Elevator car distance measuring method and device based on ultrasonic positioning, medium and equipment

Abstract

The invention discloses an elevator car distance measuring method and device based on ultrasonic positioning, a medium and equipment, and the method comprises the steps that ultrasonic positioning is conducted on an elevator car and a well wall to form a directional ultrasonic signal; collecting information of the surrounding environment of the elevator car in real time, adjusting the propagation speed of the directional ultrasonic signal, and correcting the position information of the elevator car; constructing an ultrasonic echo analysis model to analyze an ultrasonic echo signal formed after the directional ultrasonic signal is reflected by the well wall; based on the ultrasonic echo signals analyzed by the ultrasonic echo analysis model, the initial distance between the elevator car and the well wall is obtained through calculation; and integrating the propagation speed of the adjusted directional ultrasonic signal, the initial distance between the elevator car and the well wall and the corrected position information of the elevator car to obtain the final distance between the elevator car and the well wall. The range finding precision of the elevator car can be improved.

Description

Technical Field

[0001] This application belongs to the technical field of elevator control, and specifically relates to an elevator car ranging method, device, medium and equipment based on ultrasonic positioning. Background Art

[0002] Traditional ultrasonic ranging technology has various limitations in the elevator shaft environment: First, the change of temperature and humidity in the shaft will cause the drift of the sound speed, further affecting the measurement accuracy; moreover, the dense metal structures in the elevator shaft will generate strong electromagnetic interference, interfering with the ultrasonic signal; finally, when the car is moving at high speed, the dynamic error will gradually accumulate, reducing the reliability of the ranging result. Summary of the Invention

[0003] Aiming at the deficiencies in the prior art, the main purpose of this application is to provide an elevator car ranging method, device, medium and equipment based on ultrasonic positioning, aiming to improve the accuracy and stability of elevator car ranging.

[0004] To achieve the above objectives, this application provides the following technical solutions:

[0005] An elevator car ranging method based on ultrasonic positioning, the method includes: performing ultrasonic positioning on the elevator car and the shaft wall to form a directional ultrasonic signal; collecting the surrounding environment information of the elevator car in real time and adjusting the propagation speed of the directional ultrasonic signal based on the surrounding environment information, and at the same time, correcting the position information of the elevator car; constructing an ultrasonic echo analysis model to analyze the ultrasonic echo signal formed after the directional ultrasonic signal is reflected by the shaft wall to obtain an estimated value of the elevator car position information; based on the estimated value of the elevator car position information output by the ultrasonic echo analysis model, and combining the adjusted propagation speed of the directional ultrasonic signal and the time interval from the emission of the directional ultrasonic signal to the reception of the ultrasonic echo signal, calculating to obtain the initial distance between the elevator car and the shaft wall; integrating the adjusted propagation speed of the directional ultrasonic signal, the initial distance between the elevator car and the shaft wall, and the corrected position information of the elevator car to obtain the final distance between the elevator car and the shaft wall.

[0006] Optionally, the adjusting the propagation speed of the directional ultrasonic signal based on the surrounding environment information includes: adjusting the propagation speed of the directional ultrasonic signal by constructing a sound speed compensation model, and the sound speed compensation model is expressed as:

[0007] v = 331.4 + 0.6T + 0.0124H - 0.0037P

[0008] Among them, 331.4 represents the speed of sound value in air under standard conditions, with the unit of m / s; 0.6 represents the speed at which the speed of sound increases per 1°C increase, with the unit of m / s; 0.0124 represents the speed at which the speed of sound increases per 1% increase, with the unit of m / s; 0.0037 represents the speed at which the speed of sound decreases per 1 hPa increase in air pressure, with the unit of m / s; v represents the propagation speed of the directional ultrasonic signal adjusted by the speed of sound compensation model; T represents the temperature of the elevator car surrounding environment; H represents the humidity of the elevator car surrounding environment; P represents the air pressure of the elevator car surrounding environment.

[0009] Optionally, the ultrasonic echo analysis model includes, connected in sequence: a multi-modal fusion input layer, a multi-path perception convolution module, a structure-guided spatio-temporal attention module, a dynamic state adjustment layer, and an adaptive multi-task output layer. Among them, the multi-modal fusion input layer is used to preprocess the ultrasonic echo signal, and to map the elevator car surrounding environment information into an environmental embedding vector, and splice it with the preprocessed ultrasonic echo signal to form a composite feature vector; the multi-path perception convolution module is used to perform multi-scale feature extraction on the composite feature vector; the structure-guided spatio-temporal attention module is used to identify and capture key features in the multi-scale features; the dynamic state adjustment layer is used to perform time series analysis on the key features captured by the structure-guided spatio-temporal attention module to obtain an estimated value of the elevator car position information; the adaptive multi-task output layer is used to output the estimated value of the elevator car position information.

[0010] Optionally, the multi-modal fusion input layer includes a preprocessing module, an embedding layer, and a splicing layer. Among them, the preprocessing module is used to preprocess the ultrasonic echo signal; the embedding layer is used to map the elevator car surrounding environment information into an environmental embedding vector; the splicing layer is used to splice the preprocessed ultrasonic echo signal with the environmental embedding vector.

[0011] Optionally, the multi-path perception convolution module includes: multiple path branches, and the path branches are used to perform multi-scale feature extraction on the composite feature vector through different paths.

[0012] Optionally, the structure-guided spatio-temporal attention module includes: a shaft structure-guided position encoding and a spatio-temporal cross-attention mechanism. Among them, the shaft structure-guided position encoding is used to generate a position encoding matrix based on shaft geometric parameters; the spatio-temporal cross-attention mechanism is used to capture the dynamic change features of the composite feature vector in time and space.

[0013] Optionally, the ultrasonic echo analysis model is trained through the following steps: collect ultrasonic echo signals generated during the operation of the elevator car and their corresponding labels to form a data set, and divide the data set into a training set and a validation set; set model training parameters, and train the model through the training set until the maximum number of training times is reached; verify the trained model through the validation set, and use accuracy, precision, recall rate, and F1 score as performance indicators to evaluate the model. When each performance indicator reaches the threshold, the model verification passes; otherwise, adjust the model training parameters to retrain the model until the model verification passes.

[0014] This application also provides an elevator car ranging device based on ultrasonic positioning. The device includes: a positioning module for performing ultrasonic positioning on the elevator car and the shaft wall to form a directional ultrasonic signal; a collection module for real-time collecting the surrounding environment information of the elevator car and adjusting the propagation speed of the directional ultrasonic signal based on the surrounding environment information. At the same time, correct the position information of the elevator car; an analysis module for constructing an ultrasonic echo analysis model to analyze the ultrasonic echo signal formed after the directional ultrasonic signal is reflected by the shaft wall to obtain an estimated value of the elevator car position information; a first calculation module for based on the estimated value of the elevator car position information output by the ultrasonic echo analysis model, and combining the adjusted propagation speed of the directional ultrasonic signal and the time interval from the emission of the directional ultrasonic signal to the reception of the ultrasonic echo signal, calculate the initial distance between the elevator car and the shaft wall; a second calculation module for integrating the adjusted propagation speed of the directional ultrasonic signal, the initial distance between the elevator car and the shaft wall, and the corrected position information of the elevator car to obtain the final distance between the elevator car and the shaft wall.

[0015] This application also provides a storage medium, which includes instructions that, when run on a computer, cause the computer to execute the method described in any of the previous items.

[0016] This application also provides an electronic device. The electronic device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor. Wherein, when the processor executes the program, it implements the method described in any of the previous items.

[0017] This application can bring the following beneficial effects:

[0018] This application can not only adjust the ultrasonic propagation speed in real time to compensate for the influence of environmental factors on the sound speed, but also effectively identify and suppress multipath interference and other electromagnetic interference caused by complex structures in the shaft. In addition, by constructing an ultrasonic echo analysis model, this application can significantly improve the accuracy and stability of measuring the distance between the elevator car and the shaft under different operating conditions, providing a strong guarantee for the safe operation of the elevator. Description of the Drawings

[0019] Figure 1 is a schematic flowchart of a method for measuring the distance of an elevator car based on ultrasonic positioning provided by an embodiment of the present application;

[0020] Figure 2 is a schematic structural diagram of an ultrasonic echo analysis model provided by another embodiment of the present application;

[0021] Figure 3 is a schematic structural diagram of an intelligent distance measuring device for an elevator car provided by another embodiment of the present application;

[0022] Figure 4 is a schematic structural diagram of a storage medium provided by another embodiment of the present application;

[0023] Figure 5 is a schematic structural diagram of an electronic device provided by another embodiment of the present application. Detailed Embodiments

[0024] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0025] It should be noted that all directional indications (such as up, down, left, right, front, back...) in the embodiments of the present application are only used to explain the relative positional relationship and movement conditions between components in a specific posture (as shown in the drawings). If the specific posture changes, the directional indications will also change accordingly.

[0026] In the present application, unless otherwise clearly specified and limited, terms such as "connection" and "fixation" should be understood in a broad sense. For example, "fixation" can be a fixed connection, a detachable connection, or integrated; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal connection of two components or the interaction relationship between two components, unless otherwise clearly limited. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific situations.

[0027] In addition, if the descriptions such as "first" and "second" are involved in the embodiments of the present application, the descriptions of "first", "second", etc. are only for descriptive purposes and cannot be construed as indicating or implying their relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one such feature. In addition, the meaning of "and / or" appearing throughout the text includes three parallel scenarios. Taking "A and / or B" as an example, it includes Scenario A, or Scenario B, or the scenario where both A and B are satisfied simultaneously. In addition, the technical solutions between various embodiments can be combined with each other, but it must be based on what can be achieved by those of ordinary skill in the art. When the combination of technical solutions results in contradictions or cannot be achieved, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present application.

[0028] Figure 1 is a schematic flow chart of an elevator car ranging method based on ultrasonic positioning provided by an exemplary embodiment of the present application. As Figure 1 shown, the ranging method includes the following steps:

[0029] S100: Install ultrasonic sensor arrays at the top and bottom of the elevator car, and use the FPGA to analyze the structural characteristics of the hoistway wall in real time to dynamically adjust the emission angle of the ultrasonic sensor arrays, so as to form a directional ultrasonic signal pointing to the hoistway wall.

[0030] Exemplarily, the elevator hoistway has a standard rectangular cross-section with a width of 2.5 m and a depth of 3 m. There is a metal bracket 1.5 m in front of the hoistway, and its height covers half of the hoistway height and its width is 0.5 m. The FPGA can analyze the structural characteristics of the hoistway in real time, identify the metal bracket, and adjust the emission angle of the ultrasonic sensor arrays so that the emitted ultrasonic waves bypass the area of the metal bracket, or adjust the emission angle to point to the unobstructed area above or below the metal bracket for measurement. For example, the emission angle of the ultrasonic sensor arrays can be tilted downward by 10° to avoid the metal bracket and measure the distance between the elevator car and the bottom hoistway wall.

[0031] S200: Real-time collect the ambient data around the elevator car through temperature, humidity, and air pressure sensors to establish a sound speed compensation model. The ambient data includes, for example, temperature, humidity, and air pressure data; adjust the propagation speed of the directional ultrasonic signal according to the sound speed compensation model. At the same time, obtain the angular velocity data and linear acceleration data of the elevator car based on the MEMS gyroscope and accelerometer respectively, and continuously correct the position information of the elevator car through the Kalman filter to reduce the accumulation of dynamic errors during the high-speed movement of the elevator car, thereby improving the ranging accuracy between the elevator car and the hoistway.

[0032] In this step, the sound speed compensation model is expressed as follows:

[0033] v = 331.4 + 0.6T + 0.0124H - 0.0037P

[0034] Wherein, 331.4 represents the speed of sound value in air under standard conditions (0°C, sea level, relative humidity of 0%), with the unit of m / s (m / s); 0.6 represents the speed at which the speed of sound increases per 1°C increase, that is, within a given temperature range, for every 1°C increase in temperature, the speed of sound approximately increases by 0.6 m / s; 0.0124 represents the influence of humidity on the speed of sound, that is, at a certain temperature, for every 1% increase in relative humidity, the speed of sound increases by 0.0124 m / s; 0.0037 represents that for every 1 hPa increase in air pressure, the speed of sound decreases by approximately 0.0037 m / s; v represents the speed of sound; T represents the temperature of the environment surrounding the elevator car; H represents the humidity of the environment surrounding the elevator car; P represents the air pressure of the environment surrounding the elevator car.

[0035] Exemplarily, the temperature, humidity, and air pressure data obtained based on the sensors are respectively:

[0036] Temperature: 25°C; Humidity: 60%; Air pressure: 1013 hPa.

[0037] Then, based on the speed of sound compensation model shown above, the adjusted propagation speed of the directional ultrasonic signal can be obtained as:

[0038] v = 331.4 + 0.6×25 + 0.0124×60 + 0.0035×1013 = 350.7 m / s

[0039] Next, the angular velocity and linear acceleration of the elevator car obtained by using the MEMS gyroscope and accelerometer respectively are:

[0040] Angular velocity: w x = 0.01 rad / s, w y = 0.02 rad / s, w z = 0 rad / s

[0041] Linear acceleration: a x = 0.1 m / s 2 , a y = 0 m / s 2 , a z = -9.8 m / s 2

[0042] Finally, use the Kalman filter to correct the position information. Assuming the initial state of the elevator car is stationary, the initial position is p0 = 0 m, and the initial speed is v0 = 0 m / s, then after Kalman filtering combined with the above data, the corrected position of the elevator car is: p k= -4.9m + correction amount, where the correction amount = (measured value - predicted value) × Kalman gain. For example, assuming the measured value represents the position where the elevator car should be (e.g., if the elevator is supposed to stop at a certain floor, and that floor is 1.5m forward relative to the current predicted position), then: correction amount = 0.909×(1.5 - (-4.9)) = 5.8176m, that is, the position of the elevator car should be corrected upward by 5.8176m.

[0043] It should be noted that the above data are only exemplary, aiming to show how to adjust the propagation speed of the directional ultrasonic signal according to environmental data, and how to correct the position information of the elevator car according to the angular velocity and linear acceleration.

[0044] S300: Construct an ultrasonic echo analysis model to analyze the ultrasonic echo signal formed after the directional ultrasonic signal is reflected by the shaft wall, so as to obtain an estimated value of the elevator car position information;

[0045] S400: Based on the estimated value of the elevator car position information output by the ultrasonic echo analysis model, and in combination with the adjusted propagation speed of the directional ultrasonic signal and the time interval from the emission of the directional ultrasonic signal to the reception of the ultrasonic echo signal, calculate the initial distance between the elevator car and the shaft wall;

[0046] S500: Integrate the adjusted propagation speed of the directional ultrasonic signal, the initial distance between the elevator car and the shaft wall, and the corrected position information of the elevator car to obtain the final distance between the elevator car and the shaft wall.

[0047] In this application, by real-time monitoring of the environmental data around the elevator car and dynamically adjusting the propagation speed value of the directional ultrasonic signal according to the sound speed compensation model, the actual propagation speed of the directional ultrasonic signal under the current environmental conditions can be accurately reflected. In this way, when calculating the distance between the elevator car and the shaft, it can be carried out based on the sound speed closer to the actual situation, so as to effectively compensate for the sound speed drift of the ultrasonic echo caused by environmental condition changes, and thus improve the accuracy of distance measurement.

[0048] In another exemplary embodiment, as Figure 2 shown, the ultrasonic echo analysis model includes, connected in sequence: a multi-modal fusion input layer, a multi-path perception convolution module, a structure-guided spatio-temporal attention module, a dynamic state adjustment layer, and an adaptive multi-task output layer.

[0049] In this embodiment, the multimodal fusion input layer includes a preprocessing module, an embedding layer, and a splicing layer. Among them, the preprocessing module is used to preprocess the ultrasonic echo signal (including operations such as filtering and normalization). The embedding layer is used to map the environmental information around the elevator car (such as the temperature, humidity, and air pressure information collected in real time) into an environmental embedding vector through a lightweight MLP (Multi-Layer Perceptron). The splicing layer is used to splice the preprocessed ultrasonic echo signal and the environmental embedding vector. Since the ultrasonic echo signal often carries information about physical distance (such as the distance between the elevator car and the shaft wall) and obstacles, while the environmental embedding vector can provide information affecting the propagation conditions of the ultrasonic echo signal (such as temperature, humidity, etc.), through splicing, the ultrasonic echo analysis model can not only understand the distance and obstacle information directly extracted from the ultrasonic echo, but also take into account factors such as the sound speed drift caused by changes in environmental conditions, so as to obtain a more comprehensive and richer composite feature vector, which helps to improve the accuracy of elevator car ranging. Exemplarily, for a given environmental condition (25°C, 60%, 1013 hPa), the environmental embedding vector obtained after passing through the MLP is a 32-dimensional real number vector, for example, it can be expressed as: Environmental embedding vector = [0.2, -0.5,..., 0.8]. The length of the ultrasonic echo signal is 100 points, and after preprocessing, a 100-dimensional vector is obtained, for example, it can be expressed as: Ultrasonic echo signal = [0.1, 0.4,..., 0.7]. After splicing the environmental embedding vector and the ultrasonic echo signal along the dimension direction, a 132-dimensional composite feature vector is formed, for example, it can be expressed as:

[0050] Composite feature vector = [0.2, -0.5,..., 0.8, 0.1, 0.4,..., 0.7].

[0051] The multi-path perception convolution module adopts a multi-branch convolution design, including a short-path branch, a long-path branch, and a frequency-domain branch arranged in parallel. Among them, the short-path branch includes a 3×3 convolution layer, a batch normalization layer, and a Swish activation function connected in sequence. By using a smaller receptive field (3×3 convolution layer), the short-path branch can focus on capturing local detailed features in the composite feature vector (such as strong echo features like metal structure reflections within a short distance); the long-path branch includes a dilated convolution, a batch normalization layer, and a Swish activation function connected in sequence. By using dilated convolution, the long-path branch can expand the receptive field without increasing the number of parameters, thereby capturing more extensive context information in the composite feature vector (such as long-range dependencies caused by multiple reflections or multi-path effects); the frequency-domain branch first performs a Fourier transform on the composite feature vector to obtain a frequency-domain representation, and then analyzes the frequency components through a complex convolution layer, thereby extracting interference features in the frequency domain (such as irregularities or abnormal fluctuations in the frequency response caused by external electromagnetic interference). Frequency-domain analysis can help identify and suppress these interferences, thereby improving the accuracy of the ranging result. Due to the complex and variable environment in the shaft, a single type of convolution kernel cannot effectively handle all situations. The multi-path perception convolution module adopts a multi-branch design, enabling the ultrasonic echo analysis model to understand and process ultrasonic echo signals from multiple perspectives, thereby enhancing the resistance to various interference factors. In addition, the short-path branch captures local details, the long-path branch focuses on global context, and the frequency-domain branch provides interference features from the frequency-domain perspective. Fusing these different levels of features can enable the model to obtain a more comprehensive and rich feature representation, thereby improving the overall performance. For example, when a metal bracket is detected in the shaft, the short-path branch can quickly locate the obstacle, and the long-path branch can consider its impact on the surrounding environment, thereby obtaining a more accurate ranging result.

[0052] Further, the multi-path perception convolution module introduces a dynamic feature fusion mechanism to adaptively weight and fuse the outputs of the short-path branch, the long-path branch, and the frequency-domain branch. The dynamic feature fusion mechanism is specifically expressed as follows:

[0053] α = σ(W g ·Concat(F short , F long , F freq ) + b g )

[0054] F fused = α·F short + (1 - α)·(β·F long + (1 - β)·F freq )

[0055] Among them, α represents the gating coefficient, and α ∈ [0, 1], σ represents the Sigmoid function, and W g represents the learnable weight matrix, which is used to perform a linear transformation on the concatenated features. Concat represents the concatenation operation, which is used to concatenate the output feature F short of the short-path branch, the output feature F long of the long-path branch, and the output feature F freq of the frequency-domain branch along the channel dimension. b g represents the bias, which is used to adjust the output offset of the gating attention, and β represents the learnable parameter.

[0056] Through adaptive weighted fusion, the model can flexibly adjust the attention to the outputs of the above branches according to the specific requirements in different scenarios, so as to highlight the key features and reduce the influence of redundant features.

[0057] The structure-guided spatio-temporal attention module includes a shaft structure-guided position encoding and a spatio-temporal cross-attention mechanism. Among them, the shaft structure-guided position encoding can generate a position encoding matrix based on shaft geometric parameters (such as reflector distance, angle) to enhance the model's perception ability of the physical space, which helps the model to identify and capture the key features in the multi-scale features output by the multi-path perception convolution module.

[0058] Exemplarily, as described above, the elevator shaft has a standard rectangular cross-section with a width of 2.5 m and a depth of 3 m. The distance from the front-end sensor to the shaft wall is 1.5 m. The ultrasonic beams emitted by the sensor array are fan-shaped and cover an angular range of ±15 degrees (i.e., a total of 30 degrees). Next, the present application will describe in detail how to generate the position encoding matrix.

[0059] Step 1: Divide the angular range into 10 equally spaced intervals (each interval is 3 degrees), and also divide it into several segments along the distance direction. For example, select every 0.1 m as a distance segment, then there are 16 segments (including the starting point) in the distance range from 0 m to 1.5 m.

[0060] Step 2: For each grid point, calculate its position encoding value relative to the origin (i.e., the position of the sensor). For example, use trigonometric functions to calculate the x and y coordinates of each point:

[0061] x = d · cos(θ)

[0062] y = d · sin(θ)

[0063] where d represents the distance and θ represents the angle.

[0064] Then, use these coordinate values as part of the position encoding. For example, apply sine or cosine functions to increase the richness of the encoding:

[0065] For even-indexed dimensions: Use the sine function

[0066] For odd-indexed dimensions: Use the cosine function

[0067] Step 3: Based on the above calculations, a position encoding matrix can be constructed, denoted as:

[0068]

[0069] In the above matrix, each cell represents a position encoding value at a specific distance and angle (such as (x, y) coordinates). In practical applications, these values will be converted into a form suitable for neural network processing, such as encoding relative position information through sine / cosine transformation.

[0070] Once the position encoding matrix is generated, it can be fed into the subsequent spatio-temporal cross-attention mechanism to help the model better understand the dynamic change characteristics of the ultrasonic echo signal in time and space, thereby more accurately locating the key reflection regions in the wellbore and improving the ranging accuracy.

[0071] The spatio-temporal cross-attention mechanism includes time-axis attention and space-axis attention. The time-axis attention is used to further analyze the dynamic changes of the ultrasonic echo signal over time (such as the position movement or speed change of the elevator car) based on the weighted and fused composite feature vector. The space-axis attention combines the position encoding matrix to focus on the key reflection regions within the wellbore wall (such as strong reflection points in the wellbore) to capture the dynamic changes of the ultrasonic echo signal over space (such as reflection characteristics at different positions). The spatio-temporal cross-attention mechanism also introduces a gating mechanism to fuse the outputs of the time attention and space attention, enabling the model to not only identify the trends of the ultrasonic echo signal that change over time but also accurately locate the key reflection regions within the wellbore, thereby providing more accurate ranging results.

[0072] The dynamic state adjustment layer includes a filtering and fusion layer and a bidirectional LSTM layer. Among them, the filtering and fusion layer is used to filter and fuse the dynamic error correction parameters (such as speed, acceleration) output by the Kalman filter as the input of the bidirectional LSTM layer, ensuring that the bidirectional LSTM layer can not only utilize the time series characteristics of the ultrasonic echo signal but also combine accurate speed and acceleration information to estimate the position information of the elevator car.

[0073] In addition to receiving the dynamically error-corrected parameters after filtering and fusion, the bidirectional LSTM layer also receives the output from the spatio-temporal cross-attention mechanism. By processing time series data in both forward and backward directions, the bidirectional LSTM layer can capture long-term dependencies in the received data. By combining the velocity and acceleration information obtained from the filtering and fusion layer, the bidirectional LSTM layer can more accurately estimate the position of the elevator car based on the output of the spatio-temporal cross-attention mechanism.

[0074] Furthermore, the bidirectional LSTM layer also introduces a causal convolutional layer (the causal convolutional layer is set before the bidirectional LSTM layer) to preprocess the output of the Kalman filter (the causal convolutional layer performs a convolution operation on time series data through a sliding window, and each output only depends on the data points at the current moment and before, ensuring that the time order is not disrupted), so as to enhance the response ability to the mutation signals therein.

[0075] The adaptive multi-task output layer includes an interference classification branch and a distance correction regression branch. Among them, the interference classification branch is used to classify the key features obtained from the bidirectional LSTM layer (including the estimated value of the elevator car's position information and possible interference signals (such as multipath reflection, temperature and humidity changes, electromagnetic interference, etc.)), and output the probability distribution of the interference type. The interference classification branch uses a dynamic weight cross-entropy loss function, which is specifically expressed as follows:

[0076]

[0077] where L cls represents the classification loss, N represents the number of samples, w i represents the weight factor of the i-th type of interference, y i represents the true label of the i-th sample, represents the probability distribution of the interference type predicted by the model.

[0078] Through the dynamic weight cross-entropy loss function, the interference classification branch can assign higher weights to key interference types (such as multipath reflection), so as to adaptively adjust the echo detection threshold, enabling the model to more accurately identify the main interference sources that have a greater impact on the ranging accuracy in a complex environment, thereby improving the accuracy of interference classification.

[0079] Exemplarily, assume that the elevator car receives ultrasonic echo signals during operation. Among them, the environmental parameters include: temperature of 25 °C, relative humidity of 60%, and air pressure of 1013 hPa. After being detected by the model, the interference signals included in the ultrasonic echo signals can be identified as: multipath reflection (intensity level: high), sound speed drift caused by temperature change (influence degree: medium), and electromagnetic interference (intensity level: low). First, set a basic echo detection threshold, such as -40 dB. Among them, multipath reflection may cause false echoes, so it is necessary to increase the detection threshold to filter out weaker but possibly false echo signals. Assume that each time a high-intensity multipath reflection is encountered, the threshold is increased by 5 dB. In addition, although there is electromagnetic interference, due to its low intensity, the impact on echo detection is small, and a slight increase, such as 1 dB, can be considered to further reduce potential risks.

[0080] Based on the above adjustments, the final echo detection threshold = basic threshold + multipath reflection + electromagnetic interference = -40 dB + 5 dB + 1 dB = -34 dB.

[0081] It can be seen that in order to cope with multipath reflection and electromagnetic interference in the current environment, the echo detection threshold is increased from -40 dB to -34 dB. In this way, it can help the model ignore the weak signals caused by interference and below the new threshold, thereby improving the accuracy and reliability of ranging.

[0082] The distance correction regression branch is used to correct the estimated value of the elevator car position information output by the bidirectional LSTM layer to obtain a more accurate estimated value. The distance correction regression branch adopts the following loss function:

[0083]

[0084] where L reg represents the regression loss, M represents the number of data points, represents the estimated value of the elevator car position information predicted by the model for the jth data point, d j represents the corresponding true distance value, and j represents the index of the data point.

[0085] The adaptive multi-task output layer further includes a shared fusion layer. The shared fusion layer takes the outputs of the above two branches as inputs, enabling the corrected estimated value output by the distance correction regression branch to be further dynamically adjusted according to the interference classification result output by the interference classification branch, so as to further improve the accuracy of the estimated value. For example, in the case of high humidity or strong electromagnetic interference, if these factors are not considered, the estimated value may still have a large deviation even after correction; by setting the shared fusion layer, these errors can be reduced by adjusting the algorithm parameters or directly correcting the predicted value. The shared fusion layer is expressed as:

[0086] L fusion = ∑(F - T) 2

[0087] Wherein, L fusion represents the output of the shared fusion layer, F represents the new feature representation output by the shared fusion layer, and T represents the target vector or scalar determined according to the specific application scenario, for example, it can be a more accurate estimated value adjusted by combining interference information.

[0088] Then the total loss function of the adaptive multi-task output layer is expressed as:

[0089] L total = L cls + λ1·L res + λ2·L fusion

[0090] Wherein, L total represents the total loss function, and λ1 and λ2 both represent weight coefficients.

[0091] By introducing the weight coefficients λ1 and λ2, the model can flexibly adjust the importance of each sub-task, which helps to improve the ranging accuracy of the elevator car.

[0092] In another exemplary embodiment, in step S400, the present application calculates the initial distance between the elevator car and the hoistway wall through the following formula:

[0093]

[0094] Wherein, v represents the adjusted ultrasonic wave propagation speed; t represents the time interval from the ultrasonic wave emission to the reception of the echo; α represents the weight coefficient; represents the estimated value of the elevator car position output by the ultrasonic echo analysis model.

[0095] In another exemplary embodiment, the present application integrates the adjusted propagation speed of the directional ultrasonic signal, the initial distance between the elevator car and the hoistway wall, and the corrected position information of the elevator car through the following formula to obtain the final distance between the elevator car and the hoistway wall:

[0096] d final = w d ·d 初始 + w x ·x corr + w v ·β·v adj

[0097] Wherein, w d represents the weight of the initial distance d 初始 ; w xRepresents the corrected position information x of the elevator car corr weight of; w v Represents the propagation speed v of the adjusted directional ultrasonic signal adj weight of; β represents the scaling factor of the propagation speed of the adjusted directional ultrasonic signal, with the unit of s / m.

[0098] In another exemplary embodiment, the ultrasonic echo analysis model is trained through the following steps:

[0099] Collect ultrasonic echo signals generated during the operation of the elevator car and their corresponding labels (such as different interference types) to form a data set, and divide the data set into a training set and a validation set. For example, the division ratio can be 7:3;

[0100] Set the model training parameters. For example, set the batch size to 32 and the learning rate to 0.001, and train the model through the training set until the maximum number of training times is reached;

[0101] Validate the trained model through the validation set, and use accuracy, precision, recall, and F1 score as performance indicators to evaluate the model. When each performance indicator reaches 90%, the model validation passes; otherwise, adjust the model training parameters (such as adjusting the batch size to 64 or adjusting the learning rate to 0.0005) to retrain the model until the model validation passes.

[0102] In this embodiment, considering that the operating environment of the elevator car is complex and changeable, the influence degrees of different types of interference signals (such as interference caused by multipath effects, sound speed drift caused by changes in temperature and humidity in the hoistway, etc.) on the system performance may be different. Therefore, a weight factor can be assigned to each interference type to reflect its relative importance. Specifically, this adaptive weighted cross-entropy loss function can be defined as follows:

[0103]

[0104] Among them, N represents the number of samples; represents the number of categories, that is, the number of interference types; w c represents the weight factor of the c-th type of interference; represents the true label of the i-th sample for the c-th category; is the probability that the model predicts the i-th sample input to the c-th category.

[0105] In the cross-entropy loss function shown above, by assigning different weight factors to different types of interference signals, it can be ensured that the model pays more attention to those types of interference that have a significant impact on ranging accuracy during the training process. For example, in a hoistway environment, certain specific types of interference (such as interference caused by multipath effects) may cause greater errors in the final distance measurement results than other types of interference. By assigning higher weights to these key types of interference, the model can learn more effective feature representations, thereby improving its overall robustness. In addition, since the design of this loss function takes into account the influence degrees of different types of interference and optimizes the learning process of the model accordingly, it helps to improve the final distance measurement accuracy.

[0106] Next, for the sake of easy understanding, the present application exemplarily illustrates the method described in the present application in combination with the data described above.

[0107] The elevator hoistway has a standard rectangular cross-section, with a width of 2.5 m and a depth of 3 m. There is a metal bracket 1.5 m in front, whose height covers half of the hoistway height and width is 0.5 m. The environmental conditions are as follows:

[0108] Temperature: 25 °C

[0109] Humidity: 60%

[0110] Atmospheric pressure: 1013 hPa

[0111] The angular velocity of the elevator car is 0.05 rad / s, and the linear acceleration is 0.1 m / s 2

[0112] The sampling rate of the ultrasonic echo signal is 1 MHz

[0113] The time interval from the emission of the ultrasonic signal to the reception of the echo is measured as t = 0.0088 s

[0114] As shown above, the propagation speed of the adjusted directional ultrasonic signal is v = 350.7 m / s

[0115] Input the above data into the ultrasonic echo analysis model, and the specific processing process of the model for the data is as follows:

[0116] 1. Multi-path convolution output:

[0117] Short-path features: [0.12, -0.05,..., 0.08] (32 dimensions)

[0118] Long-path features: [0.30, 0.02,...,-0.15] (32 dimensions)

[0119] Frequency domain features: [0.05 + 0.02j,..., 0.10 - 0.03j] (32 dimensions)

[0120] 2. Dynamic fusion:

[0121] Gating weights = [0.4, 0.3, 0.3]

[0122] Fused features = 0.4 * short path + 0.3 * long path + 0.3 * frequency domain

[0123] 3. Initial ranging value:

[0124] d raw = (350.7 m / s × 0.0088 s) / 2 = 1.543 m

[0125] 4. Multipath error estimation:

[0126] Secondary echo delay time: 0.0092 s

[0127] Distance deviation caused by multipath: Δd = (0.0092 - 0.0088) / 2 × 350.7 = 0.07 m. Among them, the model weight is 0.85 (from interference classification), then the correction amount is: 0.07 × 0.85 = 0.06 m.

[0128] 5. Motion compensation:

[0129] 0.1 m / s 2 × 0.0088 s = 0.00088 m. The corrected value after Kalman filtering is -0.04 m.

[0130] 6. Environmental compensation:

[0131] The remaining deviation is -0.01 m

[0132] 7. The estimated value of the elevator car position information obtained after comprehensive correction is:

[0133]

[0134] Use the formula to calculate the initial distance between the elevator car and the hoistway wall. Since the ultrasonic wave travels back and forth once, the total distance needs to be divided by 2 to obtain the one-way distance. Substitute the known values into the formula, and we can get:

[0135] Among them, 0.5 is the weight coefficient, that is, α = 0.5. After rounding, the preliminary distance between the elevator car and the hoistway wall is 1.53 m.

[0136] Furthermore, based on d final = w d ·d 初始 + wx ·x corr +w v ·β·v adj Integrating the propagation speed of the adjusted directional ultrasonic signal, the initial distance between the elevator car and the hoistway wall, and the corrected position information of the elevator car, the final distance between the elevator car and the hoistway wall can be obtained as 1.52 m.

[0137] Taking the distance measured by a high-precision measuring instrument as the reference value, the ranging results obtained based on this method and the ranging results obtained based on the traditional method are shown in Table 1:

[0138] Table 1

[0139]

[0140] As can be seen from Table 1, this application takes into account the changes in environmental parameters, dynamically adjusts the sound speed, and reduces the influence of multipath effects and other interferences through advanced signal processing techniques. The measured distance is 1.52 m, with an error of +0.02 m compared with the reference value, that is, the error rate is +0.67%. Although there are certain small deviations, generally it is very close to the actual value. For the traditional ultrasonic ranging method, since a fixed sound speed value is used and the changes in environmental factors or multipath effects are not considered, the measured distance in the same environment is 1.45 m, with an error of -0.05 m compared with the reference value, that is, the error rate is -3.33%. Based on the above results, it can be shown that the method described in this application can improve the ranging accuracy between the elevator car and the hoistway wall compared with the traditional method.

[0141] In another exemplary embodiment, this application also provides an intelligent ranging device for an elevator car, such as Figure 3As shown, the device includes: a positioning module 100 for performing ultrasonic positioning on the elevator car and the hoistway wall to form a directional ultrasonic signal; a collection module 200 for collecting real-time environmental information around the elevator car and adjusting the propagation speed of the directional ultrasonic signal based on the environmental information, and at the same time, correcting the position information of the elevator car; an analysis module 300 for constructing an ultrasonic echo analysis model to analyze the ultrasonic echo signal formed after the directional ultrasonic signal is reflected by the hoistway wall to obtain an estimated value of the elevator car position information; a first calculation module 400 for calculating the initial distance between the elevator car and the hoistway wall based on the estimated value of the elevator car position information output by the ultrasonic echo analysis model, in combination with the adjusted propagation speed of the directional ultrasonic signal and the time interval from the emission of the directional ultrasonic signal to the reception of the ultrasonic echo signal; a second calculation module 500 for integrating the adjusted propagation speed of the directional ultrasonic signal, the initial distance between the elevator car and the hoistway wall, and the corrected position information of the elevator car to obtain the final distance between the elevator car and the hoistway wall.

[0142] Based on the above embodiments, with reference to Figure 4 , a computer-readable storage medium of an exemplary embodiment of the present application will be described. Please refer to Figure 4 , which shows that the computer-readable storage medium is an optical disc 40, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it will implement the steps recorded in the above method embodiments. For example, performing ultrasonic positioning on the elevator car and the hoistway wall to form a directional ultrasonic signal; collecting real-time environmental information around the elevator car and adjusting the propagation speed of the directional ultrasonic signal based on the environmental information, and at the same time, correcting the position information of the elevator car; constructing an ultrasonic echo analysis model to analyze the ultrasonic echo signal formed after the directional ultrasonic signal is reflected by the hoistway wall to obtain an estimated value of the elevator car position information; calculating the initial distance between the elevator car and the hoistway wall based on the estimated value of the elevator car position information output by the ultrasonic echo analysis model, in combination with the adjusted propagation speed of the directional ultrasonic signal and the time interval from the emission of the directional ultrasonic signal to the reception of the ultrasonic echo signal; integrating the adjusted propagation speed of the directional ultrasonic signal, the initial distance between the elevator car and the hoistway wall, and the corrected position information of the elevator car to obtain the final distance between the elevator car and the hoistway wall. The specific implementation methods of each step will not be repeated here.

[0143] It should be noted that the computer-readable storage medium includes, but is not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory, or other optical and magnetic storage media, which will not be elaborated here one by one.

[0144] Based on the above embodiments, the present application further provides an electronic device. The following refers to Figure 5 to describe the electronic device for file download according to the exemplary embodiments of the present application.

[0145] Figure 5 The block diagram of an exemplary electronic device 50 suitable for implementing the embodiments of the present application is shown. The electronic device 50 may be a computer system or a cloud server. Figure 5 The shown electronic device 50 is only an example and should not impose any limitations on the functions and usage scopes of the embodiments of the present application.

[0146] As Figure 5 shown, the electronic device 50 includes, but is not limited to: one or more processors or processing units 501, a system memory 502, and a bus 503 connecting different system components (including the system memory 502 and the processing unit 501).

[0147] The electronic device 50 typically includes various computer system-readable media. These media can be any available media accessible by the electronic device 50, including volatile and non-volatile media, removable and non-removable media.

[0148] The system memory 502 may include computer system-readable media in the form of volatile memory, such as random access memory (RAM) 5021 and / or cache memory 5022. The electronic device 50 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, ROM 5023 can be used to read and write non-removable, non-volatile magnetic media ( Figure 5 not shown in the figure, usually referred to as a "hard disk drive"). Although not shown in Figure 5As shown, a disk drive for reading and writing to a removable non-volatile disk (such as a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (such as a CD-ROM, DVD-ROM, or other optical medium) can be provided. In these cases, each drive can be connected to the bus 503 through one or more data medium interfaces. The system memory 502 may include at least one program product having a set (such as at least one) of program modules configured to perform the functions of the embodiments of the present application.

[0149] A program / utility 5025 having a set (at least one) of program modules 5024 can be stored, for example, in the system memory 502, and such program modules 5024 include but are not limited to: an operating system, one or more application programs, other program modules, and program data, and the implementation of a network environment may be included in each or some combination of these examples. The program modules 5024 generally perform the functions and / or methods in the embodiments described in the present application.

[0150] The electronic device 50 can also communicate with one or more external devices 504 (such as a keyboard, pointing device, display, etc.). Such communication can be carried out through the input / output (I / O) interface 505. Also, the electronic device 50 can further communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through the network adapter 506. As Figure 5 shown, the network adapter 506 communicates with other modules (such as the processing unit 501, etc.) of the electronic device 50 through the bus 503. It should be understood that although Figure 5 not shown in the figure, other hardware and / or software modules can be used in combination with the electronic device 50.

[0151] The processing unit 501 executes various functional applications and data processing by running the programs stored in the system memory 502. For example, it performs ultrasonic positioning on the elevator car and the hoistway wall to form a directional ultrasonic signal; it collects the ambient information around the elevator car in real time and adjusts the propagation speed of the directional ultrasonic signal based on the ambient information. At the same time, it corrects the position information of the elevator car; it constructs an ultrasonic echo analysis model to analyze the ultrasonic echo signal formed after the directional ultrasonic signal is reflected by the hoistway wall to obtain an estimated value of the elevator car position information; based on the estimated value of the elevator car position information output by the ultrasonic echo analysis model, and in combination with the adjusted propagation speed of the directional ultrasonic signal and the time interval from the emission of the directional ultrasonic signal to the reception of the ultrasonic echo signal, it calculates the initial distance between the elevator car and the hoistway wall; it integrates the adjusted propagation speed of the directional ultrasonic signal, the initial distance between the elevator car and the hoistway wall, and the corrected position information of the elevator car to obtain the final distance between the elevator car and the hoistway wall. The specific implementation methods of each step will not be repeated here. It should be noted that although several units / modules or sub-units / sub-modules of the file concurrent download device are mentioned in the above detailed description, this division is only exemplary and not mandatory. In fact, according to the embodiments of the present application, the features and functions of the two or more units / modules described above can be embodied in one unit / modules. Conversely, the features and functions of one unit / modules described above can be further divided into being embodied by multiple units / modules.

[0152] In the description of the present application, it should be noted that the terms "first", "second", and "third" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0153] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated here.

[0154] In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. Also, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some communication interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0155] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0156] In addition, each functional unit in various embodiments of the present application can be integrated in a processing unit, can also be physically present separately for each unit, or two or more units can be integrated in one unit.

[0157] If the above function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a cloud server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs, and other various media that can store program codes.

[0158] The above embodiments are only for illustrating the technical concept and features of the present application, and the purpose is to enable those who are familiar with this technology to understand the content of the present application and implement it accordingly, and it cannot be used to limit the protection scope of the present application. All equivalent changes or modifications made according to the spirit and essence of the present application should be covered within the protection scope of the present application.

Claims

1. An elevator car ranging method based on ultrasonic positioning, characterized in that, The method includes: Performing ultrasonic positioning on the elevator car and the hoistway wall to form a directional ultrasonic signal; Collecting the environmental information around the elevator car in real time and adjusting the propagation speed of the directional ultrasonic signal based on the surrounding environmental information. Meanwhile, correcting the position information of the elevator car; Constructing an ultrasonic echo analysis model to analyze the ultrasonic echo signal formed after the directional ultrasonic signal is reflected by the hoistway wall, so as to obtain an estimated value of the elevator car position information; Based on the estimated value of the elevator car position information output by the ultrasonic echo analysis model, and combining the adjusted propagation speed of the directional ultrasonic signal and the time interval from the emission of the directional ultrasonic signal to the reception of the ultrasonic echo signal, calculating to obtain the initial distance between the elevator car and the hoistway wall; Integrating the adjusted propagation speed of the directional ultrasonic signal, the initial distance between the elevator car and the hoistway wall, and the corrected position information of the elevator car to obtain the final distance between the elevator car and the hoistway wall.

2. The elevator car ranging method based on ultrasonic positioning according to claim 1, wherein The adjusting the propagation speed of the directional ultrasonic signal based on the surrounding environmental information includes: Adjusting the propagation speed of the directional ultrasonic signal by constructing a sound speed compensation model, and the sound speed compensation model is expressed as: v = 331.4 + 0.6T + 0.0124H - 0.0037P Wherein, 331.4 represents the sound speed value in air under standard conditions, with the unit of m / s; 0.6 represents the speed of sound increase per 1°C increase, with the unit of m / s; 0.0124 represents the speed of sound increase per 1% increase, with the unit of m / s; 0.0037 represents the speed of sound decrease per 1 hPa increase in air pressure, with the unit of m / s; v represents the propagation speed of the directional ultrasonic signal adjusted by the sound speed compensation model; T represents the temperature of the environment around the elevator car; H represents the humidity of the environment around the elevator car; P represents the air pressure of the environment around the elevator car.

3. The elevator car ranging method based on ultrasonic positioning according to claim 1, wherein The ultrasonic echo analysis model includes, connected in sequence: A multi-modal fusion input layer, a multi-path perception convolution module, a structure-guided spatio-temporal attention module, a dynamic state adjustment layer, and an adaptive multi-task output layer, wherein, The multi-modal fusion input layer is used for preprocessing the ultrasonic echo signal, and for mapping the environmental information around the elevator car into an environmental embedding vector, and splicing it with the preprocessed ultrasonic echo signal to form a composite feature vector; The multi-path perception convolution module is used for extracting multi-scale features from the composite feature vector; The structure-guided spatio-temporal attention module is used for identifying and capturing key features in the multi-scale features; The dynamic state adjustment layer is used for performing time series analysis on the key features captured by the structure-guided spatio-temporal attention module to obtain an estimated value of the elevator car position information; The adaptive multi-task output layer is used for outputting the estimated value of the elevator car position information.

4. The elevator car ranging method based on ultrasonic positioning according to claim 3, wherein, The multi-modal fusion input layer includes a preprocessing module, an embedding layer, and a splicing layer, wherein, The preprocessing module is used for preprocessing the ultrasonic echo signal; The embedding layer is used for mapping the environmental information around the elevator car into an environmental embedding vector; The splicing layer is used to splice the preprocessed ultrasonic echo signal and the environmental embedding vector.

5. The elevator car ranging method based on ultrasonic positioning according to claim 3, characterized in that, The multi-path perception convolutional module includes: Multiple path branches, which are used to perform multi-scale feature extraction on the composite feature vector through different paths.

6. The elevator car ranging method based on ultrasonic positioning according to claim 3, characterized in that The structure-guided spatio-temporal attention module includes: A position encoding and a spatio-temporal cross-attention mechanism guided by the shaft structure, where The position encoding guided by the shaft structure is used to generate a position encoding matrix based on the shaft geometric parameters; The spatio-temporal cross-attention mechanism is used to capture the dynamic change features of the composite feature vector in time and space.

7. The elevator car ranging method based on ultrasonic positioning according to claim 1, characterized in that, The ultrasonic echo analysis model is trained through the following steps: Collect ultrasonic echo signals generated during the operation of the elevator car and their corresponding labels to form a data set, and divide the data set into a training set and a validation set; Set the model training parameters, and train the model through the training set until the maximum number of training times is reached; Validate the trained model through the validation set, and use accuracy, precision, recall rate, and F1 score as performance indicators to evaluate the model. When each performance indicator reaches the threshold, the model is successfully validated; otherwise, adjust the model training parameters to retrain the model until the model is successfully validated.

8. An elevator car ranging device based on ultrasonic positioning, characterized in that, The device includes: A positioning module, which is used to perform ultrasonic positioning on the elevator car and the shaft wall to form a directional ultrasonic signal; An acquisition module, which is used to collect the environmental information around the elevator car in real time and adjust the propagation speed of the directional ultrasonic signal based on the surrounding environmental information. At the same time, the position information of the elevator car is corrected; An analysis module, which is used to build an ultrasonic echo analysis model to analyze the ultrasonic echo signal formed after the directional ultrasonic signal is reflected by the shaft wall to obtain an estimated value of the elevator car position information; A first calculation module, which is used to calculate the initial distance between the elevator car and the shaft wall based on the estimated value of the elevator car position information output by the ultrasonic echo analysis model, combined with the adjusted propagation speed of the directional ultrasonic signal and the time interval from the emission of the directional ultrasonic signal to the reception of the ultrasonic echo signal; A second calculation module, which is used to integrate the adjusted propagation speed of the directional ultrasonic signal, the initial distance between the elevator car and the shaft wall, and the corrected position information of the elevator car to obtain the final distance between the elevator car and the shaft wall.

9. A storage medium, characterized in that, It includes instructions that, when running on a computer, cause the computer to execute the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, The electronic device includes: A memory, a processor, and a computer program stored on the memory and executable on the processor, where When the processor executes the program, it implements the method according to any one of claims 1 to 7.

Images