Indoor positioning and temperature anomaly collaborative alarm method and system

Abstract

The invention discloses an indoor positioning and temperature anomaly collaborative alarm method and system, and the method comprises the steps: constructing a multi-dimensional data collaborative decision model through fusing living body perception, crawling ladder vibration signal feature analysis and temperature compensation personnel positioning technologies, and carrying out the collaborative alarm of the indoor positioning and temperature anomaly in combination with a safety threshold and a multi-stage braking mechanism. The problems of positioning drift and false triggering of a single sensor caused by a high-temperature environment are effectively solved, the personnel position detection precision and the abnormal behavior recognition accuracy are remarkably improved, meanwhile, the false alarm rate is reduced, the real-time performance of braking response is optimized, and finally precise safety protection of industrial equipment in a high-risk operation scene is achieved.

Description

Technical Field

[0001] The present invention belongs to the field of industrial automation, and in particular relates to a method and system for collaborative alarming of indoor positioning and temperature anomaly. Background Art

[0002] In industries like cement and chemicals, operating safety within high-temperature, enclosed environments inside equipment presents significant challenges. Traditional personnel positioning technologies rely primarily on UWB or RFID systems. However, in temperature gradient fields, electromagnetic wave signals drift due to thermal refraction, leading to inaccurate safety distance calculations.

[0003] The existing technologies have the following problems: (1) electromagnetic refraction caused by high temperature fields causes positioning drift, and the error exceeds the safety threshold; (2) the fixed threshold alarm mechanism cannot coordinate with the dynamic temperature field and changes in personnel behavior.

[0004] The above problems cause existing technologies to have defects such as frequent false alarms and braking delays in high-temperature industrial scenarios, which will seriously threaten the safety of personnel and equipment.

[0005] Therefore, how to design an intelligent, efficient, accurate and reliable indoor positioning and temperature anomaly collaborative alarm method is a technical problem to be solved. Summary of the Invention

[0006] Based on this, it is necessary to provide a method and system for collaborative alarming of indoor positioning and temperature anomaly to address the problems of the existing technology.

[0007] In a first aspect, an embodiment of the present application provides a method for collaboratively warning of indoor positioning and temperature anomalies, comprising the following steps: S1: Collect multi-source sensing data inside industrial equipment to obtain living body presence signals, ladder vibration signals, and UWB positioning personnel coordinates; S2: Determine whether the amplitude of the living body presence signal exceeds a preset threshold; S3: In response to the amplitude of the living body presence signal exceeding a preset threshold, performing refraction compensation processing on the UWB positioning person coordinates based on a temperature field distribution model to generate temperature-corrected positioning coordinates; S4: extracting features from the ladder vibration signal to obtain ladder vibration signal feature data; S5: Inputting the ladder vibration signal characteristic data into a pre-trained behavior discrimination model to output a behavior risk level; S6: Obtaining distance data between a person and a preset high-temperature area based on a spatial relationship between the temperature-corrected positioning coordinates and a preset high-temperature area of ​​the industrial equipment; S7: Combining the distance data, the behavior risk level, and the persistence indicator of the living body presence signal to make a collaborative decision and generate an alarm trigger instruction; S8: Based on the level of the alarm trigger instruction, perform a corresponding braking control operation on the industrial equipment.

[0008] In a second aspect, an embodiment of the present application provides an indoor positioning and temperature anomaly coordinated alarm system, including: The data acquisition module is used to collect multi-source sensing data from inside industrial equipment, including living body presence signals, ladder vibration signals, and UWB positioning personnel coordinates; A signal determination module, configured to determine whether the amplitude of the living body presence signal exceeds a preset threshold; a temperature compensation module, configured to, in response to the amplitude of the living body presence signal exceeding a preset threshold, perform refraction compensation processing on the coordinates of the UWB-located person based on a temperature field distribution model to generate temperature-corrected positioning coordinates; A feature extraction module is used to extract features from the ladder vibration signal to obtain feature data of the ladder vibration signal; A behavior discrimination module, configured to input the ladder vibration signal characteristic data into a pre-trained behavior discrimination model and output a behavior risk level; A distance determination module, configured to obtain distance data between a person and a preset high-temperature area based on a spatial relationship between the temperature-corrected positioning coordinates and the preset high-temperature area of ​​the industrial equipment; a collaborative decision-making module, configured to make a collaborative decision based on the distance data, the behavioral risk level, and the persistence indicator of the living body presence signal, and generate an alarm trigger instruction; A braking control module is configured to perform a corresponding braking control operation on the industrial equipment based on the level of the alarm trigger instruction.

[0009] In a third aspect, an embodiment of the present application provides an electronic device including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements a method for collaborative alarm of indoor positioning and temperature anomaly as described in the first aspect.

[0010] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for collaborative alarming of indoor positioning and temperature anomaly as in the first aspect.

[0011] Compared with the existing technology, the present invention has the following beneficial effects: the present invention provides a collaborative alarm method for indoor positioning and temperature anomalies suitable for the industrial field. The method constructs a multi-dimensional data collaborative decision-making model by integrating liveness perception, ladder vibration signal feature analysis and temperature compensation personnel positioning technology. It combines safety thresholds with multi-level braking mechanisms to effectively solve the problems of positioning drift and single sensor false triggering caused by high temperature environments, significantly improves the accuracy of personnel position detection and abnormal behavior recognition, while reducing the false alarm rate and optimizing the real-time performance of the braking response, ultimately achieving precise safety protection of industrial equipment in high-risk operation scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] A more complete understanding of the exemplary embodiments of the present invention can be obtained by referring to the following drawings. The drawings are intended to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present invention and do not constitute a limitation of the present invention. In the drawings, the same reference numerals generally represent the same components or steps.

[0013] Figure 1 A flowchart of a method for collaboratively warning indoor positioning and temperature anomalies provided by an exemplary embodiment of the present application; Figure 2 This is a flowchart of step S2 of a method for collaboratively warning indoor positioning and temperature anomalies provided by an exemplary embodiment of the present application; Figure 3 This is a flowchart of step S35 of a method for collaboratively providing an indoor positioning and temperature anomaly alarm according to an exemplary embodiment of the present application; Figure 4 This is a flowchart of step S4 of a method for collaboratively warning indoor positioning and temperature anomalies provided by an exemplary embodiment of the present application; Figure 5 A schematic diagram of an indoor positioning and temperature anomaly coordinated alarm system provided by an exemplary embodiment of the present application; Figure 6 A schematic diagram of an electronic device provided by an exemplary embodiment of the present application is shown; Figure 7 A schematic diagram of a computer-readable medium provided by an exemplary embodiment of the present application is shown. DETAILED DESCRIPTION

[0014] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0015] In the description of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limitations on the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0016] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0017] Reference Figure 1 This embodiment discloses a method for collaboratively warning indoor positioning and temperature anomaly, including the following steps: S1: Collect multi-source sensing data inside industrial equipment to obtain living body presence signals, ladder vibration signals, and UWB positioning personnel coordinates; Specifically, various sensing devices can be deployed within industrial equipment, including but not limited to liveness detection sensors, vibration sensors, and UWB (ultra-wideband) positioning devices. These sensing devices can acquire real-time biosignals (e.g., heat signatures, infrared signatures, etc.) from people to confirm the presence of living beings. Vibration sensors can be installed near or in relevant locations on ladders to monitor their vibrations, and UWB positioning technology can be used to precisely determine the coordinates of people.

[0018] S2: Determine whether the amplitude of the living body presence signal exceeds a preset threshold; Specifically, the amplitude of the living body presence signal can be used as an important reference for determining whether a person is approaching or in a dangerous area. When the signal amplitude exceeds a threshold preset by the system, the next step of operation is triggered.

[0019] The preset threshold value can be set as follows: 1) Setting based on statistical analysis of historical data. Specifically, data on the amplitude of the signal indicating the presence of living organisms inside the mixing equipment is continuously collected under different operating conditions, such as normal operation of the industrial equipment and the presence of a person. This data covers a large number of samples at different times and in different operating scenarios. Statistical analysis is performed on the collected historical data, and a frequency distribution histogram of the signal amplitude is plotted to observe its distribution patterns. For example, it was found that during normal operation, the signal amplitude is mostly concentrated in a lower range, but increases significantly when a person approaches. Based on the signal amplitude distribution, a preset threshold is selected that effectively distinguishes between normal conditions and abnormal conditions caused by the presence of a person. A value between the maximum signal amplitude in normal conditions and the minimum signal amplitude when a person approaches can be selected as the preset threshold. The threshold can be determined based on a certain probability confidence interval, such as the upper limit of a 95% confidence interval. This ensures that the probability of the signal amplitude exceeding this threshold is extremely low under normal conditions, but is highly likely to exceed this value when a person approaches.

[0020] 2) Set according to the equipment operation characteristics and personnel safety regulations. Specifically, analyze the impact of interference signals generated by various normal operations during the operation of industrial equipment itself on the living body presence signal, and determine the typical amplitude range of these interference signals in the living body presence signal detection system. Consult relevant industrial safety standards, regulations and industrial equipment operation safety specifications to understand the relevant requirements such as the safe distance for personnel activities near industrial equipment and the strength of safety signals. For example, it is stipulated that when personnel enter a certain dangerous range from the mixing equipment, the corresponding safety protection measures must be triggered. Taking into account the maximum allowable amplitude of the equipment operation interference signal and the corresponding signal strength requirements in the personnel safety regulations, set a preset threshold that is slightly higher than the normal interference signal amplitude of the equipment and meets the requirements of the safety regulations to ensure that when personnel really approach the dangerous range, they can be detected in a timely and accurate manner.

[0021] 3) Calibrate through experiments and simulate settings. Specifically, build an experimental platform in a laboratory or actual production environment that is similar to the actual operating environment of industrial equipment, including installing the same equipment such as living presence signal detection sensors. Arrange personnel to approach the mixing equipment at different speeds, postures, and distances, while recording the changes in the amplitude of the living presence signal, and obtain sample data of the signal amplitude under various possible actual human approach scenarios. Starting from an initial estimate, gradually adjust the preset threshold, conduct multiple simulation experiments at each threshold, and count indicators such as the number of correct detections of human approach and the number of false alarms. By comparing the experimental results under different thresholds, select an optimal threshold that ensures high detection accuracy while having a low false alarm rate as the final setting value.

[0022] S3: In response to the amplitude of the living body presence signal exceeding a preset threshold, refraction compensation processing is performed on the UWB positioning person coordinates based on the temperature field distribution model to generate temperature-corrected positioning coordinates; In this embodiment, taking into account the influence of temperature on signal transmission and positioning, the personnel coordinates obtained by UWB positioning are calibrated based on the established mathematical model of temperature field distribution to compensate for possible positioning offsets caused by temperature, thereby obtaining more accurate personnel positioning coordinates after temperature correction.

[0023] S4: extracting features of the ladder vibration signal to obtain ladder vibration signal feature data; Specifically, feature extraction is performed on the ladder vibration signal, and the extracted feature data may include time domain or frequency domain features such as vibration frequency, amplitude, and zero-crossing rate.

[0024] S5: Input the ladder vibration signal feature data into the pre-trained behavior discrimination model and output the behavior risk level; Specifically, the characteristic data of the ladder vibration signal is input into a pre-trained behavior discrimination model. The behavior discrimination model of this embodiment is a behavior discrimination model that has been trained in advance with a large amount of sample data. The model can identify whether the corresponding behavior is a safe normal behavior or a dangerous abnormal behavior through the input characteristic data, and generate a behavior risk level based on this, such as low risk, medium risk, high risk, etc.

[0025] Specifically, the training steps of the behavior discrimination model can be as follows: 1) Collect sample data. Specifically, vibration sensors are installed on the ladders of industrial equipment to collect vibration signals from multiple people climbing normally (both ascending and descending). Vibration signals generated by people climbing at different speeds and in different postures (e.g., holding tools, climbing barehanded), and these samples are labeled as "normal behavior." Samples are collected by simulating various abnormal climbing situations, such as those caused by prolonged pausing, sudden falls, large shaking caused by unstable climbing posture, and chaotic vibration caused by multiple people climbing simultaneously. These samples are also labeled as "abnormal behavior." The specific types of abnormal behavior (e.g., falls, crowding, etc.) can be further broken down to enable the model to more accurately identify different types of abnormal behavior. In addition to vibration signals generated by people climbing, environmental interference vibration signals unrelated to climbing can also be collected, such as slight vibrations from the industrial equipment itself, vibrations caused by nearby machinery, and vibrations caused by natural factors such as wind blowing on the ladder. These samples are used to train the model to distinguish between human behavior vibration and environmental interference vibration, thereby improving the model's robustness and accuracy.

[0026] 2) Preprocess the sample data. Specifically, clean the collected raw vibration signal to remove noise and invalid data points. For example, remove abnormal spike signals caused by sensor failure or signal transmission errors, and data in time periods with many missing values. Features extracted from the cleaned vibration signal can include time domain features (such as mean, variance, peak value, root mean square value, zero crossing rate, etc.) and frequency domain features (such as spectral energy, main frequency, frequency bandwidth, etc.). These features can effectively characterize the characteristics of the vibration signal and provide input for subsequent model training. For example, the root mean square value can reflect the energy of the vibration signal, the zero crossing rate can reflect the frequency of signal change, and the spectral energy can represent the energy distribution of the signal at different frequencies.

[0027] 3) Construct and train the model. Specifically, a variety of machine learning or deep learning models can be selected for training and comparison, such as support vector machine (SVM), random forest (RF), convolutional neural network (CNN), etc. Preferably, the present invention can select CNN because it is particularly suitable for processing vibration signal data with spatiotemporal correlation, and can automatically extract features in the signal and classify them. The extracted feature data and the corresponding labels are input into the selected model for training. During the training process, methods such as cross-validation are used to evaluate the performance of the model to avoid overfitting or underfitting. For example, the sample data is divided into a training set and a validation set. By training the model on the training set and testing its accuracy, recall rate, F1 value and other indicators on the validation set, the model's hyperparameters (such as the convolution kernel size and learning rate in CNN, etc.) are continuously adjusted to optimize the model's performance so that the model can achieve better classification results on both training data and validation data.

[0028] 4) Evaluate and optimize the model. Specifically, metrics such as accuracy, recall, F1-score, ROC curve (Receiver Operating Characteristic Curve), and AUC (Area Under ROC Curve) can be used to comprehensively evaluate model performance. Accuracy indicates the proportion of samples correctly classified by the model to the total number of samples. Recall reflects the model's ability to identify samples of a certain category (such as abnormal behavior), that is, the proportion of samples that are correctly identified among samples of that category. F1-score is the harmonic mean of accuracy and recall, comprehensively considering the balance between the two. The ROC curve is plotted with the false positive rate on the horizontal axis and the true positive rate on the vertical axis. The AUC represents the area under the ROC curve and measures the overall classification performance of the model. A larger AUC value indicates better model performance. Based on the results of model evaluation, further optimize the model. If the model performs well on the training set but poorly on the validation set, there may be an overfitting problem, which can be alleviated by adding regularization terms, reducing model complexity, increasing the amount of training data, etc. If the model performs poorly on both the training set and the validation set, there may be an underfitting problem, and you can try increasing the complexity of the model, adjusting the model structure, increasing the number of features, etc.

[0029] S6: Obtaining distance data between the personnel and the preset high-temperature area based on the spatial relationship between the temperature-corrected positioning coordinates and the preset high-temperature area of ​​the industrial equipment; Specifically, the temperature-corrected personnel positioning coordinates are spatially compared with the high-temperature area pre-set in the system to determine whether the personnel is in the high-temperature area or close to the high-temperature area.

[0030] S7: Combine distance data, behavioral risk level, and persistence indicators of live presence signals to make collaborative decisions and generate alarm trigger instructions; Specifically, in combination with a series of indicators such as the above-mentioned behavioral risk level and whether the live presence signal is continuously and stably present, a decision is made by comprehensively considering multiple factors. When all conditions meet the preset alarm trigger conditions, the corresponding alarm trigger instruction is generated.

[0031] Specifically, the persistence metric characterizes whether the presence signal is stable, rather than intermittent or sporadic. A stable signal indicates the continuous presence of a living being within industrial equipment, helping to accurately determine whether a person is inside. This metric also measures the temporal continuity of the signal, specifically whether the signal experiences frequent interruptions or fluctuations. A continuous signal indicates that the presence of a living being is uninterrupted or unchanging, helping to distinguish true human presence signals from transient signals caused by environmental noise or other interference.

[0032] Specifically, continuity indicators can include signal presence time percentage, signal strength fluctuation range, and signal continuity parameters. The signal presence time percentage refers to the ratio of the time the liveness signal is detected to the total monitoring time within a specific monitoring period. The signal strength fluctuation range refers to the difference between the maximum and minimum liveness signal strength (e.g., amplitude) within a specific period. Signal continuity parameters may include, but are not limited to, continuous presence time percentage and interruption time percentage. The continuous presence time percentage indicates the percentage of time within a specific time interval during which the liveness signal is continuously present. For example, if the signal is continuously present for 50 seconds within a 1-minute interval, the continuous presence time percentage is 50 / 60 ≈ 0.83, or 83%. The interruption time percentage indicates the percentage of time within a specific time interval during which the liveness signal is interrupted. This is the ratio of the remaining time after subtracting the signal presence time from the total monitoring time to the total monitoring time. This parameter complements the continuous presence time percentage and measures the degree of signal interruption.

[0033] S8: Based on the level of the alarm trigger instruction, perform corresponding braking control operations on the industrial equipment.

[0034] Specifically, depending on the level of the alarm trigger (e.g., level 1, level 2, etc.), corresponding braking control measures are taken for industrial equipment. For example, for industrial equipment such as mixing equipment, a low-level alarm could initiate measures such as reducing the mixing speed, while a high-level alarm could directly command the equipment to stop operating immediately, ensuring effective braking in abnormal situations to protect the safety of personnel and equipment.

[0035] The above-mentioned indoor positioning and intelligent alarm method suitable for the industrial field, by integrating live perception, ladder vibration signal feature analysis and temperature compensation personnel positioning technology, constructs a multi-dimensional data collaborative decision-making model, and combines safety thresholds with multi-level braking mechanisms. It effectively solves the problems of positioning drift and single sensor false triggering caused by high temperature environment, significantly improves the accuracy of personnel position detection and abnormal behavior recognition, while reducing the false alarm rate and optimizing the real-time performance of braking response, and ultimately realizes the precise safety protection of industrial equipment in high-risk operation scenarios.

[0036] Reference Figure 2 In some optional embodiments, refraction compensation processing is performed on the UWB positioning personnel coordinates based on the temperature field distribution model to generate temperature-corrected positioning coordinates, including the following steps: S31: Collect internal temperature data of industrial equipment through a distributed thermocouple array and build a temperature field distribution model based on the internal temperature data; Among them, the temperature field distribution model is a three-dimensional temperature gradient field, and the three-dimensional temperature gradient field is: (1); Specifically, a distributed thermocouple array can be reasonably arranged inside the industrial equipment. The thermocouple can sense the temperature conditions at different locations inside the equipment in real time and transmit the collected temperature data back. Then, the temperature data is used to construct a three-dimensional temperature gradient field to obtain formula (1). Is a reference temperature value, which serves as a basic reference temperature for the entire temperature field; Indicates the The intensity coefficient of each heat source. Different heat sources have different heating intensities, which are quantified by this coefficient. is the heat diffusion radius, which reflects the range of heat diffusion from the heat source; Then accurately give the The coordinate position of each heat source in three-dimensional space can clearly identify the specific location of the heat source inside the equipment; Represents the total number of heat sources, that is, how many heat source points are inside the device.

[0037] S32: Construct the electromagnetic wave refractive index field based on the temperature field distribution model; Among them, the refractive index field of electromagnetic waves is: (2); Specifically, based on the constructed three-dimensional temperature gradient field, the refractive index field of the electromagnetic wave is further calculated, and its expression is formula (2). In this step, Is the temperature refractive index coefficient, a constant related to the material properties and temperature changes inside industrial equipment, used to characterize the degree of influence of temperature changes on the refractive index of electromagnetic waves. Through this formula, we can get the refractive index of different locations inside the equipment. The refractive index of electromagnetic waves at , thereby clarifying the changes in refractive characteristics caused by temperature differences when electromagnetic waves propagate inside the device.

[0038] S33: Obtaining a path length deviation based on the electromagnetic wave refractive index field and the path length deviation model; wherein the path length deviation is caused by a refractive index change of the UWB signal used to locate the coordinates of the person on its propagation path; Among them, the path length deviation model is: (3); Specifically, based on the determined electromagnetic wave refractive index field, the path length deviation of the UWB signal caused by the refractive index change on the propagation path is calculated using formula (3): .in, is the UWB signal incident angle, which represents the angle between the UWB signal and the normal when it enters different media (in this scenario, regions with different electromagnetic wave refractive indices caused by different temperature regions); The standard refractive index of air is used as a known reference value to compare with the actual refractive index inside industrial equipment; It is the propagation path of the UWB signal, which clearly defines the specific route the signal takes from the transmitter to the receiver inside the device; Represents a small length element on the propagation path. By integrating the refractive index difference effect corresponding to each small length element on the entire path, the total path length deviation is finally obtained. ,This deviation reflects the impact of refraction changes caused by temperature on the propagation distance of UWB signals.

[0039] S34: Compensating the TOA measurement value of the UWB signal according to the path length deviation to obtain a compensated TOA value of the UWB signal; The compensated TOA value is expressed by the following formula: (4); Specifically, according to the path length deviation calculated above , the TOA (Time of Arrival) measurement value of the UWB signal is compensated by formula (4). Among them, is the original UWB signal TOA measurement value; is the speed of light in a vacuum, which serves as a benchmark for the speed of electromagnetic wave propagation; The temperature drift compensation term takes into account the drift caused by temperature changes on the UWB signal characteristics and other potential influencing factors during the measurement process. After processing with this compensation formula, the compensated TOA value is obtained. The compensated value can more accurately reflect the actual propagation time of the UWB signal inside the industrial equipment, thus providing a basis for subsequent more accurate positioning.

[0040] S35: Acquire the personnel feature points of the industrial equipment and input them into the visual synchronous positioning and mapping algorithm for processing to obtain the three-dimensional personnel point cloud coordinates; Specifically, binocular cameras can be installed inside industrial equipment. Using their imaging principles, they can capture the characteristic points of people inside the equipment. These points can include key body parts and contour features. Then, using a simultaneous localization and mapping (SLAM) algorithm, a point cloud coordinate system is constructed in three-dimensional space based on these acquired characteristic points. These point cloud coordinates accurately represent the spatial distribution of people within the industrial equipment, providing a visual positioning reference for subsequent fusion with UWB positioning data.

[0041] S36: The temperature-corrected positioning coordinates are obtained by fusing the compensated TOA value and the three-dimensional personnel point cloud coordinates through an extended Kalman filter.

[0042] Specifically, the previously compensated UWB signal TOA values ​​and the 3D person point cloud coordinates generated by the binocular camera are input into an Extended Kalman Filter (EKF) for data fusion. The EKF effectively handles state estimation problems in nonlinear systems. In this step, it comprehensively considers the characteristics of UWB and visual positioning data, as well as the error characteristics between them. By fusing the two data, it ultimately generates temperature-corrected positioning coordinates. These temperature-corrected positioning coordinates combine the advantages of both positioning methods while accounting for the impact of temperature on UWB positioning. This allows for more accurate determination of the person's actual position within the industrial equipment, providing a reliable basis for subsequent collaborative decision-making and alarm triggering.

[0043] Reference Figure 3 In an optional embodiment, step S35 includes the following steps: S351: Use binocular cameras to capture images of the internal environment of industrial equipment and obtain raw visual data streams; Specifically, binocular cameras can be installed inside industrial equipment, allowing the two cameras to simultaneously capture images of the environment inside the equipment from slightly different angles. This method can capture a raw visual data stream containing internal scene information, including image information of various objects and people inside the equipment.

[0044] S352: Perform ORB feature extraction on each frame of the original visual data stream to obtain person feature points; Specifically, the ORB (Oriented FAST and Rotated BRIEF) feature extraction algorithm is used to process each frame of the raw visual data stream captured by the binocular camera. The ORB algorithm first uses the FAST (Features from Accelerated Segment Test) algorithm to quickly detect feature points such as corners in the image. It then calculates the orientation information of these feature points and uses the BRIEF (Binary Robust Independent Elementary Features) algorithm to generate descriptors for these feature points. This method extracts representative feature points of individuals from each image frame, which uniquely identify key parts or characteristic regions of individuals in the image.

[0045] S353: Performing three-dimensional coordinate conversion on the person's feature points based on parallax calculation to obtain sparse point cloud data; Specifically, the method utilizes the parallax principle of binocular cameras. Due to the different viewing angles of the two cameras, the position of the same insider's feature point in the two images will differ (parallax). By calculating this parallax and combining it with the internal and external parameters of the binocular cameras (such as the focal length and baseline distance), the insider's feature points in the two-dimensional image can be converted into coordinate points in three-dimensional space. The resulting series of three-dimensional coordinate points constitutes sparse point cloud data. Although the point cloud data is relatively sparsely distributed in three-dimensional space, it can initially provide information on the approximate spatial position and posture of the insider within the industrial equipment.

[0046] S354: Perform IPC registration on the sparse point cloud data and the global map to obtain the visual SLAM pose transformation matrix; wherein the global map is a dense environment model generated by fusing all historical valid frame point cloud data through voxel filtering; Specifically, the sparse point cloud data obtained above is aligned with a pre-built global map using IPC (Iterative Point Cloud Registration). The global map is a dense environment model generated by fusing the point cloud data from all historical valid frames through voxel filtering (which removes noise and reduces data redundancy). It contains relatively complete 3D environmental information within the industrial equipment. The IPC registration algorithm calculates the optimal matching transformation between the current sparse point cloud data and the global map, thereby solving the visual SLAM (Simultaneous Localization and Mapping) pose transformation matrix. This pose transformation matrix represents the current camera position and pose in the global map coordinate system.

[0047] Specifically, the process of constructing the global map can be as follows: The binocular camera continuously captures the interior of industrial equipment at a certain frame rate (e.g., 30 frames per second), generating a series of images. Each frame contains visual information about the interior of the equipment at that moment.

[0048] For each image frame, feature extraction algorithms such as ORB are used to successfully extract a sufficient number of well-distributed feature points. Only frames that successfully extract a sufficient number of feature points are considered valid. For example, if an image frame is too dark, too bright, or blurred, and the extracted feature points are too few or too concentrated, and cannot accurately describe the internal scene of the device, the frame may be considered invalid.

[0049] In a binocular camera, feature points in the left and right images must be matched. Only frames with a matching success rate exceeding a certain threshold (e.g., 80%) are considered valid. Frames with a low matching success rate may be caused by factors such as occlusion or reflections from internal objects, making it impossible to accurately construct a 3D point cloud.

[0050] To further ensure the accuracy of the point cloud data, the disparity after feature point matching can be checked for consistency. For example, the disparity between the left and right views should satisfy specific geometric constraints. If the disparity data of a frame does not meet these constraints, the frame will be excluded and not included in the valid frames.

[0051] For valid frames, the parallax principle of binocular vision is used to convert the successfully matched feature points into coordinate points in three-dimensional space, generating sparse point cloud data. These point cloud data points contain the position information of objects inside industrial equipment in three-dimensional space.

[0052] Voxel filtering is performed on the sparse point cloud data of all historical valid frames. Voxel filtering divides the three-dimensional space into small cubes (voxels), and counts and filters the point cloud data within each voxel, removing noise points and retaining representative points, thereby obtaining cleaner and more regular point cloud data.

[0053] The point cloud data after voxel filtering is integrated with valid frame point cloud data at different times and perspectives through fusion algorithms (such as fusion based on spatial consistency, fusion based on time series, etc.) to generate a dense internal environment model of industrial equipment.

[0054] S355: Convert the visual SLAM pose transformation matrix to the coordinate system of the industrial equipment to obtain the three-dimensional personnel point cloud coordinates.

[0055] Specifically, since the visual SLAM pose transformation matrix is ​​obtained in the camera coordinate system, while what is actually needed is the three-dimensional coordinates of the personnel in the industrial equipment coordinate system, a preset transformation matrix is ​​required to perform the coordinate system conversion. The preset transformation matrix is ​​pre-calculated based on information such as the structure of the industrial equipment and the installation location of the camera. It can accurately convert the data in the camera coordinate system to the industrial equipment coordinate system. After this conversion, the three-dimensional point cloud coordinates of the internal personnel in the industrial equipment coordinate system can be obtained, which can accurately represent the spatial position of the personnel inside the industrial equipment.

[0056] Reference Figure 4 In an optional embodiment, step S4 includes the following steps: S41: Perform wavelet packet decomposition on the ladder vibration signal based on Daubechies 4 wavelet basis to obtain wavelet coefficients; Specifically, the Daubechies 4 wavelet basis is used to perform wavelet packet decomposition on the ladder vibration signal. In the prior art, the Daubechies 4 wavelet basis is a wavelet function used as a basis function in wavelet transforms. Wavelet transforms are a signal processing method that utilizes wavelet bases to decompose and reconstruct signals. Wavelet packet decomposition is an analytical method that can gradually decompose a signal into different frequency bands. This decomposition yields wavelet coefficients for each frequency band. The wavelet coefficients reflect the energy distribution of the signal in each frequency band.

[0057] S42: Based on the wavelet coefficient and energy proportion model, obtaining the energy proportion of the ladder vibration signal in a preset frequency band; wherein the energy proportion is used to capture the impact signal generated by abnormal behavior when the person climbs; The energy proportion model is expressed by the following formula: (5); Formula (5) is used to calculate the energy proportion of the ladder vibration signal in the preset frequency band (0 to P Hz). The preset frequency band is set based on the frequency range of normal vibration signals when people climb in actual application scenarios and the frequency range of impact signals caused by possible abnormal behavior. is the wavelet coefficient, is the wavelet scale factor, which is inversely proportional to the frequency of the signal. The larger the scale, the lower the corresponding frequency. is a translation factor used to shift the wavelet function on the time axis to analyze the signal at different locations. By calculating the energy fraction, we can effectively capture the impact signal generated by abnormal climbing behavior. This is because abnormal behavior often causes a sudden change or increase in energy within a specific frequency band, which in turn changes the energy fraction of that frequency band.

[0058] S43: Obtaining a sample entropy feature of the ladder vibration signal based on a sample entropy calculation model; The sample entropy calculation model is expressed by the following formula: (6); Specifically, the sample entropy algorithm is used to calculate the sample entropy characteristics of the ladder vibration signal. is the embedding dimension, which represents the length of the continuous time series segment extracted from the ladder vibration signal. This length determines the size of the time window considered when analyzing the signal; Represents the similarity tolerance threshold, which is used to measure whether two signal segments are similar, and is usually a certain percentage of the signal standard deviation; is the number of data points of the ladder vibration signal, that is, the total length of the signal. is the matching probability, which means that the length of the ladder vibration signal is The similarity between consecutive signal segments is less than The probability of climbing a certain height can be calculated, and calculating this probability can reflect the complexity and regularity of the signal. The sample entropy feature can quantify the randomness of human behavior patterns, because different behavior patterns (normal climbing or abnormal climbing) will lead to different complexity and regularity of the vibration signal, thus causing the sample entropy value to change.

[0059] S44: Integrate the energy proportion and the sample entropy characteristics to obtain ladder vibration signal characteristic data.

[0060] Specifically, the previously calculated energy percentage and sample entropy features are integrated to form complete feature data. The energy percentage reflects the energy variation of the signal within a specific frequency band and is primarily used to detect abnormal impact signals. The sample entropy feature, on the other hand, quantifies the complexity and randomness of the signal, reflecting differences in human behavior patterns. By combining these two features, a more comprehensive description of the characteristics of ladder vibration signals can be achieved, providing richer and more effective input information for the subsequent behavioral discrimination model, enabling more accurate identification of the risk level of climbing behavior.

[0061] The feature integration operation in this step can be feature concatenation, which creates a feature vector containing two elements: the first element is the energy contribution, and the second element is the sample entropy feature. For example, if the energy contribution is 0.6 and the sample entropy feature is 0.8, the integrated feature vector is F = [0.6, 0.8].

[0062] In an optional embodiment, The calculation steps include: (1) Divide the ladder vibration signal into indivual dimensional vector; The dimensional vector is represented as ,in, is the ladder vibration signal, .

[0063] Specifically, first determine the total number of data points of the ladder vibration signal as N. Then, according to the set embedding dimension m, the entire ladder vibration signal is divided into N-m+1 m-dimensional vectors. Each m-dimensional vector is represented as ,in is the amplitude of the ladder vibration signal at time t, where t ranges from 1 to N-m+1. This segmentation method aims to convert the one-dimensional vibration signal into a multidimensional vector sequence, enabling better analysis of the signal's characteristics and patterns. By considering combinations of m consecutive data points, the dynamic temporal characteristics of the signal can be captured.

[0064] (2) Targeting indivual dimensional vectors, and the Chebyshev distance between the vectors is calculated according to the following formula: (7); Specifically, for the N-m+1 m-dimensional vectors obtained by the previous segmentation, calculate the Chebyshev distance between them. In the calculation formula (7) of Chebyshev distance , Represents a vector With vector The Chebyshev distance between two vectors is used. This distance metric focuses on the maximum difference between two vectors in each dimension and effectively reflects the overall degree of difference between the two vectors. By calculating the Chebyshev distance between all vectors, a distance matrix is ​​generated, providing distance information for subsequent operations such as statistical matching logarithms, thereby further analyzing signal similarities and regularities.

[0065] (3) According to the following formula, statistics satisfy The number of matching pairs : (8); Specifically, according to formula (8), the statistics satisfying the Chebyshev distance are Less than the similarity tolerance threshold The number of matching pairs .in, is the characteristic function, when When the conditions in the brackets are met, The value is 1, otherwise the value is 0. That is, for each vector , the statistical distance is less than Other vectors of The sum of the quantities is The purpose of this statistical process is to find the number of other vectors that are similar to each vector within the similarity tolerance range, thereby reflecting the frequency of occurrence of similar patterns in the signal, providing a basis for subsequent calculation of matching probability, and thus better quantifying the complexity and randomness characteristics of the signal.

[0066] (4) According to the formula , calculate the matching probability .

[0067] Specifically, finally, according to the formula Calculating matching probability This formula represents the matching number of all vectors Sum and then divide by , get the matching probability. Matching probability Reflecting the similarity tolerance In the range, the length of the ladder vibration signal is The probability of similarity between consecutive signal segments can quantify the regularity and repetitiveness of the signal. By calculating the matching probability, we can further analyze the random differences in human behavior patterns, providing important characteristic parameters for subsequent behavior discrimination models, thereby more accurately identifying the risk level of human behavior during climbing.

[0068] In an optional embodiment, step S6 includes: S61: Based on the temperature-corrected positioning coordinates and the temperature field distribution of the preset high-temperature area, obtaining distance data between the person and the preset high-temperature area; the distance data is Mahalanobis distance.

[0069] Specifically, the temperature-corrected coordinates of the personnel's location are first obtained. These coordinates account for the impact of temperature on positioning accuracy and more accurately reflect the personnel's actual position within the industrial equipment. Furthermore, the temperature field distribution of the preset high-temperature area is determined. This preset high-temperature area is pre-determined based on the process characteristics and safety requirements of the industrial equipment. The temperature field distribution within this area can be uniform or have a gradient. The Mahalanobis distance formula is then used to calculate the Mahalanobis distance between the personnel's location and the preset high-temperature area. The Mahalanobis distance considers the covariance structure of the data and can effectively measure the distance between two points in multidimensional space, while also accounting for the correlations between dimensions and the scale differences between different dimensions. In this step, by calculating the Mahalanobis distance between the personnel's location and the preset high-temperature area, the relative positional relationship between the personnel and the high-temperature area can be quantitatively assessed, providing an important indicator for subsequent alarm decisions. If the Mahalanobis distance between the personnel's location and the high-temperature area is small, it indicates that the person is close to the high-temperature area, potentially posing a safety risk.

[0070] In an optional embodiment, step S7 includes: S71: Mapping the behavior risk level to a risk coefficient according to a preset mapping relationship; Specifically, based on the behavioral risk level obtained previously through the behavioral discrimination model, combined with the preset mapping relationship, the behavioral risk level is converted into a risk coefficient. The preset mapping relationship can be determined based on actual experience, safety standards, and an assessment of the possible consequences of different behavioral risk levels. For example, behavioral risk levels can be divided into low risk, medium risk, high risk, etc. The corresponding mapping relationship may be that low risk is mapped to a smaller risk coefficient, such as 0.2; medium risk is mapped to a medium risk coefficient, such as 0.5; high risk is mapped to a larger risk coefficient, such as 0.8. By quantifying the behavioral risk level as a risk coefficient, it can be more convenient to comprehensively consider and calculate it with other indicators in the subsequent alarm trigger instruction generation process, making the alarm decision more scientific, reasonable and quantitative.

[0071] S72: Generate an interference suppression flag based on the persistence indicator of the living body presence signal; Specifically, the persistence indicator of the live presence signal is analyzed, which reflects the temporal stability of the live presence signal. If the live presence signal is stable and continuous, it indicates that there is indeed human activity within the industrial equipment. If the live presence signal is intermittent or exhibits abnormal fluctuations, it may be due to external interference or other abnormal conditions. Based on the persistence indicator of the live presence signal, an interference suppression flag is generated. The interference suppression flag can be used to determine the reliability of the current live presence signal. If the signal persistence is good, the interference suppression flag can be 0, indicating that the signal is normal and not subject to interference. If the signal persistence is poor, possibly due to interference, the interference suppression flag can be 1, indicating that interference suppression processing is required or that the influence of interference factors should be considered in the alarm decision-making, thereby avoiding false alarms caused by misjudgments and improving the accuracy and reliability of the alarm system.

[0072] S73: Generate an alarm trigger instruction based on the Mahalanobis distance, risk coefficient, and interference suppression flag.

[0073] Specifically, the previously calculated Mahalanobis distance, risk coefficient, and generated interference suppression flag are comprehensively considered, and an alarm trigger instruction is generated based on preset alarm triggering rules and algorithms. The alarm triggering rules can be formulated based on the safe operation requirements of industrial equipment, personnel safety standards, and a comprehensive assessment of various risk factors.

[0074] Specifically, the alarm triggering steps may be as follows: 1) Normalize the parameters. Specifically, since the dimensions and numerical ranges of the Mahalanobis distance, risk coefficient, and interference suppression flag are different, they are first normalized and their values ​​are mapped to the [0,1] interval.D The smaller the value, the closer the person is to the high temperature area. . Risk Factor R Already in the range [0,1]. Interference suppression flag I Perform logical inversion and normalization .

[0075] 2) Perform a weighted summation of the parameters. Specifically, different weights are assigned to the normalized parameters based on the importance and urgency of the security risk. Assume that the Mahalanobis distance weight is 0.5, the risk coefficient is 0.3, and the interference suppression flag is 0.2. The weighted summation formula is: .

[0076] 3) Trigger alarms based on thresholds. Specifically, set alarm trigger thresholds based on historical data and security standards. For example, set the threshold to 0.7. If the value is less than 0, the alarm will be triggered; otherwise, it will not be triggered.

[0077] In an optional embodiment, the indoor positioning and temperature anomaly coordinated alarm method further includes the following steps: When a preset brake control release condition is met, the brake control operation on the industrial equipment is released.

[0078] Among them, the braking control release conditions are: the rate of change of the three-dimensional temperature gradient field is less than the first threshold, the Mahalanobis distance is greater than the second threshold, and the vibration energy spectrum density of the industrial equipment is less than the third threshold.

[0079] Specifically, the rate of change of the three-dimensional temperature gradient field reflects how quickly the temperature distribution inside the industrial equipment changes over time. When the rate of change is less than the first threshold, it means that the temperature situation inside the industrial equipment tends to be stable and no longer changes drastically. For example, during the normal operation of industrial equipment such as stirring equipment, the internal temperature may fluctuate due to factors such as stirring action and material reaction, causing changes in the three-dimensional temperature gradient field. When the production process enters a stable stage, or after the industrial equipment stops running for a period of time, the temperature gradually stabilizes, and the rate of change of the three-dimensional temperature gradient field will decrease. The first threshold is set to determine a stable range of temperature changes. When the temperature changes within this range, it can be considered that the temperature factor has little impact on the safe operation of the industrial equipment, and meets a condition for releasing the braking control.

[0080] The Mahalanobis distance has been mentioned in the previous content and is used to measure the relative position relationship between the personnel position and the preset high-temperature area. When the Mahalanobis distance is greater than the second threshold, it means that the personnel has moved away from the preset high-temperature area and is in a relatively safe position. For example, inside industrial equipment, the preset high-temperature area may be close to the heating device or the high-temperature material reaction area. When personnel are moving normally inside the equipment, they may be close to the high-temperature area, and the Mahalanobis distance is small at this time. When the personnel completes the relevant operations and leaves the high-temperature area, the Mahalanobis distance will increase. The second threshold is set to define a safe distance range. When the distance between the personnel and the high-temperature area exceeds this range, the safety risk caused by the personnel position is reduced, and another condition for releasing the braking control is met.

[0081] The vibration energy spectrum density reflects the distribution of the vibration energy of industrial equipment at different frequencies. When the vibration energy spectrum density is less than the third threshold, it indicates that the vibration of the industrial equipment is within the normal range, without abnormal vibration energy concentration or excessive vibration amplitude. For example, when industrial equipment (such as mixing equipment) is operating normally, the rotation of the mixing blades and the stirring of the materials will generate a certain amount of vibration, and the vibration energy spectrum density is within a reasonable range. However, if the equipment fails, such as an unbalanced mixing shaft or bearing wear, the vibration energy spectrum density may increase, resulting in abnormal vibration. The third threshold is set to determine a normal vibration energy range. When the vibration energy spectrum density is within this range, the mechanical operation of the equipment is in good condition and will not cause safety hazards due to abnormal vibration. The third condition for releasing the brake control is met.

[0082] When all three conditions are met simultaneously—the rate of change of the three-dimensional temperature gradient field is less than the first threshold, the Mahalanobis distance is greater than the second threshold, and the vibration energy spectral density of the industrial equipment is less than the third threshold—the various safety risks facing the industrial equipment can be considered reduced to a safe range. At this point, the previously executed braking control operation can be released, allowing the equipment to resume normal operation. This braking control release condition ensures the safety of both personnel and equipment, allowing for the proper resumption of equipment operation and improved production efficiency.

[0083] The above-mentioned method for collaborative alarm of indoor positioning and temperature anomalies of industrial equipment collects the presence signals of living things, ladder vibration signals and UWB-located personnel coordinates inside the industrial equipment in real time, and combines the temperature field distribution model to dynamically compensate for the electromagnetic wave refraction effect to correct the UWB-located personnel coordinates. It uses wavelet packet decomposition and sample entropy algorithm to extract the low-frequency characteristics and complexity parameters of human behavior in the ladder vibration signal, and constructs a dynamic collaborative decision-making model that integrates spatial distance, behavioral risk level and living body continuity. Finally, through a multi-level braking verification mechanism and safety release conditions, it effectively solves the problems of positioning drift and multi-sensor false triggering in high temperature environments, significantly improves positioning accuracy, abnormal behavior recognition accuracy and overall system reliability, and realizes accurate monitoring and rapid active protection of the safety status of personnel in industrial confined spaces.

[0084] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0085] Based on the same inventive concept, the present application also provides a system for implementing the aforementioned indoor positioning and temperature anomaly coordinated alarm method. The solution provided by this system is similar to the solution described in the aforementioned method. Therefore, the specific limitations of the embodiments of the indoor positioning and temperature anomaly coordinated alarm system provided below can be found in the above-mentioned limitations of the indoor positioning and temperature anomaly coordinated alarm method, and will not be repeated here.

[0086] In an exemplary embodiment, Figure 5 As shown, an indoor positioning and temperature anomaly coordinated alarm system 20 is provided, comprising: The data acquisition module 201 is used to collect multi-source sensing data inside the industrial equipment, obtain living body presence signals, ladder vibration signals, and UWB positioning personnel coordinates; The signal determination module 202 is configured to determine whether the amplitude of the living body presence signal exceeds a preset threshold; a temperature compensation module 203 for performing refraction compensation processing on the UWB-located person coordinates based on a temperature field distribution model to generate temperature-corrected positioning coordinates in response to the amplitude of the living body presence signal exceeding a preset threshold; A feature extraction module 204 is used to extract features from the ladder vibration signal to obtain ladder vibration signal feature data; A behavior discrimination module 205 is configured to input the ladder vibration signal characteristic data into a pre-trained behavior discrimination model and output a behavior risk level; A distance determination module 206 is configured to obtain distance data between a person and a preset high-temperature area based on a spatial relationship between the temperature-corrected positioning coordinates and the preset high-temperature area of ​​the industrial equipment; A collaborative decision module 207 is configured to make a collaborative decision based on the distance data, the behavior risk level, and the persistence indicator of the living body presence signal, and generate an alarm trigger instruction; The braking control module 208 is configured to perform a corresponding braking control operation on the industrial equipment based on the level of the alarm trigger instruction.

[0087] Please refer to Figure 6 , which shows a schematic diagram of an electronic device provided by some embodiments of the present application. Figure 6 As shown, the electronic device 30 includes: a processor 300, a memory 301, a bus 302 and a communication interface 303, and the processor 300, the communication interface 303 and the memory 301 are connected via the bus 302; the memory 301 stores a computer program that can be run on the processor 300, and the processor 300 executes the aforementioned method of this application when running the computer program.

[0088] Memory 301 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. Communication between the system network element and at least one other network element is achieved through at least one communication interface 303 (which may be wired or wireless), and may utilize the Internet, a wide area network, a local area network, a metropolitan area network, or the like.

[0089] The bus 302 may be an ISA bus, a PCI bus, or an EISA bus. The bus may be divided into an address bus, a data bus, a control bus, etc. The memory 301 is used to store programs. The processor 300 executes the programs upon receiving execution instructions. The methods disclosed in any of the aforementioned embodiments of the present application may be applied to or implemented by the processor 300.

[0090] The processor 300 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by hardware integrated logic circuits in the processor 300 or by software instructions. The above processor 300 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in conjunction with the embodiments of this application can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium mature in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory 301 , and the processor 300 reads the information in the memory 301 and completes the steps of the above method in combination with its hardware.

[0091] The electronic device provided in the embodiments of the present application and the method provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, operated or implemented by them.

[0092] The present application also provides a computer-readable medium corresponding to the method provided in the above embodiment. Figure 4 The computer-readable storage medium shown is a CD 40 on which a computer program (ie, a program product) is stored. When the computer program is run by a processor, the aforementioned method is executed.

[0093] It should be noted that examples of the computer-readable storage medium may also include, but are 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 or magnetic storage media, which are not listed here one by one.

[0094] It should be noted that the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, program segment or part of code, and the module, program segment or part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0095] Those skilled in the art will 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 aforementioned method embodiments and will not be repeated here.

[0096] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. 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 mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interface, the indirect coupling or communication connection of the device or unit can be electrical, mechanical or other forms.

[0097] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0098] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0099] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and they should all be included in the scope of the claims and description of the present application.

Claims

1. A method for collaborative alarming of indoor positioning and temperature anomaly, characterized in that: The steps include: S1: Collect multi-source sensing data inside industrial equipment to obtain living body presence signals, ladder vibration signals, and UWB positioning personnel coordinates; S2: Determine whether the amplitude of the living body presence signal exceeds a preset threshold; S3: In response to the amplitude of the living body presence signal exceeding a preset threshold, performing refraction compensation processing on the UWB positioning person coordinates based on a temperature field distribution model to generate temperature-corrected positioning coordinates; S4: extracting features from the ladder vibration signal to obtain ladder vibration signal feature data; S5: Inputting the ladder vibration signal characteristic data into a pre-trained behavior discrimination model to output a behavior risk level; S6: Obtaining distance data between a person and a preset high-temperature area based on a spatial relationship between the temperature-corrected positioning coordinates and a preset high-temperature area of ​​the industrial equipment; S7: Combining the distance data, the behavior risk level, and the persistence indicator of the living body presence signal to make a collaborative decision and generate an alarm trigger instruction; S8: Based on the level of the alarm trigger instruction, perform a corresponding braking control operation on the industrial equipment.

2. The method according to claim 1, characterized in that The performing refraction compensation processing on the UWB positioning personnel coordinates based on the temperature field distribution model to generate temperature-corrected positioning coordinates includes: S31: collecting internal temperature data of the industrial equipment through a distributed thermocouple array, and constructing a temperature field distribution model based on the internal temperature data, wherein the temperature field distribution model is a three-dimensional temperature gradient field; S32: Constructing an electromagnetic wave refractive index field according to the temperature field distribution model; S33: Obtaining a path length deviation based on the electromagnetic wave refractive index field and the path length deviation model; wherein the path length deviation is caused by a refractive index change of the UWB signal used to locate the coordinates of the person by UWB on its propagation path; S34: Compensating the TOA measurement value of the UWB signal according to the path length deviation to obtain a compensated TOA value of the UWB signal; S35: Acquire the personnel feature points of the industrial equipment and input them into the visual synchronous positioning and mapping algorithm for processing to obtain the three-dimensional personnel point cloud coordinates; S36: Fusing the compensated TOA value and the three-dimensional personnel point cloud coordinates through an extended Kalman filter to obtain the temperature-corrected positioning coordinates.

3. The method according to claim 2, characterized in that Step S35 includes: S351: Capturing an internal environment image of the industrial equipment through a binocular camera to obtain an original visual data stream; S352: Perform ORB feature extraction on each frame of the original visual data stream to obtain person feature points; S353: performing three-dimensional coordinate conversion on the person's feature points based on parallax calculation to obtain sparse point cloud data; S354: Performing IPC registration on the sparse point cloud data and the global map to obtain a visual SLAM pose transformation matrix; wherein the global map is a dense environment model generated by fusing all historical valid frame point cloud data through voxel filtering; S355: Convert the visual SLAM pose transformation matrix to the coordinate system of the industrial equipment to obtain the three-dimensional personnel point cloud coordinates.

4. The method according to any one of claims 1 to 3, characterized in that Step S4 includes: S41: performing wavelet packet decomposition on the ladder vibration signal based on Daubechies 4 wavelet basis to obtain wavelet coefficients; S42: Based on the wavelet coefficients and the energy proportion model, obtaining the energy proportion of the ladder vibration signal in a preset frequency band; wherein the energy proportion is used to capture the impact signal generated by abnormal behavior when a person climbs; S43: Obtaining a sample entropy feature of the ladder vibration signal based on a sample entropy calculation model; S44: Integrate the energy proportion and the sample entropy feature to obtain the ladder vibration signal characteristic data.

5. The method according to claim 2, characterized in that Step S6 includes: S61: Based on the temperature-corrected positioning coordinates and the temperature field distribution of the preset high-temperature area, obtaining distance data between the person and the preset high-temperature area; the distance data is Mahalanobis distance.

6. The method according to claim 5, characterized in that Step S7 includes: S71: Mapping the behavior risk level to a risk coefficient according to a preset mapping relationship; S72: Generate an interference suppression flag based on the persistence indicator of the living body presence signal; S73: Generate the alarm trigger instruction by combining the Mahalanobis distance, the risk coefficient and the interference suppression flag.

7. The method according to claim 6, characterized in that The method further comprises: When a preset brake control release condition is met, releasing the brake control operation on the industrial equipment; The braking control release condition is as follows: the rate of change of the three-dimensional temperature gradient field is less than a first threshold, the Mahalanobis distance is greater than a second threshold, and the vibration energy spectrum density of the industrial equipment is less than a third threshold.

8. An indoor positioning and temperature anomaly coordinated alarm system, characterized in that: include: The data acquisition module is used to collect multi-source sensing data from inside industrial equipment, including living body presence signals, ladder vibration signals, and UWB positioning personnel coordinates; A signal determination module, configured to determine whether the amplitude of the living body presence signal exceeds a preset threshold; a temperature compensation module, configured to, in response to the amplitude of the living body presence signal exceeding a preset threshold, perform refraction compensation processing on the coordinates of the UWB-located person based on a temperature field distribution model to generate temperature-corrected positioning coordinates; A feature extraction module is used to extract features from the ladder vibration signal to obtain feature data of the ladder vibration signal; A behavior discrimination module, configured to input the ladder vibration signal characteristic data into a pre-trained behavior discrimination model and output a behavior risk level; A distance determination module, configured to obtain distance data between a person and a preset high-temperature area based on a spatial relationship between the temperature-corrected positioning coordinates and the preset high-temperature area of ​​the industrial equipment; a collaborative decision-making module, configured to make a collaborative decision based on the distance data, the behavioral risk level, and the persistence indicator of the living body presence signal, and generate an alarm trigger instruction; A braking control module is configured to perform a corresponding braking control operation on the industrial equipment based on the level of the alarm trigger instruction.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images