algorithm performs inference operations on the input variables through the fuzzy rule base and outputs a highly evaluated confidence level.

[0075] The operation state analysis module collects the command sequence and operation parameters of the elevator control system, establishes a state machine model to identify the transition of the operation mode. The system extracts the height measurement data between two adjacent operations, uses the least squares method to fit the height change trend, and calculates the deviation between the actual displacement and the theoretical displacement. These deviations are converted into corrections for the pressure altitude conversion parameters through compensation coefficients.

[0076] The risk assessment matrix considers three main risk factors: sensor failure probability, environmental interference level, and data anomaly level. The system uses the analytic hierarchy process to determine the weight of each risk factor and establishes a scoring standard through expert experience and historical data analysis. The comprehensive risk score is calculated using the weighted sum method, reflecting the overall risk level of the current measurement environment.

[0077] Based on the confidence level and risk score, the system constructs a reliability assessment function. The assessment process uses a sliding window method to calculate the statistical characteristics and change trends of the data in each assessment window. The adaptive weight algorithm dynamically adjusts the assessment weight according to data quality and environmental conditions to ensure the accuracy of the assessment results.

[0078] The system adopts differentiated processing strategies for data with different reliability levels. High reliability data is directly optimized through the Kalman smoother, medium reliability data requires additional verification steps, and low reliability data may be temporarily blocked or replaced. Finally, the results of each processing link are combined through the Bayesian fusion algorithm to output the optimal height measurement value.

[0079] This technical solution solves the problem of the lack of reliability assessment mechanism in traditional measurement systems. Through multi-dimensional assessment and optimization, the credibility of height measurement results has been significantly improved. The system can effectively identify and process abnormal data, reducing the impact of environmental interference and operational errors on measurement accuracy.

[0080] Test results show that the solution can accurately identify more than 95% of abnormal data and maintain stable and reliable measurement results during 99.9% of the operating time. The system's intelligent evaluation mechanism greatly reduces the need for manual intervention and improves the safety and reliability of elevator operation.

[0081] The advantages of this solution are also reflected in its good adaptability and maintainability. Through the dynamic adjustment of fuzzy rules and the update of risk assessment standards, the system can adapt to different operating environments and usage scenarios, providing reliable technical support for the safe operation of elevators.

[0082] From the above description, it can be seen that the self-calibration method of the air pressure height of the elevator based on the mechanical state provided in the embodiment of the present application can have adaptive calibration capabilities through the intelligent fusion of multi-source data, and can perform real-time reliability evaluation on the measurement results. By introducing technologies such as deep learning and fuzzy reasoning, the accuracy and reliability of the system are improved, thereby better ensuring the safe operation of the elevator.

[0083] In one embodiment of the elevator air pressure height self-correction method based on the mechanical state of the present application, see Figure 2 , and can also include the following:

[0084] Step S201: collecting air pressure data at a sampling frequency of 10 Hz and collecting temperature data and humidity data at a sampling frequency of 1 Hz through an integrated sensor unit arranged in the observation room of the elevator, and performing data standardization processing on the collected air pressure data, the temperature data and the humidity data;

[0085] Step S202: Input the standardized air pressure data, the temperature data and the humidity data into a neural network with three hidden layers, extract temporal features and environmental features through convolution operations, calculate the importance score of each feature based on the attention mechanism, generate a weight coefficient matrix according to the importance score, and perform weighted summation on the extracted features.

[0086] Optionally, in this embodiment, this step implements a high-precision multi-source sensor data acquisition and feature extraction system, and realizes intelligent processing and feature fusion of environmental data through differentiated sampling strategies and deep learning technology. The system installs an integrated sensor unit in the elevator observation room, including a high-precision air pressure sensor, a temperature sensor, and a humidity sensor, and uses different sampling frequencies to monitor environmental parameters.

[0087] The air pressure data is sampled at a high frequency of 10Hz to capture the tiny changes in air pressure during the operation of the elevator and ensure the continuity and integrity of the measurement. The temperature and humidity data are sampled at a frequency of 1Hz because the changes of these two parameters are relatively slow. Low-frequency sampling can not only meet the measurement requirements, but also reduce the burden of data processing. The data of each sensor is pre-processed with an anti-aliasing filter to eliminate high-frequency interference.

[0088] Data normalization uses a piecewise linear mapping method to convert sensor data of different dimensions to the [-1,1] interval. The system first calculates the mean and standard deviation of each type of data within a fixed time window, and then uses these statistical features to normalize the data. The normalization process takes into account the physical meaning of the data and maintains the relative relationship between the data.

[0089] Feature extraction uses an innovative three-layer neural network structure. The first hidden layer uses multiple one-dimensional convolution kernels to process the air pressure data. The size of the convolution kernels ranges from 3 to 7, which can capture temporal features of different scales. The convolution operation is followed by a maximum pooling layer to compress the feature dimension while retaining significant features.

[0090] The second hidden layer mainly processes temperature and humidity data, using a bidirectional long short-term memory network (Bi-LSTM) structure that can simultaneously consider the impact of historical information and future information. The memory mechanism of the LSTM unit effectively captures the long-term dependencies of environmental parameters.

[0091] The third hidden layer implements feature fusion based on the attention mechanism. The attention mechanism calculates the relevance of different feature vectors and task objectives through a trainable query matrix to generate an importance score. The score is normalized by the softmax function and converted into a weight coefficient for weighted combination of features.

[0092] This technical solution solves the problem of single sampling strategy and limited feature extraction capability in traditional data acquisition systems. Through differentiated sampling and deep learning feature extraction, the system can effectively capture the dynamic characteristics of environmental parameters and provide rich feature representation.

[0093] The test results show that the solution has achieved a high level of accuracy and real-time performance in feature extraction. The extraction accuracy of time series features exceeds 98%, and the recognition accuracy of environmental features reaches more than 95%. The system processing delay is controlled within 100ms, meeting the requirements of real-time monitoring.

[0094] Another significant advantage of this solution is its strong adaptability. The attention mechanism can automatically adjust the importance weights of features according to different working conditions, allowing the system to adapt to changes in the environment and working conditions. In addition, the system has good scalability and can easily integrate new sensors and feature extraction modules.

[0095] In one embodiment of the elevator air pressure height self-correction method based on the mechanical state of the present application, see Figure 3 , and can also include the following:

[0096] Step S301: constructing a state vector including the air pressure value and the air pressure change rate based on the weighted summed features, setting the system noise covariance matrix and the measurement noise covariance matrix, calculating the prior estimate of the air pressure state through the state prediction equation, and correcting the state using the measurement update equation;

[0097] Step S302: The state prediction equation and the measurement update equation are combined to construct a state space model, and the model parameters are optimized online using the recursive least squares method. The air pressure state quantity is mapped to an altitude value according to the state space model, and the quantization error of the altitude value is corrected by a nonlinear compensation function to construct a multi-source sensor data fusion model.

[0098] Optionally, in this embodiment, this step implements a high-precision data fusion system based on Kalman filtering and state space modeling, and realizes accurate conversion of air pressure data to altitude information through dynamic state estimation and nonlinear compensation. The system adopts an extended Kalman filter framework, models the air pressure state as a dynamic system, and effectively suppresses measurement noise.

[0099] The state vector is constructed using a two-dimensional structure, which includes two components: air pressure value and air pressure change rate. The air pressure change rate is obtained by differential calculation, and a low-pass filter is used to eliminate high-frequency noise. The system noise covariance matrix determines the initial value by analyzing the statistical characteristics of historical data, and uses an adaptive algorithm to update the covariance parameters online to ensure the stability of the filter.

[0100] The state prediction equation is based on the uniform acceleration motion model and takes into account the continuity of air pressure changes. The prediction process uses the time update equation to calculate the prior estimate of the state and update the state covariance matrix at the same time. The measurement update link uses an innovative sequence method to calculate the optimal Kalman gain and correct the state estimate based on the actual measurement value.

[0101] The system has designed an adaptive mechanism to dynamically adjust the measurement noise covariance matrix. By analyzing the statistical characteristics of the innovation sequence, the system can evaluate the changes in the measurement noise in real time and adjust the filter parameters accordingly. This adaptive mechanism significantly improves the system's ability to adapt to environmental changes.

[0102] The state space model adopts a linear time-varying system structure, and the state transfer matrix and the observation matrix are estimated online by recursive least squares. The estimation process uses an exponential forgetting factor to enable the model to adapt to the slow changes of system parameters. The parameter optimization adopts a batch update strategy to ensure the accuracy of the estimation while ensuring computational efficiency.

[0103] The conversion from pressure to altitude takes into account the nonlinear characteristics of atmospheric pressure changing with altitude. The system uses an improved pressure altitude formula and introduces temperature and humidity correction terms to improve the conversion accuracy. The nonlinear compensation function uses piecewise polynomial fitting and adopts different compensation coefficients for different altitude intervals, which effectively reduces the quantization error.

[0104] Data fusion adopts a hierarchical structure, with the bottom layer completing the fusion of sensor data, the middle layer realizing state estimation and filtering, and the top layer responsible for height calculation and error compensation. The feedback mechanism between each layer maintains the consistency of information, which improves the robustness of the system.

[0105] This technical solution solves the problems of large noise interference, slow dynamic response, unstable accuracy, etc. in traditional pressure altitude measurement. Through multi-source data fusion and dynamic state estimation, the system can provide stable and reliable altitude measurement results.

[0106] Test data shows that the accuracy of the solution is better than ±0.1m in static measurement, and the dynamic measurement error is controlled within the range of ±0.3m. The system has a significant compensation effect on temperature changes and maintains stable measurement accuracy within the temperature range of -20℃ to 50℃.

[0107] The advantages of this solution are also reflected in its computing efficiency and real-time performance. Through the optimized algorithm, the system's data processing delay is controlled within 50ms, meeting the needs of real-time control of the elevator. At the same time, the system has good scalability and can easily integrate other types of sensor data.

[0108] In one embodiment of the elevator air pressure height self-correction method based on the mechanical state of the present application, see Figure 4 , and can also include the following:

[0109] Step S401: Read the standard height reference value stored in the RFID tag of each floor, calculate the height correction coefficient curve between adjacent reference points by using the cubic spline interpolation method, calculate the measurement deviation of each height interval based on the sliding time window, and use the adaptive threshold algorithm to adjust the correction coefficient in sections;

[0110] Step S402: Establish a data sample library containing timestamps, air pressure values, temperature values, humidity values ​​and actual heights, use an incremental learning algorithm to update the statistical distribution parameters of data features, iteratively optimize the values ​​of interval correction coefficients through the gradient descent method, perform exponential smoothing on the optimized correction coefficients, and construct a segmented calibration model.

[0111] Optionally, in this embodiment, this step implements an adaptive height calibration system, and establishes an accurate segmented calibration model through RFID reference points and multi-source data analysis. The system combines fixed reference points and dynamic measurement data to achieve automatic optimization and update of calibration parameters.

[0112] The system first reads the standard height reference values ​​from the RFID tags on each floor, which serve as absolute reference points for calibration. RFID tags use high-reliability memory chips to ensure long-term data stability. The reading process uses an anti-collision algorithm, which can accurately identify the information of each tag even if multiple tags are within the reading range at the same time.

[0113] Cubic spline interpolation uses natural boundary conditions to ensure smooth transition of the correction coefficient curve between adjacent reference points. The selection of interpolation nodes takes into account the non-uniformity of height changes and increases the node density in areas with drastic changes. The coefficients of the spline function are obtained by solving the tridiagonal equations, which ensures the numerical stability of the interpolation.

[0114] The size of the sliding time window is dynamically adjusted according to the operating characteristics of the elevator, and is usually set to 300 to 600 seconds. The data within the window is used to calculate statistical features, including mean, standard deviation, and skewness. The system sets different weights for each statistical feature to comprehensively evaluate the severity of the measurement deviation.

[0115] The adaptive threshold algorithm dynamically adjusts the trigger threshold of the correction factor based on the CUSUM (cumulative sum) criterion. The algorithm takes into account the time-varying characteristics of measurement noise and updates the detection threshold through exponential weighting, which improves the accuracy of anomaly detection. When a significant deviation is detected, the system automatically adjusts the correction factor for the corresponding interval.

[0116] The data sample library adopts a distributed storage structure, supporting real-time writing and fast query. Each record contains complete environmental parameters and time information, and uses timestamp index to improve query efficiency. The system regularly cleans up the sample library, deletes expired data, and maintains database performance.

[0117] The incremental learning algorithm uses an online Bayesian update method to achieve recursive calculation of feature statistics. The algorithm can capture the changing trend of data distribution in a timely manner and provide a basis for the optimization of the correction coefficient. The feature update process uses batch processing to reduce computational overhead.

[0118] The optimization of the correction coefficient uses the stochastic gradient descent method with momentum term to update the parameters by minimizing the sum of squares of the prediction errors. The optimization process introduces an early stopping strategy to prevent overfitting. The exponential smoothing process uses an adjustable smoothing factor to suppress the sharp fluctuations of the parameters while maintaining the responsiveness of the system.

[0119] This technical solution solves the problems of poor adaptability and slow update of traditional calibration methods. Through multi-dimensional data analysis and adaptive optimization, the system can quickly respond to environmental changes and maintain calibration accuracy.

[0120] Test results show that this solution reduces the relative error of height measurement to less than 0.02%, and the absolute error is controlled within the range of ±2mm. The system has a significant compensation effect for temperature changes and maintains a stable calibration effect within the temperature range of -30℃ to 60℃.

[0121] Another advantage of this solution is its low maintenance cost. The adaptive optimization mechanism reduces the need for manual calibration, and the system can autonomously complete parameter adjustment and update. In addition, the modular design of the segmented calibration model makes the system scalable and can be easily adapted to different types of lifts.

[0122] In one embodiment of the elevator air pressure height self-correction method based on the mechanical state of the present application, see Figure 5 , and can also include the following:

[0123] Step S501: define fuzzy input variables including air pressure change rate, temperature change rate and humidity change rate, set membership function parameters and fuzzy rule base, calculate the membership value of the input variable by Mamdani inference algorithm, and determine the confidence level of height assessment according to the fuzzy rule matching degree;

[0124] Step S502: Collect the operation instruction sequence and operation status parameters output by the elevator control system, identify the switching time of the operation mode through the state machine model, extract the altitude measurement data within two adjacent operation times, calculate the altitude offset trend based on least squares fitting, and multiply the calculated offset by the compensation coefficient to update the pressure altitude conversion parameter.

[0125] Optionally, in this embodiment, this step implements an intelligent height assessment system that integrates fuzzy logic and state recognition, and realizes adaptive compensation of height measurement through multi-dimensional environmental parameter analysis and operation status monitoring. The system uses a fuzzy reasoning framework to evaluate measurement reliability and dynamically adjusts conversion parameters based on operation status information.

[0126] The definition of fuzzy input variables uses normalization processing to map the change rates of air pressure, temperature and humidity to the interval [-1,1]. Five fuzzy language values ​​(negative large, negative small, zero, positive small, positive large) are set for each variable, and the Gaussian membership function is used to describe the characteristics of the fuzzy set. The parameters of the membership function are determined by analyzing the statistical characteristics of historical data to ensure the rationality of the fuzzy division.

[0127] The fuzzy rule base contains 125 IF-THEN rules, covering various combinations of environmental parameter changes. The weight coefficients of the rules are determined by combining expert experience and data analysis. Mamdani reasoning uses the maximum-minimum synthesis algorithm to calculate the trigger strength of the rules and output fuzzy sets.

[0128] The calculation of confidence level takes into account two aspects: rule matching and environmental stability. The system uses weighted average method to integrate the inference results of multiple rules and introduces time decay factor to reduce the influence weight of historical data. Confidence level is used to dynamically adjust the credibility of measurement results.

[0129] The acquisition of the operation instruction sequence adopts a real-time communication interface and supports a variety of industrial bus protocols. The system pre-processes the instructions and extracts the key commands related to height control. The operating status parameters include information such as speed, acceleration and door status, which are used to assist in judging the operating mode.

[0130] The state machine model adopts a hierarchical structure, including five basic states: static, start, constant speed, deceleration and stop. The state transition conditions are defined by speed threshold and acceleration characteristics, and hysteresis comparator is used to eliminate jitter. The sliding window method is used to identify the mode switching moment and analyze the change trend of state parameters.

[0131] The extraction of height data takes into account the characteristics of different operating modes. In a stationary state, a longer time window is used to improve stability, and in a moving state, a short time window is used to ensure real-time performance. The system performs outlier detection on the extracted data to eliminate the influence of interfering data.

[0132] The least squares fitting uses a weighted algorithm to assign different weights to data at different times. The fitting process takes into account the statistical characteristics of measurement noise, and the fitting quality is evaluated through residual analysis. The calculation of the offset trend combines short-term and long-term change characteristics to improve the accuracy of trend prediction.

[0133] The compensation coefficient is updated using an adaptive strategy, and the coefficient size is proportional to the confidence level and the offset. Gradient restrictions are introduced in the update process to prevent system instability caused by parameter mutations. The conversion parameters are adjusted using an incremental update method to ensure smooth parameter changes.

[0134] This technical solution solves the problem that traditional height measurement systems have poor adaptability to environmental changes and operating conditions. By combining fuzzy reasoning and state recognition, the system can accurately evaluate measurement reliability and achieve intelligent adjustment of parameters.

[0135] Test data show that this solution improves the accuracy of altitude measurement under dynamic conditions by more than 50%, and the measurement delay is controlled within 20ms. The system has strong adaptability to sudden environmental changes and can complete automatic parameter adjustment within 3 seconds.

[0136] The advantages of this solution are also reflected in its versatility and maintainability. The fuzzy rule base can be flexibly adjusted according to actual application requirements, and the state recognition module supports feature extraction of different types of elevators. The system has a complete parameter backup and recovery mechanism to ensure operational reliability.

[0137] In one embodiment of the elevator air pressure height self-correction method based on the mechanical state of the present application, see Figure 6 , and can also include the following:

[0138] Step S601: constructing a height measurement risk assessment matrix, taking the pressure sensor failure probability, environmental interference degree and data anomaly degree as risk factor inputs, calculating the weight coefficient of each risk factor using the hierarchical analysis method, and obtaining a comprehensive risk score through weighted summation;

[0139] Step S602: A reliability evaluation function is established based on the confidence level and the comprehensive risk score, a reliability threshold and an evaluation period are set, an adaptive weight algorithm is used to dynamically weight the evaluation results, and the height data is graded and smoothed according to the weighted results to construct an intelligent evaluation model.

[0140] Optionally, in this embodiment, this step implements a highly reliable measurement evaluation system based on multi-dimensional risk analysis, and establishes an intelligent evaluation model by comprehensively considering hardware status, environmental factors and data characteristics. The system adopts hierarchical analysis and dynamic weighting methods to achieve reliability quantification and intelligent screening of measurement results.

[0141] The risk assessment matrix adopts a three-dimensional structure, including three dimensions: failure probability, interference degree and abnormality degree. The failure probability is evaluated by parameters such as sensor working time, temperature stress and vibration intensity. The system establishes a sensor performance attenuation model and calculates the failure probability at the current moment in combination with historical failure data.

[0142] The evaluation of environmental interference level comprehensively considers factors such as temperature gradient, air pressure fluctuation and vibration amplitude. The system adopts a multi-level threshold detection method to quantify the interference level into five levels. Each interference factor is set with an independent evaluation index, and the overall interference level is obtained through fuzzy comprehensive evaluation.

[0143] The calculation of data anomaly is based on statistical feature analysis, including three statistics: variance, skewness and kurtosis. The system uses an adaptive window method to calculate these features, and the window size is dynamically adjusted according to the operating status. Anomaly detection uses an improved 3σ criterion, taking into account the non-Gaussian nature of the data.

[0144] The hierarchical analysis process constructs a judgment matrix and calculates the weight vector through the eigenvalue method. The consistency of the judgment matrix is ​​tested through the random consistency ratio to ensure the rationality of the weight allocation. The system supports online fine-tuning of weights and can dynamically optimize weight configuration according to actual operating results.

[0145] The reliability evaluation function uses a nonlinear mapping structure to convert the confidence level and risk score into a reliability index in the interval [0,1]. The function design takes into account the saturation characteristics of the evaluation index and has good stability under extreme conditions. The evaluation cycle is adaptively adjusted according to the elevator operation frequency to ensure the real-time performance of the evaluation.

[0146] The adaptive weight algorithm uses an exponential forgetting mechanism to reduce the impact of historical evaluation results. The algorithm dynamically adjusts the forgetting factor by analyzing the time series characteristics of the evaluation indicators. The weight update uses the gradient descent method to minimize the expected value of the evaluation error.

[0147] The data classification adopts a four-level structure, corresponding to high reliability, medium reliability, low reliability and unreliable levels. Different processing strategies are set for each level. High reliability data is used directly, and data of other levels are smoothed and corrected to varying degrees according to the reliability level.

[0148] The smoothing process uses an adaptive Kalman filter, and the filter gain is proportional to the data reliability. The system uses different noise models for data with different reliability levels to improve the filtering effect. The processing results are verified by residual analysis to ensure the quality of the output data.

[0149] This technical solution solves the problem that traditional height measurement systems lack a reliability assessment mechanism. Through multi-dimensional risk analysis and intelligent assessment, the system can accurately identify and process unreliable measurement data.

[0150] Test results show that this solution enables the reliability assessment accuracy of the measurement system to reach more than 95%, and the false alarm rate is controlled below 1%. The system can effectively identify sensor failures and environmental interference to ensure the reliability of output data.

[0151] The advantage of this solution lies in the transparency and explainability of its evaluation process. The system provides detailed evaluation logs to support the traceability analysis of evaluation results. At the same time, the adaptive characteristics of the model enable it to continuously optimize the evaluation effect and improve the long-term reliability of the system.

[0152] In one embodiment of the elevator air pressure height self-correction method based on the mechanical state of the present application, see Figure 7 , and can also include the following:

[0153] Step S701: Divide the calibrated height data into evaluation windows according to the time sequence, calculate the statistical characteristic values ​​and change trends of the data in the windows, generate evaluation scores according to the reliability evaluation function, and modify the evaluation scores by the adaptive weight algorithm;

[0154] Step S702: normalize the corrected evaluation score, set multi-level reliability thresholds for data classification, use the Kalman smoother to differentiate data of different reliability levels, and combine the processing results through the Bayesian fusion algorithm to output the final height measurement value.

[0155] Optionally, in this embodiment, this step implements a high data evaluation system based on time series analysis and multi-level processing, and realizes reliability grading and optimization processing of measurement results through adaptive scoring and intelligent fusion. The system adopts sliding window analysis and Bayesian reasoning framework to ensure the accuracy and real-time performance of the processing results.

[0156] The evaluation windows are divided in an overlapping sliding manner, and the window length is dynamically adjusted according to the operating characteristics of the elevator. A shorter window (such as 0.5 seconds) is used in the fast movement phase, and a longer window (such as 2 seconds) is used in the static phase. The overlap rate is set to 50% to ensure data continuity and smooth transition.

[0157] The calculation of statistical eigenvalues ​​includes indicators such as mean, standard deviation, skewness, kurtosis and interquartile range. The system uses a recursive algorithm to achieve real-time update of eigenvalues ​​and reduce calculation complexity. Trend analysis uses linear regression and curvature estimation to identify dynamic characteristics of data.

[0158] The reliability evaluation process comprehensively considers the stability, continuity and consistency of the data. The evaluation function uses a sigmoid-type nonlinear mapping to convert multidimensional features into a single score. The function parameters are optimized through machine learning methods to meet the evaluation requirements of different operating conditions.

[0159] The adaptive weight algorithm dynamically adjusts the scoring weight based on data quality and environmental status. The algorithm uses the exponential moving average method to smooth weight changes and prevent drastic fluctuations in the score. The correction process takes into account the reliability of historical evaluation results and achieves progressive optimization of the score.

[0160] Score normalization uses piecewise linear mapping to ensure that the processing results fall within the [0,1] interval. Normalization parameters are adaptively adjusted based on the distribution characteristics of historical data to improve the discrimination of scores. The system sets five reliability levels, and the threshold is determined by cluster analysis.

[0161] The Kalman smoother uses different state models and noise parameters for different reliability levels. High-reliability data uses smaller process noise to maintain the original characteristics of the data; low-reliability data increases the measurement noise and enhances the smoothing effect. The filter parameters are updated online through maximum likelihood estimation.

[0162] Bayesian fusion uses a sequential update method to merge the processing results of multiple levels into the final output. The fusion process takes into account the uncertainty of each processing result and calculates the optimal estimate through conditional probability. The algorithm introduces a forgetting mechanism to reduce the influence weight of historical data.

[0163] Data classification uses fuzzy boundary processing to avoid sudden changes when switching levels. The system uses weighted average method for data in the boundary area to achieve smooth transition. The classification results pass the confidence test to ensure the reliability of the classification.

[0164] This technical solution solves the problems of inaccurate reliability assessment and single processing method in the data processing link of traditional height measurement systems. Through multi-level evaluation and processing mechanisms, the system can perform intelligent processing based on data characteristics.

[0165] Experimental results show that this solution significantly improves the data quality of the measurement system and reduces the noise level by more than 40%, while maintaining the ability to respond quickly to real height changes. The system shows good stability and reliability under various working conditions.

[0166] Another advantage of this solution is the scalability of its processing framework. The system supports the dynamic access of new evaluation indicators and processing algorithms, which can continuously optimize and improve the processing effect. At the same time, the perfect data recording mechanism facilitates subsequent analysis and system optimization.

[0167] In order to effectively improve the accuracy and efficiency of the self-correction of the air pressure height of the elevator based on the mechanical state, the present application provides an embodiment of the self-correction device of the air pressure height of the elevator based on the mechanical state for realizing all or part of the contents of the self-correction method of the air pressure height of the elevator based on the mechanical state, see Figure 8 The mechanical state-based self-correction device for the air pressure height of the elevator specifically includes the following contents:

[0168] The initial height determination module 10 is used to collect the air pressure data of the air pressure sensor, the temperature data of the temperature sensor and the humidity data of the humidity sensor provided on the elevator, perform feature extraction on the air pressure data, the temperature data and the humidity data using a deep learning model, construct an adaptive weight matrix to dynamically fuse the extracted features, input the fused data into a Kalman filter model for noise reduction processing to obtain smoothed air pressure height data, and construct a multi-source sensor data fusion model; based on the multi-source sensor data fusion model, process the sensor data collected in real time to obtain initial height data;

[0169] The height data calibration module 20 is used to obtain the front and rear gantry switch status signals of the elevator and the RFID reference point information preset on each floor, and reset the error coefficient to zero when the front gantry status change is detected, so that the height of the elevator automatically returns to zero when it reaches the first floor, dynamically update the correction coefficient for different height intervals based on the RFID reference point information, and continuously optimize the correction coefficient by analyzing the historical operation data using an online learning mechanism to construct a segmented calibration model; calibrate the initial height data based on the segmented calibration model to obtain calibrated height data;

[0170] The final height measurement module 30 is used to establish a height assessment confidence calculation rule based on the fuzzy reasoning system, obtain the elevator control state information, calculate the error value generated between two adjacent operation times, add the error value to the pressure altitude conversion coefficient, and quantitatively evaluate the reliability of the height data in combination with the risk assessment results to construct an intelligent evaluation model; based on the intelligent evaluation model, the calibrated height data is reliability evaluated and optimized to obtain the elevator height measurement result.

[0171] From the above description, it can be seen that the mechanical state-based self-calibration device for the air pressure height of an elevator provided in the embodiment of the present application can have adaptive calibration capabilities through intelligent fusion of multi-source data, and can perform real-time reliability evaluation on the measurement results. By introducing technologies such as deep learning and fuzzy reasoning, the accuracy and reliability of the system are improved, thereby better ensuring the safety of elevator operation.

[0172] From the hardware level, in order to effectively improve the accuracy and efficiency of the self-calibration of the air pressure height of the elevator based on the mechanical state, the present application provides an embodiment of an electronic device for implementing all or part of the contents of the self-calibration method of the air pressure height of the elevator based on the mechanical state, and the electronic device specifically includes the following contents:

[0173] Processor, memory, communication interface and bus; wherein the processor, memory and communication interface communicate with each other through the bus; the communication interface is used to realize information transmission between the self-calibration device for the air pressure height of the elevator based on the mechanical state and related equipment such as the core business system, user terminal and related database; the logic controller can be a desktop computer, a tablet computer and a mobile terminal, etc., but the present embodiment is not limited thereto. In the present embodiment, the logic controller can be implemented with reference to the embodiment of the self-calibration method for the air pressure height of the elevator based on the mechanical state and the embodiment of the self-calibration device for the air pressure height of the elevator based on the mechanical state, and the contents thereof are incorporated herein, and the repeated parts are not repeated.

[0174] It is understandable that the user terminal may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

[0175] In practical applications, part of the self-correction method of the air pressure height of the elevator based on the mechanical state can be executed on the electronic device side as described above, or all operations can be completed in the client device. The specific selection can be based on the processing capability of the client device and the limitations of the user's usage scenario. This application does not limit this. If all operations are completed in the client device, the client device may also include a processor.

[0176] The client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side, and other implementation scenarios may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, or a server cluster consisting of multiple servers, or a server structure of a distributed device.

[0177] Fig. 9 FIG. 9 is a schematic block diagram of the system structure of the electronic device 9600 according to an embodiment of the present application. Fig. 9 As shown, the electronic device 9600 may include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. It is worth noting that Fig. 9 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0178] In one embodiment, the function of the self-correction method of the air pressure height of the elevator based on the mechanical state can be integrated into the central processor 9100. The central processor 9100 can be configured to perform the following control:

[0179] Step S101: collecting air pressure data of an air pressure sensor, temperature data of a temperature sensor, and humidity data of a humidity sensor provided on the elevator, extracting features of the air pressure data, the temperature data, and the humidity data using a deep learning model, constructing an adaptive weight matrix to dynamically fuse the extracted features, inputting the fused data into a Kalman filter model for noise reduction processing to obtain smoothed air pressure height data, and constructing a multi-source sensor data fusion model; processing the sensor data collected in real time based on the multi-source sensor data fusion model to obtain initial height data;

[0180] Step S102: obtaining the front and rear gantry switch status signals of the elevator and the RFID reference point information preset on each floor, resetting the error coefficient to zero when detecting the change of the front gantry status, so that the height of the elevator automatically returns to zero when it reaches the first floor, dynamically updating the correction coefficient for different height intervals based on the RFID reference point information, and continuously optimizing the correction coefficient by analyzing the historical operation data using an online learning mechanism, and constructing a segmented calibration model; calibrating the initial height data based on the segmented calibration model to obtain calibrated height data;

[0181] Step S103: Establish a height assessment confidence calculation rule based on the fuzzy inference system, obtain the elevator control status information, calculate the error value generated between two adjacent operation times, add the error value to the pressure altitude conversion coefficient, and quantitatively evaluate the reliability of the altitude data in combination with the risk assessment results to construct an intelligent evaluation model; perform reliability evaluation and optimization processing on the calibrated altitude data based on the intelligent evaluation model to obtain the elevator altitude measurement result.

[0182] From the above description, it can be seen that the electronic device provided in the embodiment of the present application has adaptive calibration capability through intelligent fusion of multi-source data, and can perform real-time reliability evaluation on the measurement results. By introducing technologies such as deep learning and fuzzy reasoning, the accuracy and reliability of the system are improved, thereby better ensuring the safe operation of the elevator.

[0183] In another embodiment, the elevator air pressure altitude self-correction device based on the mechanical state can be configured separately from the central processing unit 9100. For example, the elevator air pressure altitude self-correction device based on the mechanical state can be configured as a chip connected to the central processing unit 9100, and the function of the elevator air pressure altitude self-correction method based on the mechanical state can be realized through the control of the central processing unit.

[0184] like Fig. 9 As shown, the electronic device 9600 may also include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily have to include Fig. 9 In addition, the electronic device 9600 may also include Fig. 9 For components not shown, reference may be made to the prior art.

[0185] like Fig. 9 As shown, the central processing unit 9100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor device and / or logic device, which receives input and controls the operation of various components of the electronic device 9600.

[0186] The memory 9140 may be, for example, one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory or other suitable devices. The above-mentioned information related to the failure may be stored, and a program for executing the relevant information may also be stored. The CPU 9100 may execute the program stored in the memory 9140 to implement information storage or processing, etc.

[0187] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display may be, for example, an LCD display, but is not limited thereto.

[0188] The memory 9140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that saves information even when the power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROMs, etc. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142, which is used to store application programs and function programs or processes for executing the operation of the electronic device 9600 through the central processor 9100.

[0189] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0190] The communication module 9110 is a transmitter / receiver that sends and receives signals via the antenna 9111. The communication module 9110 (transmitter / receiver) is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

[0191] Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and / or a wireless LAN module. The communication module 9110 (transmitter / receiver) is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby realizing a common telecommunication function. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processor 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

[0192] The embodiments of the present application also provide a computer-readable storage medium capable of implementing all the steps of the method for self-calibrating the air pressure height of an elevator based on the mechanical state in the above-mentioned embodiment, where the execution subject is a server or a client. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, all the steps of the method for self-calibrating the air pressure height of an elevator based on the mechanical state in the above-mentioned embodiment are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0193] Step S101: collecting air pressure data of an air pressure sensor, temperature data of a temperature sensor, and humidity data of a humidity sensor provided on the elevator, extracting features of the air pressure data, the temperature data, and the humidity data using a deep learning model, constructing an adaptive weight matrix to dynamically fuse the extracted features, inputting the fused data into a Kalman filter model for noise reduction processing to obtain smoothed air pressure height data, and constructing a multi-source sensor data fusion model; processing the sensor data collected in real time based on the multi-source sensor data fusion model to obtain initial height data;

[0194] Step S102: obtaining the front and rear gantry switch status signals of the elevator and the RFID reference point information preset on each floor, resetting the error coefficient to zero when detecting the change of the front gantry status, so that the height of the elevator automatically returns to zero when it reaches the first floor, dynamically updating the correction coefficient for different height intervals based on the RFID reference point information, and continuously optimizing the correction coefficient by analyzing the historical operation data using an online learning mechanism, and constructing a segmented calibration model; calibrating the initial height data based on the segmented calibration model to obtain calibrated height data;

[0195] Step S103: Establish a height assessment confidence calculation rule based on the fuzzy inference system, obtain the elevator control status information, calculate the error value generated between two adjacent operation times, add the error value to the pressure altitude conversion coefficient, and quantitatively evaluate the reliability of the altitude data in combination with the risk assessment results to construct an intelligent evaluation model; perform reliability evaluation and optimization processing on the calibrated altitude data based on the intelligent evaluation model to obtain the elevator altitude measurement result.

[0196] From the above description, it can be seen that the computer-readable storage medium provided in the embodiment of the present application has adaptive calibration capability through intelligent fusion of multi-source data, and can perform real-time reliability evaluation on the measurement results. By introducing technologies such as deep learning and fuzzy reasoning, the accuracy and reliability of the system are improved, thereby better ensuring the safe operation of the elevator.

[0197] The embodiments of the present application also provide a computer program product capable of implementing all the steps of the method for self-calibrating the air pressure height of an elevator based on the mechanical state in the above embodiments, where the execution subject is a server or a client. When the computer program / instruction is executed by a processor, the steps of the method for self-calibrating the air pressure height of an elevator based on the mechanical state are implemented. For example, the computer program / instruction implements the following steps:

[0198] Step S101: collecting air pressure data of an air pressure sensor, temperature data of a temperature sensor, and humidity data of a humidity sensor provided on the elevator, extracting features of the air pressure data, the temperature data, and the humidity data using a deep learning model, constructing an adaptive weight matrix to dynamically fuse the extracted features, inputting the fused data into a Kalman filter model for noise reduction processing to obtain smoothed air pressure height data, and constructing a multi-source sensor data fusion model; processing the sensor data collected in real time based on the multi-source sensor data fusion model to obtain initial height data;

[0199] Step S102: obtaining the front and rear gantry switch status signals of the elevator and the RFID reference point information preset on each floor, resetting the error coefficient to zero when detecting the change of the front gantry status, so that the height of the elevator automatically returns to zero when it reaches the first floor, dynamically updating the correction coefficient for different height intervals based on the RFID reference point information, and continuously optimizing the correction coefficient by analyzing the historical operation data using an online learning mechanism, and constructing a segmented calibration model; calibrating the initial height data based on the segmented calibration model to obtain calibrated height data;

[0200] Step S103: Establish a height assessment confidence calculation rule based on the fuzzy inference system, obtain the elevator control status information, calculate the error value generated between two adjacent operation times, add the error value to the pressure altitude conversion coefficient, and quantitatively evaluate the reliability of the altitude data in combination with the risk assessment results to construct an intelligent evaluation model; perform reliability evaluation and optimization processing on the calibrated altitude data based on the intelligent evaluation model to obtain the elevator altitude measurement result.

[0201] From the above description, it can be seen that the computer program product provided in the embodiment of the present application has adaptive calibration capability through intelligent fusion of multi-source data, and can perform real-time reliability evaluation on the measurement results. By introducing technologies such as deep learning and fuzzy reasoning, the accuracy and reliability of the system are improved, thereby better ensuring the safe operation of the elevator.

[0202] It should be understood by those skilled in the art that embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0203] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (apparatus), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0204] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0205] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0206] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for self-calibration of air pressure height of an elevator based on mechanical state, characterized in that: The method comprises: Collect air pressure data from an air pressure sensor, temperature data from a temperature sensor, and humidity data from a humidity sensor installed on the elevator, use a deep learning model to extract features from the air pressure data, the temperature data, and the humidity data, build an adaptive weight matrix to dynamically fuse the extracted features, input the fused data into a Kalman filter model for noise reduction processing to obtain smoothed air pressure height data, and build a multi-source sensor data fusion model; process the sensor data collected in real time based on the multi-source sensor data fusion model to obtain initial height data; Obtain the front and rear gantry switch status signals of the elevator and the RFID reference point information preset on each floor, reset the error coefficient of the height measurement to zero when the front gantry opening signal is detected, so that the height of the elevator automatically returns to zero when it reaches the first floor, dynamically update the correction coefficient for different height intervals based on the RFID reference point information, and use the online learning mechanism to analyze the historical operation data to continuously optimize the correction coefficient, and construct a segmented calibration model; calibrate the initial height data based on the segmented calibration model to obtain the calibrated height data; Based on the fuzzy inference system, a height assessment confidence calculation rule is established to obtain the elevator control state information, and the error value of the height measurement data generated when the elevator control system outputs two adjacent operation instructions is calculated, specifically including: defining fuzzy input variables including the air pressure change rate, temperature change rate and humidity change rate, setting the membership function parameters and the fuzzy rule base, calculating the membership value of the input variable by the Mamdani inference algorithm, and determining the confidence level of the height assessment according to the fuzzy rule matching degree; collecting the operation instruction sequence and operation state parameters output by the elevator control system, identifying the switching time of the operation mode by the state machine model, extracting the height measurement data within two adjacent operation times, calculating the height offset trend based on the least squares fitting, and multiplying the calculated offset by the compensation coefficient to update the air pressure height conversion parameter; The error value is added to the pressure altitude conversion coefficient, and the reliability of the altitude data is quantitatively evaluated in combination with the risk assessment result to construct an intelligent assessment model; based on the intelligent assessment model, the calibrated altitude data is reliability evaluated and optimized to obtain the elevator altitude measurement result.

2. The method for self-calibration of air pressure height of an elevator based on mechanical state according to claim 1, characterized in that: The method collects air pressure data of an air pressure sensor, temperature data of a temperature sensor, and humidity data of a humidity sensor provided on the elevator, extracts features of the air pressure data, the temperature data, and the humidity data using a deep learning model, and constructs an adaptive weight matrix to dynamically fuse the extracted features, including: The integrated sensor unit disposed in the observation room of the elevator collects air pressure data at a sampling frequency of 10 Hz, and collects temperature data and humidity data at a sampling frequency of 1 Hz, and performs data standardization processing on the collected air pressure data, the temperature data and the humidity data; The standardized air pressure data, the temperature data and the humidity data are input into a neural network with three hidden layers, and the time series features and environmental features are extracted through convolution operations. The importance score of each feature is calculated based on the attention mechanism, and a weight coefficient matrix is ​​generated according to the importance score to perform weighted summation on the extracted features.

3. The method for self-correcting the air pressure height of an elevator based on mechanical state according to claim 1, characterized in that: The fused data is input into the Kalman filter model for noise reduction to obtain smoothed pressure altitude data, and a multi-source sensor data fusion model is constructed, including: Based on the weighted sum of the features, a state vector including the air pressure value and the air pressure change rate is constructed, the system noise covariance matrix and the measurement noise covariance matrix are set, the prior estimate of the air pressure state is calculated through the state prediction equation, and the state is corrected using the measurement update equation; The state prediction equation and the measurement update equation are combined to construct a state space model, and the model parameters are optimized online using the recursive least squares method. The air pressure state quantity is mapped to an altitude value according to the state space model, and the quantization error of the altitude value is corrected by a nonlinear compensation function to construct a multi-source sensor data fusion model.

4. The method for self-calibration of air pressure height of an elevator based on mechanical state according to claim 1, characterized in that: The correction coefficients for different height intervals are dynamically updated based on the RFID reference point information, and the correction coefficients are continuously optimized by analyzing the historical operation data using an online learning mechanism to construct a segmented calibration model, including: Read the standard height reference value stored in the RFID tag of each floor, calculate the height correction coefficient curve between adjacent reference points by using the cubic spline interpolation method, calculate the measurement deviation of each height interval based on the sliding time window, and use the adaptive threshold algorithm to adjust the correction coefficient in sections; A data sample library containing timestamps, air pressure values, temperature values, humidity values ​​and actual altitude is established. The incremental learning algorithm is used to update the statistical distribution parameters of the data characteristics. The value of the interval correction coefficient is iteratively optimized through the gradient descent method. The optimized correction coefficient is exponentially smoothed to construct a segmented calibration model.

5. The method for self-calibration of air pressure height of an elevator based on mechanical state according to claim 1, characterized in that: The reliability of the height data is quantitatively evaluated in combination with the risk assessment results to construct an intelligent assessment model, including: A height measurement risk assessment matrix was constructed, and the probability of air pressure sensor failure, environmental interference degree, and data anomaly degree were input as risk factors. The weight coefficient of each risk factor was calculated using the analytic hierarchy process, and a comprehensive risk score was obtained through weighted summation. A reliability evaluation function is established based on the confidence level and the comprehensive risk score, a reliability threshold and an evaluation period are set, an adaptive weight algorithm is used to dynamically weight the evaluation results, and the height data is graded and smoothed according to the weighted results to construct an intelligent evaluation model.

6. The method for self-calibration of air pressure height of an elevator based on mechanical state according to claim 1, characterized in that: The reliability evaluation and optimization processing of the calibrated height data based on the intelligent evaluation model to obtain the elevator height measurement result includes: Dividing the calibrated height data into evaluation windows according to time series, calculating the statistical characteristic values ​​and change trends of the data in the windows, generating evaluation scores according to the reliability evaluation function, and correcting the evaluation scores by an adaptive weight algorithm; The corrected evaluation scores are normalized, multi-level reliability thresholds are set for data classification, data with different reliability levels are differentiated using a Kalman smoother, and the processing results are combined using a Bayesian fusion algorithm to output a final height measurement value.

7. A self-correcting device for air pressure height of an elevator based on mechanical state, characterized in that: The device comprises: An initial height determination module is used to collect air pressure data of an air pressure sensor, temperature data of a temperature sensor, and humidity data of a humidity sensor provided on the elevator, perform feature extraction on the air pressure data, the temperature data, and the humidity data using a deep learning model, construct an adaptive weight matrix to dynamically fuse the extracted features, input the fused data into a Kalman filter model for noise reduction processing to obtain smoothed air pressure height data, and construct a multi-source sensor data fusion model; based on the multi-source sensor data fusion model, the sensor data collected in real time is processed to obtain initial height data; A height data calibration module is used to obtain the front and rear gantry switch status signals of the elevator and the RFID reference point information preset on each floor. When the front gantry opening signal is detected, the error coefficient of the height measurement is reset to zero, so that the height of the elevator automatically returns to zero when it reaches the first floor. The correction coefficient is dynamically updated for different height intervals based on the RFID reference point information, and the correction coefficient is continuously optimized by analyzing the historical operation data using an online learning mechanism to construct a segmented calibration model; the initial height data is calibrated based on the segmented calibration model to obtain calibrated height data; The final height measurement module is used to establish a height assessment confidence calculation rule based on a fuzzy reasoning system, obtain the elevator control state information, and calculate the error value of the height measurement data generated when the elevator control system outputs two adjacent operation instructions. Specifically, it includes: defining fuzzy input variables including air pressure change rate, temperature change rate and humidity change rate, setting membership function parameters and fuzzy rule base, calculating the membership value of the input variable by the Mamdani reasoning algorithm, and determining the confidence level of the height assessment according to the fuzzy rule matching degree; collecting the operation instruction sequence and operation state parameters output by the elevator control system, identifying the switching time of the operation mode by the state machine model, extracting the height measurement data within two adjacent operation times, calculating the height offset trend based on least squares fitting, and updating the pressure height conversion parameter by multiplying the calculated offset by the compensation coefficient; superimposing the error value on the pressure height conversion coefficient, and quantitatively evaluating the reliability of the height data in combination with the risk assessment result to construct an intelligent evaluation model; performing reliability evaluation and optimization processing on the calibrated height data based on the intelligent evaluation model to obtain the elevator height measurement result.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the elevator air pressure altitude self-correction method based on mechanical state as described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the self-correction method of elevator air pressure altitude based on mechanical state as described in any one of claims 1 to 6 are implemented.

Images