CN-121978971-A - Lifting platform obstacle avoidance control method and system suitable for transformer substation live working complex space

CN121978971ACN 121978971 ACN121978971 ACN 121978971ACN-121978971-A

Abstract

The invention provides a lifting platform obstacle avoidance control method and system suitable for a complex space of live working of a transformer substation, and relates to the technical field of obstacle avoidance control. In the live lifting operation process, environment sensing data of an operation space are obtained in parallel through a distributed multi-mode sensor network, sensing flow data and mechanical flow data with uniform time stamps are generated, synchronous alignment is carried out on the sensing flow data and the mechanical flow data, a feature alignment matrix is constructed, a collaborative environment self-adaptive entropy index for representing the operation reliability of a lifting platform is calculated based on the feature alignment matrix, a differentiated lifting driving control strategy is triggered according to the matching relation between the collaborative environment self-adaptive entropy index and a preset risk threshold, and the weighting parameters of the collaborative environment self-adaptive entropy index are dynamically optimized on line based on the execution feedback of the lifting driving control strategy so as to realize closed-loop self-adaptive adjustment. The method can be used for deeply coupling electromagnetic sensing, visual topology and dynamic stress and has closed-loop optimization capability.

Inventors

CHEN JINGSHI
ZHANG SHUO
LI QUANQUAN
WU YONG
CHEN GONGXUN
DUAN ZHUCHAO
YANG ZENGXIANG
ZHANG SHUAIFEI
GU ZHAOYANG
ZHENG YUHANG

Assignees

国网江苏省电力有限公司
国网江苏省电力有限公司建设分公司

Dates

Publication Date: 20260505
Application Date: 20260409

Claims (10)

1. The lifting platform obstacle avoidance control method suitable for the complex space of the live working of the transformer substation is characterized by comprising the following steps of: Step S1, in the live lifting operation process, environment sensing data of an operation space are obtained in parallel through a distributed multi-mode sensor network, and sensing flow data and mechanical flow data with uniform time stamps are generated; S2, synchronously aligning the perceived flow data and the mechanical flow data to construct a characteristic alignment matrix; Step S3, calculating a collaborative environment self-adaptive entropy index for representing the operation reliability of the lifting platform based on the characteristic alignment matrix; Step S4, triggering a differentiated lifting drive control strategy according to the matching relation between the self-adaptive entropy index of the collaborative environment and a preset risk threshold; and S5, dynamically optimizing the weighting parameters of the adaptive entropy indexes of the collaborative environment on line based on the execution feedback of the lifting driving control strategy so as to realize closed-loop adaptive adjustment.
2. The lifting platform obstacle avoidance control method suitable for the transformer substation live working complex space is characterized in that in the step S1, the sensing flow data comprise visual sensing data, laser radar ranging data and electric field sensing data, and the mechanical flow data comprise angular speed data acquired by an inertia measuring unit and current feedback data of a motor driver.
3. The lifting platform obstacle avoidance control method suitable for the complex space of the live working of the transformer substation, which is characterized in that the visual sensing data comprises a live body edge intensity mapping chart and a reflectivity distribution gradient of the surface of an insulator string; analyzing laser radar ranging data of a preset quantity scale to form space obstacle topology information, wherein the space obstacle topology information comprises a three-dimensional space bounding box of a charged body under a preset coordinate system and a minimum discharge safety critical distance between the charged body and a lifting platform; The electric field sensing data are acquired in real time through an electric field sensor array arranged on the lifting platform.
4. The lifting platform obstacle avoidance control method suitable for the complex space of the live working of the transformer substation according to claim 3, wherein the compensation moment applied by a driving motor for maintaining the posture balance of the lifting platform is obtained based on the current feedback data mapping of a motor driver, the posture shaking variable quantity of the lifting platform is reflected by the angular velocity data acquired by the inertial measurement unit, and the environmental load information of the lifting platform is obtained by combining the compensation moment and the posture shaking variable quantity; From the spatial obstacle topology information And environmental load information to which the lifting platform is subjected And forming a characteristic alignment matrix.
5. The method for controlling obstacle avoidance of a lifting platform suitable for a complex space of live working of a transformer substation according to claim 4, wherein in step S3, the collaborative environment adaptive entropy index is The calculation formula of (2) is as follows: In the formula, The method comprises the steps of quantifying the authenticity of perceived stream data in an electromagnetic interference environment for a perceived confidence evaluation factor; the dynamic instability entropy is used for quantifying the instability degree of the lifting platform, which is disturbed by the external environment when the lifting platform executes the operation; The gradient of the electric field intensity in space, which is obtained in real time for the electric field sensor array, is used for representing the change rate of the electric field in local space; Is a weighting parameter; Is an environmental sensitivity coefficient; Is the minimum discharge safety critical distance between the charged body and the lift platform.
6. The lifting platform obstacle avoidance control method suitable for a complex space of live working of a transformer substation according to claim 5, wherein the perception confidence evaluation factor is And (3) extracting signal-to-noise ratio characteristics of the edge profile of the charged body and reflectivity distribution characteristics of the surface of the insulator string, and performing weighted mapping to obtain the surface-mounted insulator string.
7. The lifting platform obstacle avoidance control method suitable for the complex space of live working of the transformer substation according to claim 5, wherein the mechanical dynamic instability entropy is The method comprises the steps of acquiring angular velocity data by an inertia measurement unit, and carrying out integral operation in a time domain on the product of the angular velocity data acquired by the inertia measurement unit and a compensation moment applied by a motor driver to maintain the posture balance of a lifting platform, wherein the compensation moment is obtained by mapping current feedback data of the motor driver in real time, and the frequency of the integral operation is kept synchronous with the sampling frequency of the angular velocity data acquired by the inertia measurement unit.
8. The method for controlling obstacle avoidance of a lifting platform suitable for a complex space of live working of a transformer substation according to claim 1 or 5, wherein in step S4, when the environment adaptive entropy index is coordinated When the motion track is smaller than a preset first-level risk threshold value, the lifting platform keeps a preset motion track; when (when) When the lift platform is in the first-level and second-level risk threshold value interval, the lift platform executes a deceleration distance increasing strategy And when the emergency braking and attitude locking actions are larger than the secondary risk threshold, the lifting platform executes the emergency braking and attitude locking actions.
9. The lifting platform obstacle avoidance control method suitable for the complex space of live working of the transformer substation according to claim 6, wherein step S5 specifically comprises: Real-time comparison of the collaborative environment adaptive entropy index And (3) calculating and controlling tracking errors according to the early warning level of the lifting platform and the residual shaking amplitude after the action of the lifting platform is executed ; Building tracking error based on the control Is an adaptive update function of (a) On-line iterative updating of the weighting parameters by gradient descent And (3) with 。
10. A lifting platform obstacle avoidance control system for executing the lifting platform obstacle avoidance control method applicable to the complex space of live working of a transformer substation according to any one of claims 1 to 9, the lifting obstacle avoidance control system comprising: The distributed multi-mode sensor network is composed of a visual sensor, a laser radar array, an electric field sensor array and an inertial measurement unit and is used for acquiring environmental sensing data of a working space; the system comprises a lifting platform, a visual sensor, an inertial measurement unit, a laser radar array, an electric field sensor array, an inertial measurement unit and a control unit, wherein the visual sensor is arranged at the tail end of the lifting platform and is used for acquiring the reflectivity distribution gradient of the surface of an insulator string and acquiring an electrified body edge intensity mapping diagram; The data processing module is used for summarizing all data acquired by the distributed multi-mode sensor network, generating sensing flow data and mechanical flow data with uniform time stamps, synchronously aligning the sensing flow data and the mechanical flow data, and constructing a characteristic alignment matrix; the operation module is used for calculating a collaborative environment self-adaptive entropy index for representing the operation reliability of the lifting platform based on the characteristic alignment matrix; The execution driving module is used for triggering a differentiated lifting driving control strategy according to the matching relation between the self-adaptive entropy index of the collaborative environment and a preset risk threshold; And the execution feedback module dynamically optimizes the weighting parameters of the adaptive entropy indexes of the collaborative environment on line based on the execution feedback of the lifting drive control strategy so as to realize closed-loop adaptive adjustment.

Description

Lifting platform obstacle avoidance control method and system suitable for transformer substation live working complex space Technical Field The invention relates to the technical field of obstacle avoidance control, in particular to an obstacle avoidance control method and system for a lifting platform suitable for a complex space of live working of a transformer substation. Background In substation live working, the lifting platform is in a high-voltage electric field, strong electromagnetic interference, complex metal member stacking and turbulence wind field environment. The bottleneck of the existing obstacle avoidance technology is as follows: 1. The strong high-voltage electromagnetic field exists in the transformer substation, so that noise is easily generated by the visual sensor, and tiny deviation is easily generated in laser radar ranging, and accurate and reliable obstacle information is difficult to provide by the traditional single sensing means. 2. In high-altitude operation, unexpected shaking of the platform can be caused by disturbance of wind load and transmission clearance of the lifting mechanism. If the obstacle avoidance algorithm cannot quantify the mechanical instability in real time, discharge accidents are easily caused by instant displacement when the electric charge body is approaching. 3. The existing obstacle avoidance logic is mostly based on hard cutting of static distance, lacks dynamic sensing and closed-loop self-adaptive adjusting capability for environmental risk, and cannot optimize subsequent control weight according to the fluctuation of environment response after execution. Therefore, a cooperative obstacle avoidance scheme capable of deep coupling electromagnetic sensing, visual topology, dynamic stress and having closed loop optimization capability is needed. Disclosure of Invention The invention aims to provide a lifting platform obstacle avoidance control method suitable for a complex space of live working of a transformer substation, and further provides a system for realizing the method, wherein quantitative characterization and intelligent hedging of perceived errors and dynamic interference in the live working lifting obstacle avoidance process of the transformer substation are realized through a distributed multi-mode perception and closed-loop entropy self-adaptive mechanism. In order to achieve the above purpose, the present invention provides the following technical solutions: a lifting platform obstacle avoidance control method suitable for a transformer substation live working complex space comprises the following steps: Step S1, in the live lifting operation process, environment sensing data of an operation space are obtained in parallel through a distributed multi-mode sensor network, and sensing flow data and mechanical flow data with uniform time stamps are generated; S2, synchronously aligning the perceived flow data and the mechanical flow data to construct a characteristic alignment matrix; Step S3, calculating a collaborative environment self-adaptive entropy index for representing the operation reliability of the lifting platform based on the characteristic alignment matrix; Step S4, triggering a differentiated lifting drive control strategy according to the matching relation between the self-adaptive entropy index of the collaborative environment and a preset risk threshold; and S5, dynamically optimizing the weighting parameters of the adaptive entropy indexes of the collaborative environment on line based on the execution feedback of the lifting driving control strategy so as to realize closed-loop adaptive adjustment. The invention is further improved in that in the step S1, the sensing flow data comprise visual sensing data, laser radar ranging data and electric field sensing data, and the mechanical flow data comprise angular velocity data acquired by an inertia measuring unit and current feedback data of a motor driver. The invention further improves that the visual sensing data specifically comprises a charged body edge intensity mapping chart and a reflectivity distribution gradient of the surface of the insulator string; analyzing laser radar ranging data of a preset quantity scale to form space obstacle topology information, wherein the space obstacle topology information comprises a three-dimensional space bounding box of a charged body under a preset coordinate system and a minimum discharge safety critical distance between the charged body and a lifting platform; The electric field sensing data are acquired in real time through an electric field sensor array arranged on the lifting platform. The invention is further improved in that the compensation moment applied by the driving motor for maintaining the posture balance of the lifting platform is obtained based on the current feedback data mapping of the motor driver, the posture shaking variable quantity of the lifting platform is reflected by the angular speed data acquired by