CN-122022637-A - Unattended intelligent coal shipment control method and device

CN122022637ACN 122022637 ACN122022637 ACN 122022637ACN-122022637-A

Abstract

The embodiment of the invention provides an unattended intelligent coal shipment control method and device, which relate to the technical field of automatic control, wherein the method comprises the following steps of generating a digital freight task based on a reservation request, generating a first control instruction of vehicle access authorization based on the digital freight task, performing cross verification based on entry perception data and the digital freight task, and generating a second control instruction after verification is passed; the method comprises the steps of comparing and analyzing first weighing data, generating a scheduling instruction based on an analysis result, collecting carriage form data of a vehicle in a loading position, generating a loading control instruction based on the carriage form data and a target loading capacity, dynamically adjusting the loading control instruction based on real-time feedback data, performing multidimensional verification on the generated calculation result, and generating electronic certificate data and a settlement triggering instruction. According to the scheme, the safety of the unattended intelligent coal shipment process is improved through full-flow automation and data automatic acquisition and binding.

Inventors

SUN JUN
Zuo Chunchang
YIN YUANYUAN
WANG JIXIN
GUO HUA
DUAN ZHIFEI
YANG XIMING

Assignees

华电煤业集团有限公司

Dates

Publication Date: 20260512
Application Date: 20260410

Claims (10)

1. An unattended intelligent coal shipment control method is characterized by comprising the following steps: Based on a reservation request of a client, generating and binding a digital freight task comprising a unique task code, generating a first control instruction of vehicle access authorization based on the digital freight task, collecting entry perception data of a vehicle, performing cross verification based on the entry perception data and the digital freight task, and generating a second control instruction allowing the vehicle to enter a weighing area after the verification is passed; in the weighing area, first weighing data of the vehicle are collected, the tare condition of the vehicle is analyzed according to the first weighing data and dynamic tare influence data of the vehicle, the task state is updated based on an analysis result, and a dispatching instruction is generated, wherein the dynamic tare influence data comprises current fuel quantity data and parking posture inclination angle of the vehicle; Based on the scheduling instruction, acquiring real-time carriage form data of a vehicle in a loading position, generating a loading control instruction based on the carriage form data and a target loading capacity, dynamically regulating and controlling loading operation in the loading process based on the loading control instruction so as to ensure the loading speed, the loading capacity and the leveling of the material level of coal in the carriage, dynamically regulating the loading control instruction based on real-time feedback coal loading state data in the loading process, and recording accumulated loading weight after loading is finished; And collecting second weighing data of the vehicle, calculating the net weight of the load based on the second weighing data, the accumulated loading weight and the first weighing data, and executing multi-dimensional verification on the generated calculation result, and if the verification is passed, generating electronic certificate data and a settlement triggering instruction.
2. The unattended intelligent coal shipment control method of claim 1, wherein collecting entry perception data of a vehicle, cross-checking with the digital freight mission based on the entry perception data, and generating a second control instruction allowing entry into a weighing area after passing the check, comprises: obtaining entrance perception data, wherein the entrance perception data comprises entrance license plate image data, driver identity text data and driver face image data; Identifying the entering license plate image data to obtain a first license plate number, and comparing the first license plate number with the reserved license plate number bound in the digital freight task for the first time; analyzing the driver identity text data to obtain first identity information, and performing second comparison on the first identity information and reserved driver information; Extracting features of the facial image data of the driver to obtain a first biological feature vector, and comparing the first biological feature vector with a standard biological feature vector extracted from the first identity information for the third time; And when the first comparison, the second comparison and the third comparison are successful, determining that the weighing area is allowed to be accessed, and generating the second control instruction.
3. The unattended intelligent coal shipment control method of claim 2, wherein the feature extraction of the driver facial image data to obtain a first biometric vector, and the third comparison of the first biometric vector with a standard biometric vector extracted from the first identity information comprises: Acquiring a time sequence image sequence of N frame face images through a face recognition camera with fixed sampling frequency f in a preset continuous verification time window T, wherein N=T×f; Extracting features of the face image of the driver of the first identity information to generate a standard biological feature vector; Performing living body detection and feature extraction on each frame image in the time sequence image sequence to obtain N biological feature vectors, and respectively calculating cosine similarity of the biological feature vectors and the standard biological feature vectors to generate N matching scores S i , wherein i is a serial number of a frame; Constructing a time sequence of the matching score S i based on a time axis, and inputting the time sequence into a dynamic decision fusion model; Calculating a standard deviation of a weighted moving average matching score S avg of the time sequence and the matching score S i of the time sequence within the verification time window T in the dynamic decision fusion model, taking the standard deviation as a stability index sigma, wherein S avg =(Σ(w i ×S i ))/Σw i ,w i is a weight coefficient, Alpha is a preset attenuation factor; If S avg ≥θ 1 is satisfied and sigma is less than or equal to theta 2 , the third comparison is successful, wherein theta 1 is a preset high matching degree threshold value, and theta 2 is a preset matching stability threshold value.
4. The unattended intelligent coal shipment control method according to claim 1, wherein analyzing the tare condition of the vehicle based on the first weighing data and the dynamic tare impact data of the vehicle, updating the task state based on the analysis result and generating the scheduling instruction, comprises: Extracting multiple historical tare data of a current vehicle through a first license plate number, and calculating a tare average mu and a tare standard deviation sigma of the historical tare data; calculating a dynamic tolerance threshold delta by the tare standard deviation sigma based on an intelligent fault tolerance decision method of multi-source sensor data fusion and regression analysis; Calculating a tare deviation aw of the first weighing data W current from the tare mean μ, wherein aw= |w current - μ|; comparing the tare deviation delta W with the dynamic tolerance threshold delta, if delta W is less than or equal to delta, judging that the tare deviation delta W is normal, and generating the scheduling instruction pointing to the loading area; If delta W > delta, judging that the device is abnormal, generating an abnormal alarm and triggering manual review.
5. The unattended intelligent coal shipment control method according to claim 4, wherein the dynamic tolerance threshold δ is calculated by the tare standard deviation σ based on an intelligent fault-tolerant decision method of multi-source sensor data fusion and regression analysis, comprising: when the first weighing data are collected, synchronously collecting current fuel quantity data F curr of the vehicle and a parking posture inclination angle theta on a weighing wagon balance; An expected tare deviation model is built, and expected tare deviation delta W expected =β 0 +β 1 ×(F curr -F ref )×ρ fuel ×V tank +β 2 multiplied by sin (theta) under the influence of multi-source working condition compensation data is calculated through the expected tare deviation model, wherein F curr is current fuel quantity data of a vehicle, F ref is reference fuel quantity adopted by the vehicle in standard tare statistics, rho fuel is standard density of the fuel, V tank is nominal total volume of a fuel tank of the vehicle, and beta 0 、β 1 、β 2 is a coefficient of the expected tare deviation model obtained based on multiple linear regression analysis training; And outputting a dynamic tolerance threshold delta by the intelligent fault tolerance decision method, wherein delta=K multiplied by delta++ delta W expected |, K is a expansion factor, and sigma is a tare standard deviation.
6. The unattended intelligent coal shipment control method of claim 1, wherein the generating of the loading control instruction based on the carriage morphology data and the target loading capacity, and the dynamic regulation of the loading operation during the loading based on the loading control instruction, to ensure the loading speed, the loading capacity and the level of the coal in the carriage, the dynamic regulation of the loading control instruction based on the coal loading state data fed back in real time during the loading, comprises: Panoramic scanning is carried out on the interior of a carriage of the vehicle to obtain dense point cloud data, and a carriage three-dimensional grid model is generated through the dense point cloud data; generating an initial loading control instruction through a cloth simulation algorithm based on the carriage three-dimensional grid model and the target loading capacity, wherein the initial loading control instruction comprises the moving speed of a loading chute, a preset moving track of the loading chute, the opening of a feed opening and a preset flow curve; Executing the initial loading control instruction, and collecting first feedback data and second feedback data in real time, wherein the first feedback data comprises instantaneous flow data and accumulated weight data of coal, and the second feedback data is real-time material surface profile data acquired by scanning equipment; Calculating weight progress deviation based on the accumulated weight data in the first feedback data and the preset flow curve, and calculating charge level flatness deviation based on the real-time charge level profile data in the second feedback data and the carriage form data; Dynamically correcting the moving speed of the loading chute and the opening of the discharging opening in the loading control instruction through a control algorithm based on the weight progress deviation and the charge level flatness deviation, so that the loading process converges to the preset moving track and the preset flow curve; and when the loading process is finished, recording the final accumulated weight data in the first feedback data as accumulated loading weight fed back by the loading equipment.
7. The unattended intelligent coal shipment control method of claim 6, wherein generating an initial loading control instruction by a cloth simulation algorithm based on the carriage three-dimensional grid model and the target loading amount comprises: performing discretization voxel processing on the carriage three-dimensional grid model to generate a three-dimensional voxel space for simulation; In the three-dimensional voxel space, establishing a kinematic and dynamic simulation model of the coal particle material based on a discrete element method; taking the target loading capacity as constraint, taking the highest flatness of the loading surface and/or the shortest loading time after loading is completed as an optimization target, and iteratively calculating the movement track of a loading chute and a global flow curve matched with the movement track in a kinematic and dynamic simulation model of the coal particle material; And reversely resolving and generating the initial loading control instruction according to the moving track and the global flow curve.
8. The unattended intelligent coal shipment control method according to claim 6, wherein dynamically correcting the movement speed of the loading chute and the opening of the discharge opening in the loading control instruction by a control algorithm based on the weight progress deviation and the charge level flatness deviation so that the loading process converges to the preset movement track and the preset flow rate curve comprises: defining a state vector X (T) = [ W acc (t),H avg (t),H std (t),P real (t)] T ], wherein W acc (T) is accumulated weight, H avg (T) is average material height calculated according to the real-time material surface profile data, H std (T) is standard deviation of material height, P real (T) is actual pose of a loading chute, T is continuous time, and T is matrix transposition; defining a control input vector U (t) = [ V (t), A (t) ] T , wherein V (t) is the moving speed of the loading chute, and A (t) is the opening of the feed opening; the dynamic correction is repeated in each control period k and in the next control period k+1 until the end: Defining an objective function J, and solving the objective function J as a target to obtain a control sequence meeting the objective function J based on a prediction model and a current state X (k), wherein, , N is the predicted time domain length, M is the control time domain length, For the accumulated weight of the ith step based on the state prediction of cycle k, As the accumulated weight reference value of the i-th step, The standard deviation of the material height of the ith step of the state prediction of the period k, For the actual pose of the loading chute of the ith step predicted based on the state of cycle k, Is the position and posture reference value of the loading chute in the ith step, In order to control the amount of change in the input, In order to weigh the weight coefficients of the weight tracking error, Is a weight coefficient of the flatness error of the material surface, Is a weight coefficient of the tracking error of the chute pose, Weight coefficients for controlling incremental changes; and controlling the execution mechanisms of the loading chute and the discharging opening through control instructions in the control sequence.
9. The unattended intelligent coal shipment control method of claim 1, wherein calculating the net weight of the load based on the second time weighing data, the accumulated loading weight, and the first time weighing data, and performing a multi-dimensional check on the generated calculation result, comprises: calculating a first difference value between the second weighing data and the first weighing data to obtain a net weight, calculating a second difference value between the net weight and the accumulated loading weight, and judging whether the second difference value is smaller than a preset physical loss threshold value; Comparing the net weight with the legal check load quality of the vehicle, and judging whether overload exists or not; according to the specific loading bin number for loading, searching actual loading coal types from a bin number and coal type mapping table, and comparing the actual loading coal types with designated coal types in the digital freight task; And when the second difference value is smaller than the physical loss threshold value, overload does not exist, and the actual coal type is consistent with the designated coal type, judging that the verification passes.
10. An unattended intelligent coal shipment control device, comprising: the admission verification module is used for generating and binding a digital freight task containing a unique task code based on a reservation request of a client, generating a first control instruction of vehicle admission authorization based on the digital freight task, collecting admission sensing data of a vehicle, performing cross verification based on the admission sensing data and the digital freight task, and generating a second control instruction allowing the vehicle to enter a weighing area after the verification is passed; the weighing control module is used for collecting first weighing data of the vehicle in the weighing area, analyzing the tare condition of the vehicle according to the first weighing data and dynamic tare influence data of the vehicle, updating the task state based on an analysis result and generating a scheduling instruction, wherein the dynamic tare influence data comprises current fuel quantity data and parking posture inclination angle of the vehicle; The automatic loading control module is used for acquiring real-time carriage form data of a vehicle in a loading place based on the scheduling instruction, generating a loading control instruction based on the carriage form data and a target loading capacity, dynamically regulating and controlling loading operation in the loading process based on the loading control instruction so as to ensure loading speed, loading capacity and leveling of a material level of coal in a carriage, dynamically regulating the loading control instruction based on real-time feedback coal loading state data in the loading process, and recording accumulated loading weight after loading is finished; the signing and settling module is used for collecting second weighing data of the vehicle, calculating the net weight of the vehicle based on the second weighing data, the accumulated loading weight and the first weighing data, executing multidimensional verification on the generated calculation result, and generating electronic certificate data and a settlement triggering instruction if the verification is passed.

Description

Unattended intelligent coal shipment control method and device Technical Field The invention relates to the technical field of automatic control, in particular to an unattended intelligent coal shipment control method and device. Background At present, road shipment links of bulk cargos such as coal, ore and the like generally depend on a large number of manual field operations and management. Conventional shipping modes typically include manual window handling, on-site personnel commanding the weighing and loading of the vehicle, driver interactive confirmation with the loader driver or under-the-shelf personnel, and the like. This highly dependent manual work and coordination mode suffers from the following significant drawbacks: 1. in the process of moving, weighing and loading vehicles in a factory, drivers are in close contact with on-site commanders and loading staff, serious safety accidents such as vehicle collision and personnel rolling are easily caused, and the safety risk is high. 2. In the weighing link, the problems of coal replacement, entrainment, insufficient loading or overload can occur in the loading link. These activities seriously impair the economic benefits of enterprises and are costly to manage due to difficulty in evidence collection. 3. The links of manual handling, queuing and waiting of vehicles, manual command and loading and the like are time-consuming and labor-consuming, and the turnover of vehicles is slow. Information of each link is split (such as a pound room, a dispatching and loading point), communication is dependent on telephones or interphones, information errors and dispatching confusion are easy to occur, peak period congestion is serious, and operation efficiency is low. 4. From reservation, entering, weighing, loading and settlement, the data of each link are recorded manually or dispersed in different systems, and a complete, real and untampered 'one-vehicle-one-gear' data chain is difficult to form. Once disputes occur, tracing and auditing are extremely difficult, so that a data chain is broken, and the tracing is difficult. Some existing improvement technologies, such as individual license plate recognition, unmanned pound-room or semi-automatic loading equipment, are often only optimized for a single link, cannot realize full-flow closed-loop automation and intellectualization from reservation to delivery, and cannot radically stop human interference, serial cheating and security management loopholes. Disclosure of Invention In view of the above, the embodiment of the invention provides an unattended intelligent coal shipment control method, which aims to solve the technical problems of artificial interference, serial cheating and security management loopholes in the coal shipment process in the prior art. The method comprises the following steps: Based on a reservation request of a client, generating and binding a digital freight task comprising a unique task code, generating a first control instruction of vehicle access authorization based on the digital freight task, collecting entry perception data of a vehicle, performing cross verification based on the entry perception data and the digital freight task, and generating a second control instruction allowing the vehicle to enter a weighing area after the verification is passed; in the weighing area, first weighing data of the vehicle are collected, the tare condition of the vehicle is analyzed according to the first weighing data and dynamic tare influence data of the vehicle, the task state is updated based on an analysis result, and a dispatching instruction is generated, wherein the dynamic tare influence data comprises current fuel quantity data and parking posture inclination angle of the vehicle; Based on the scheduling instruction, collecting real-time carriage form data of a vehicle in a loading position, generating a loading control instruction based on the carriage form data and a target loading capacity, dynamically regulating and controlling loading operation in the loading process based on the loading control instruction so as to ensure loading speed, loading capacity and leveling of a material level of coal in a carriage, dynamically regulating the loading control instruction based on coal loading state data fed back in real time in the loading process, and recording accumulated loading weight after loading is finished; And collecting second weighing data of the vehicle, calculating the net weight of the load based on the second weighing data, the accumulated loading weight and the first weighing data, and executing multi-dimensional verification on the generated calculation result, and if the verification is passed, generating electronic certificate data and a settlement triggering instruction. The embodiment of the invention also provides an unattended intelligent coal shipment control device, which is used for solving the technical problems of artificial interference, serial cheating an