A Design Method of Electronic Fence for Electric Power Materials Considering Positioning Errors

Abstract

The present invention discloses a design method for an electronic fence of electric power materials considering positioning errors, including: determining the tag position by using the received signal strength indicator value of low-power Bluetooth; measuring the time difference of arrival by using ultra-wideband technology; using unscented Kalman filter to fuse the BLE received signal strength and the TDOA value measured by UWB to calculate the updated material positioning state vector; designing the regional range of the electronic fence considering the measurement error of material positioning; and dynamically tracking, monitoring and warning the electric power materials through the electronic fence system. The present invention integrally uses two positioning technologies of low-power Bluetooth and ultra-wideband, greatly improving the tracking accuracy of the system; at the same time, perfectly integrating the electronic fence technology to effectively prevent and timely warn against violations or emergencies of materials, significantly improving the accuracy rate, management efficiency and flexibility of the material inbound and outbound trajectories.

Description

Technical Field

[0001] The present application relates to a design method for an electronic fence for high-precision warning of power material violations or emergencies, specifically a design method for an electronic fence for power materials considering positioning errors. Background Art

[0002] At present, the actual environment of power warehouses is complex. Using positioning methods based on positioning algorithms results in inaccurate positioning accuracy, untimely alarm signals, and waste of warning resources. The design method for an electronic fence for power materials considering positioning errors perfectly integrates Bluetooth (BLE) and ultra-wideband (UWB) indoor positioning technologies. Different from indoor positioning technologies such as infrared, ultrasonic, and radio frequency identification, UWB technology has many advantages such as fast transmission rate, low power consumption, relatively low overall cost, good anti-interference effect, ideal safety indicators, and ideal accuracy. In addition, UWB technology can efficiently develop existing spectrum resources, effectively address defects such as overcrowding in the spectrum, and at the same time can effectively solve the positioning requirements in complex indoor scenarios. Therefore, UWB technology is very suitable for the real-time positioning of power warehouse materials. BLE technology is a short-range radio technology that can achieve point-to-point communication through Bluetooth pairing. Integrating the use of two positioning technologies, low-power Bluetooth (BLE) and ultra-wideband (UWB), can greatly improve the tracking accuracy of the system.

[0003] The electronic fence technology uses a virtual fence to enclose a virtual geographical boundary on an electronic map. By comparing the position information returned by the material positioning system integrating the BLE-UWB technology with the pre-set electronic fence area, the current state of the target material is determined, and an alarm signal is sent or closed in a timely manner, thereby improving the monitoring and management efficiency of the material's inbound and outbound trajectories. Summary of the Invention

[0004] Aiming at the phenomena of inaccurate positioning accuracy, untimely alarm signals, and waste of warning resources in the actual complex environment of power warehouses based on positioning algorithms, the purpose of the present invention is to provide a design method for an electronic fence for power materials considering positioning errors to improve the accuracy, management efficiency, and flexibility of the material's inbound and outbound trajectories.

[0005] The purpose of the present invention is achieved through the following technical solutions:

[0006] A design method for an electronic fence for power materials considering positioning errors includes:

[0007] (1) Using the received signal strength indicator (RSSI) value of low-power Bluetooth (BLE) to determine the tag position;

[0008] (2) Using ultra-wideband (UWB) technology to measure the time difference of arrival (TDOA);

[0009] (3) Use the Unscented Kalman Filter (UKF) to fuse the BLE Received Signal Strength Indicator (RSSI) and the TDOA value measured by UWB to calculate the updated material positioning state vector;

[0010] (4) Consider the measurement error of material positioning and design the area range of the electronic fence;

[0011] (5) Dynamically track, monitor, and warn about power materials through the electronic fence system.

[0012] Further, the BLE Received Signal Strength Indicator (RSSI) in step (1) estimates the signal attenuation distance through the Friis Free-Space Path Loss (FSPL) model.

[0013] Further, in step (1), the likelihood constraint positioning model is used to impose linear constraints on the positioning.

[0014] Further, in step (2), the UWB part measures the time of arrival through the base station, and further calculates the time difference of arrival. All base stations should maintain unified time bases.

[0015] Further, in step (2), calibration measurements are performed on the clock frequency deviation caused by the crystal oscillator tolerance and stability through clock drift and nominal frequency difference correction.

[0016] Further, step (3) specifically includes:

[0017] (3-1) Model the local material as a dynamic system, and the state of the system at a given time k is described as a state vector x = [x v x y v y .

[0018] (3-2) In the time update stage of the UKF, use the Discrete White Noise Acceleration (DWNA) motion model to predict the speed and position of the current material.

[0019] (3-3) In the measurement update stage of the UKF, calculate the predicted measurement result and the corresponding covariance matrix by propagating the state vector through the assumed sensor model, and then update the prediction of the state vector according to the measurement result returned by the system.

[0020] (3-4) The mean value of the updated state vector and the covariance matrix P k(+) are the final results of the UKF iteration.

[0021] Further, step (4) specifically includes:

[0022] (4-1) Since the movement of materials follows a multivariate normal distribution at each sampling point, considering the measurement error of material positioning, the electronic fence can be set as Under different measurement errors, the set electronic fence area will also be different.

[0023] (4-2) The S of each partition unit r can be flexibly set according to the importance of the deviation in each direction or the actual warning requirements of power enterprises r,1 , S r,2 ,..., S r,n .

[0024] Furthermore, in step (5), according to the characteristics of the integrated BLE-UWB technology and the actual needs of power enterprises, the proposed electronic fence system mainly consists of front-end devices and a background integrated management platform server to achieve the purpose of dynamically tracking, monitoring, and warning power materials.

[0025] The present invention is based on Bluetooth (BLE) and ultra-wideband (UWB) technologies, uses RSSI and TDOA values, and fully considers the positioning error, thereby improving the efficiency and accuracy of material management. The electronic fence system for power materials considering positioning error aims to process security warning information, can set electronic fence detectors for materials outside the established transportation track, form an electronic wall, and effectively prevent and timely warn against violations or emergencies of materials. Circle the range on the system construction map, compare with the electronic fence range by means of the coordinates returned by the material positioning system, continuously calculate whether the target exceeds this range, and if it exceeds this range, an early warning signal will be automatically sent, which can play a certain positive role in ensuring the safety of materials. According to the actual application requirements, the threshold of the warning area can also be flexibly set to achieve the intelligent management of the material warehouse.

[0026] Advantages of the present invention:

[0027] Through the method of the present invention, by integrating and using two positioning technologies of low-power Bluetooth (BLE) and ultra-wideband (UWB), the tracking accuracy of the system is greatly improved; at the same time, the electronic fence technology is perfectly integrated to effectively prevent and timely warn against violations or emergencies of materials, significantly improving the accuracy, management efficiency, and flexibility of the material inbound and outbound trajectory. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 is the flowchart of the method of the present invention;

[0029] Figure 2 is the working data flow diagram of the electronic fence designed by using the method of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0030] This embodiment provides a design method for an electronic fence of electric power materials considering positioning errors, as Figure 1 shown, including:

[0031] (1) Determine the tag position using the received signal strength indication (RSSI) index value of low-power Bluetooth (BLE), specifically including:

[0032] (1-1) In this embodiment, an algorithm commonly used in the BLE positioning field is selected, that is, the signal attenuation distance is estimated using the RSSI index value, and then the tag position is determined. The attenuation of the signal with distance can be described by the Friis free-space path loss (FSPL) model. According to different operating environments, the secondary path loss exponent is an adjustable number γ. In addition, assuming that the antennas of the transmitter and receiver are isotropic, the specific signal strength 0 at a distance d can be calibrated:

[0033]

[0034] (1-2) In the case of performing positioning in a road or a similar geometry, linear constraints are usually imposed on the positioning. The likelihood constraint positioning model is used, that is, the likelihood function is derived from the RSSI distance model, a circular density function L(x) is created around each beacon, the radius is determined according to the measurement data, and each given measurement value is summarized:

[0035]

[0036]

[0037] where b i is the beacon position and A is the linear constraint matrix. This non-convex function can be minimized under the given constraints.

[0038] (2) Measure the time difference of arrival (TDOA) using ultra-wideband (UWB) technology, specifically including:

[0039] (2-1) In this embodiment, the TDOA value is calculated using the arrival time measured by the base station. To obtain correct results, all independently measured arrival times should be based on the same period. The base stations are equipped with oscillators with different frequency tolerances and stabilities, and their frequencies will drift due to temperature changes. To provide high positioning accuracy, a synchronization program under a hybrid solution is used to compensate for these offsets. The synchronization of the base stations is achieved by using reference base stations placed at known positions.

[0040] (2-2) In this embodiment, calibration measurement is performed through clock drift and nominal frequency difference correction. Two reference base stations, a standard base station, and tags are placed at known positions, and a set of TDOA measurement results are collected. By averaging these results and subtracting them, the TDOA is calculated from the base station coordinates and the internal reception delay δ of the base station m,R Calculate the TDOA n,m :

[0041]

[0042] Wherein, represents the reception time when the transmitter l sends the first data packet to the base station j (l can represent PR (primary base station), SR (slave base station), or T (tag)); T D represents the transmission delay of the synchronization data packet at the slave base station; T R1 represents the reference period when the primary base station sends the first data packet, T Rm represents the reference period transmitted by the slave base station, T Mm and T Mn respectively represent the reference periods measured by AN m and AN n ; and represent the periods from the previous transmission period, so the "-" is marked as a superscript; t Sn represents the propagation time between the slave base station and the base station AN n .

[0043] (3) Use the unscented Kalman filter (UKF) to fuse the BLE received signal strength (RSSI) and the TDOA value measured by UWB, and calculate the updated material positioning state vector, specifically including:

[0044] (3-1) In this embodiment, when applying the unscented Kalman filter algorithm, the local material is modeled as a dynamic system. The state of the system at a given time k is described as a state vector, which contains information about the material position (x, y coordinates) and the moving speed (speed components v x , v y ):

[0045] x = [x v x y v y (5)

[0046] (3-2) In this embodiment, in the time update stage of the UKF, the discrete white noise acceleration (DWNA) motion model is used to predict the speed and position of the current material. According to this model, the movement of the material between the two analyzed moments is uniformly linear, and the acceleration of the material is regarded as white noise. The time update stage is implemented using the following two formulas:

[0047]

[0048] P k(-) = FP k-1(+) F T + Q(7)

[0049] wherein, and are the predicted state vector value at a given time k and the result of the previous UKF iteration respectively; the matrices P k(-) and P k-1(+) are the corresponding covariance matrices; F is the state transition matrix, and Q is the process noise covariance matrix.

[0050] (3-3) In this embodiment, in the measurement update stage of the UKF, the prediction of the state vector is updated according to the measurement result returned by the system. The updated state vector is a combination of the predicted value and the difference between the actual measurement result and the result obtained by multiplying the predicted material position by the Kalman gain K. Performing the above operations requires pre-computing the predicted measurement result and the corresponding covariance matrix. It can be completed by propagating the state vector through the assumed sensor model:

[0051] h(x k ) + v k = [RSSI 1 (x k )... RSSI m (x k ) TDOA 1 (x k )... TDOA n (x k )] + v k (8)

[0052] wherein, RSSI i (x) and TDOA j (x) are the RSSI value of BLE and the TDOA value measured by UWB respectively.

[0053] (3-4) In this embodiment, the updated state vector mean and the covariance matrix P k(+) , are the final results of the UKF iteration:

[0054]

[0055] P k(+) = P k(-) - KPK T (10)

[0056] wherein, K is the Kalman gain; z kis a vector that contains all the measurement results returned by the system.

[0057] (4) Considering the measurement error of material positioning, design the area range of the electronic fence, specifically including:

[0058] (4-1) In this embodiment, from the above Kalman filter estimation formula, the estimation formula for the updated material positioning based on RSSI and TDOA values is and the corresponding estimated covariance matrix is P k(+) ; at the same time, from the discrete white noise acceleration (DWNA) motion model, it can be known that the movement of the material is regarded as a white noise process. Therefore, the movement of the material should follow a multivariate normal distribution at each sampling point. Considering the measurement error of material positioning, the electronic fence can be set as:

[0059]

[0060]

[0061] Among them, B represents the set area of the electronic fence. By comparing with B, it can be judged whether the movement track of the material meets the requirements of the established transportation range. If it does not meet the requirements, the electronic fence will immediately send out a warning signal to indicate that there is an abnormality in the material, form and send an alarm record to the background integrated management platform of the system; if it meets the setting requirements, the electronic fence will not send out an alarm message, but form a normal movement track of the material.

[0062] (4-2) In this embodiment, represents the transpose of P k(+) , S r is the diagonal matrix of each sample point, then represents the measurement error term of material positioning. Under different measurement errors, the set area of the electronic fence will also be different. Set S of each partition unit r according to the actual warning requirements of the power enterprise r,1 , S r,2 ,..., S r,n , S r,i (i = 1,..., n). The smaller the value, the smaller the warning area, and the stricter the monitoring of the material. In addition, the value of S r,i can also be flexibly set according to the importance of the deviation in each direction to maximize the utilization of the base station signal, thereby avoiding ineffective alarms and waste of public communication resources.

[0063] (5) Dynamically track, monitor and warn power materials through the electronic fence system, specifically including:

[0064] (5-1) In this embodiment, according to the characteristics of the integrated BLE-UWB technology and the actual needs of power enterprises, the proposed electronic fence system mainly consists of front-end devices and a background integrated management platform server. The background server sends instructions to the front-end devices. After receiving the instructions, the front-end devices perform corresponding operations and upload the collected material status data to the server to achieve the communication and transmission process between the two, so as to achieve the purpose of dynamically tracking, monitoring, and warning power materials.

[0065] (5-2) In this embodiment, when the detected material activity position is within the preset electronic fence range, the front-end device displays normal transportation, and the background server forms a material activity trajectory after receiving the data. Once the detector detects an abnormality, that is, the detected material activity position does not fall within the range preset by the electronic fence, the front-end device will immediately send an alarm signal to warn of the abnormal material, form and send an alarm record to the background integrated management platform, and track whether this event is properly resolved within a certain time interval in the subsequent period, thus playing a role in ensuring the safety of materials in a timely manner.

[0066] (5-3) In this embodiment, the system can formulate a flexible and unified alarm level according to the importance of materials to restrict the rationality of sending alarm information. The background integrated management platform server will save the data of each alarm information to facilitate the warehouse administrator to conduct data analysis and problem tracking and feedback regularly. Figure 2 It is the working data flow of the electronic fence.

[0067] Through the method of the present invention, by integrating and using two positioning technologies of low-power Bluetooth (BLE) and ultra-wideband (UWB), the tracking accuracy of the system is greatly improved; at the same time, the electronic fence technology is perfectly integrated to effectively prevent and timely warn of violations or emergencies of materials, significantly improving the accuracy, management efficiency, and flexibility of the material inbound and outbound trajectories.

Claims

1. A design method of an electronic fence for electric power materials considering positioning errors, characterized in that the method comprises the following steps: (1) Determine the tag position using the received signal strength indicator value of low-power Bluetooth; (2) Measure the time difference of arrival using ultra-wideband technology; (3) Use unscented Kalman filtering to fuse the received signal strength of low-power Bluetooth and the measured time difference of arrival of ultra-wideband, and calculate the updated material positioning state vector; (4) Consider the measurement error of material positioning and design the area range of the electronic fence; (5) Dynamically track, monitor and give early warnings to electric power materials through the electronic fence system; Step (3) specifically includes: (3-1)Model the local materials as a dynamic system. The state of the system at a given time k is described as a state vector, which contains information about the x and y coordinates of the material's position and the moving speed, with speed components v x , v y : The state of the system at a given time k is described as a state vector x = [x v x y v y ; (3-2) In the time update stage of UKF, use the discrete white noise acceleration motion model to predict the speed and position of the current material; (3-3) In the measurement update stage of UKF, calculate the predicted measurement result and the corresponding covariance matrix by propagating the state vector through the assumed sensor model, and then update the prediction of the state vector according to the measurement result returned by the system; (3-4) The mean of the updated state vector and the covariance matrix P k(+) , which is the final result of the UKF iteration; Step (4) specifically includes: (4-1) Since the movement of materials follows a multivariate normal distribution at each sampling point, considering the measurement error of material positioning, the electronic fence is set as Under different measurement errors, the set area of the electronic fence will also be different; in the formula, B represents the set area of the electronic fence, represents the transpose of P k(+) S r is the diagonal matrix of each sample point, then represents the measurement error term of material positioning. Under different measurement errors, the set area of the electronic fence will also be different; (4-2) Flexibly set the S of each partition unit r according to the importance of the deviations in each direction or the actual warning requirements of the power enterprise r,1 , S r,2 , …, S r,n .

2. The design method of an electronic fence for electric power materials considering positioning errors according to claim 1, characterized in that the signal attenuation distance of the received signal strength of low-power Bluetooth in step (1) is estimated by the Friis free space path loss model.

3. The design method of an electronic fence for electric power materials considering positioning errors according to claim 1, characterized in that a likelihood constraint positioning model is used in step (1) to impose linear constraints on the positioning.

4. The design method of an electronic fence for electric power materials considering positioning errors according to claim 1, characterized in that in step (2), the ultra-wideband part measures the time of arrival through the base station, further calculates the time difference of arrival, and all base stations keep the time base unified.

5. The design method of an electronic fence for electric power materials considering positioning errors according to claim 1, characterized in that in step (2), calibration measurement of the clock frequency deviation caused by the crystal oscillator tolerance and stability is carried out through clock drift and nominal frequency difference correction.

6. The design method of an electronic fence for electric power materials considering positioning errors according to claim 1, characterized in that in step (5), according to the integration of the technical characteristics of BLE-UWB and the actual needs of electric power enterprises, the proposed electronic fence system is composed of front-end devices and a background integrated management platform server, realizing dynamic tracking, monitoring and early warning of electric power materials.

Images