A method for controlling the speed of a stage device in a gradual manner with safety constraints

Abstract

This invention relates to the field of stage equipment control technology, specifically disclosing a progressive speed control method for stage equipment that integrates safety constraints. By acquiring the motion characteristic information of the equipment in the current environment, performing risk assessment on abnormal information, constructing segmented safety thresholds based on dynamic safety position sensitivity, and finally obtaining a progressive speed with safety constraint mechanisms, this method solves the safety hazards caused by control system failures in existing methods. The method of this invention achieves low cost through lightweight algorithm design, significantly reducing hardware investment and deployment costs. Simultaneously, by combining dynamic safety distance updates with characteristic anomaly risk assessment, it greatly improves the resilience and fault tolerance of the control system, providing a high-precision and highly robust safety control solution for large stage machinery (such as stage wagons and rotating stages).

Description

Technical Field

[0001] The present invention relates to the technical field of stage equipment control, and particularly relates to a progressive stage equipment speed control method integrating safety constraints, and is particularly applicable to large-stage equipment such as seat carriages and rotating stages that require high-precision movement and have low safety redundancy. Background Art

[0002] With the realization of scale expansion in the performing arts market through all-round coverage, multi-dimensional deep cultivation, and high-level breakthroughs, it not only makes the complexity and challenge coefficient of performances continue to rise, but also makes the danger threshold of safety risks continuously decrease. Once there is an oversight in a certain link, it may trigger a chain reaction and lead to a tragedy, which is precisely the key to determining whether the industry can move forward steadily. As is well known, in iconic large-scale performances such as shows, drama tours, and variety shows, stage mechanical equipment, as the core equipment carrying the immersive experience of the audience, its operation accuracy and safety redundancy are directly related to the project presentation effect and the safety of the audience during the performance. Among them, the audience seat carriage, a special equipment integrating mechanical transmission, intelligent control, and ergonomic design, is not simply a carrying tool, but more like a link connecting the interaction between the audience and the stage, allowing the audience to experience the wonderful feeling of different scenes changing while moving. It is worth mentioning that the application scenario of the seat carriage is also particularly crucial in the overall system design. It generally needs to move to the required position at a high speed. Affected by the overall layout of the stage, the distance between this position and the building wall has been compressed to the critical value, almost without any physical buffer space, and there is no adjustment margin for emergency operations. Although this extreme design can bring a strong sense of plot substitution and visual impact to the audience, it also hides extremely high safety risks: once the control system of the equipment fails (such as program instruction errors, drive module failures, signal transmission interruptions, etc.), the carriage will lose effective restraint and rush towards the rear wall at full speed. In the light case, it will cause serious damage to the mechanical structure of the carriage and force the performance to be interrupted. In the heavy case, it may cause the seats to loosen due to a violent collision, the audience to be受惊 or even injured, resulting in immeasurable economic losses and adverse social impacts.

[0003] Therefore, there is an urgent need for a method that can dynamically sense the state of the equipment and the environment, multi-dimensionally evaluate risks, and achieve safety control with a progressive strategy to solve the problems of low safety redundancy, poor fault tolerance, and rigid control in the prior art. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing stage equipment control methods, such as static safety constraints, single risk assessment, and rigid control strategies. It provides a progressive stage equipment speed control method that integrates safety constraints. Through multi-source feature acquisition, dynamic risk assessment, dynamic safety distance calculation, and progressive speed adjustment, it achieves high-precision and robust control of the equipment under complex working conditions, while reducing hardware costs and improving system fault tolerance, thereby solving the aforementioned problems in the prior art.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A progressive stage equipment speed control method incorporating safety constraints includes:

[0007] The motion feature set is generated by collecting real-time state values ​​of stage equipment during movement using multi-source sensors.

[0008] Based on the motion feature set, determine the dynamic design index value and calculate the outlier parameter of each feature;

[0009] The outlier parameters are classified into different levels;

[0010] Using the anomaly level of each feature as the initial node, the process involves repeatedly merging the two nodes with the smallest outlier parameters until only one root node remains. The sum of weights W, excluding accumulated nodes, is calculated along the path from the child node to the root node. sum The sum of weights W including the accumulated nodes max The characteristic anomaly risk parameters are obtained;

[0011] The dynamic safety distance when the front end of the computing device is close to a wall;

[0012] The actual distance is calculated based on the total operating distance of the equipment to obtain the range risk coefficient;

[0013] The characteristic anomaly risk parameters and the range risk coefficients are normalized, and the overall system risk is calculated based on the normalization results.

[0014] Based on the overall system risk and equipment status, a progressive speed control strategy is determined, and the target speed for adjustment is determined.

[0015] Preferably, it includes the following steps:

[0016] S1: Real-time acquisition of state values ​​of stage equipment during movement via multi-source sensors to generate a motion feature set M(t) = [P(t), V(t), A(t), L(t), G(t), I(t)] T Where: P(t) is the current position of the device, V(t) is the current velocity, and A(t) is the current acceleration. For the current total load, Fi (t) Force sensor data acquisition value; Center of gravity offset G(t) = |Δx(t)| + |Δy(t)|, where Δx and Δy are the offsets of the load's center of gravity relative to the equipment's center, and x i y i The installation coordinates of the i-th sensor are represented; I(t) represents the current of the device.

[0017] S2: Determine the dynamic design index value based on the motion feature set M(t). Calculate the outlier parameters for each feature. ΔM max The maximum permissible deviation for the corresponding feature, i∈(1…N) + ), N + These are positive integers, corresponding to position, velocity, acceleration, load, center of gravity offset, and current characteristics, respectively.

[0018] S3: Regarding the outlier parameter M i un The levels are divided according to the following rules: If M i un =0, marked as level 0, no anomaly; if 0 <M i un ≤low(M i un ), marked as level 1, mild abnormality; if low(M i un ) <M i un ≤upp(M i un ), marked as level 2, moderately abnormal; if M i un >upp(M i un ), marked as level 3, severely abnormal, where low(M) i un ), upp(M i un These represent the lower and upper threshold values ​​for the corresponding outlier features;

[0019] S4: Using the anomaly level of each feature as the initial node, repeatedly merge the two nodes with the smallest outlier parameters until only one root node remains. The outlier value of the new node is the sum of the outlier values ​​of its two child nodes. Calculate the sum W of the weights of the child nodes on the path to the root node that does not include the accumulated nodes. sum The sum of weights W including the accumulated nodes max The characteristic anomaly risk parameter R is obtained. fea =W sum / W max ;

[0020] S5: Dynamic safety distance D when the front end of the calculation device is close to the wall. safe (t)=D react (t)+D per (t)+D buff (t), where D react (t)=V(t)*t react (t), t react (t) represents the response time. a eff (t) represents the maximum safe acceleration. k lod k fri k agi These are the load correction factor, friction correction factor, and equipment aging correction factor, respectively, and k lod +k fri +k agi =1, η=0.8 D is the importance coefficient. buff (t)=max(ΔD err ΔD off ),ΔD err For the positioning error, ΔD off (This refers to the lateral offset error);

[0021] S6: Based on the total operating distance of the equipment, D sum Calculate the actual distance D real (t)=D sum -P(t) yields the range risk coefficient R. dis =D real (t) / D safe (t);

[0022] S7: For R fea R dis After normalization, we obtain R respectively. nor-fea R nor-dis , and press R sys =γ fea ·R nor-fea +γ dis ·R nor-dis Comprehensive risk of computing system, γ fea =1-γ dis , and γ dis >1 / 2;

[0023] S8: According to R sys Determine a progressive speed control strategy based on equipment status, and follow... Adjust the target velocity, Ta = (V(t) - Vgoal ) / a eff (t) is the variable speed time, and μ = 8 / Ta is the speed curve change coefficient.

[0024] Preferably, the multi-source sensors described in step S1 include: a laser displacement sensor, an optoelectronic encoder, a piezoelectric force sensor, a gravity sensor, and a Hall current sensor; the laser displacement sensor collects P(t), the optoelectronic encoder collects V(t), A(t), the piezoelectric force sensor collects F i (t) to calculate L(t), the gravity sensor assists in correcting G(t), and the Hall current sensor collects I(t).

[0025] Preferably, the dynamic design index value in step S2 where is the initial design value of the feature, λ is the attenuation coefficient, the value range of λ is 0.1 - 0.3, T dec is the attenuation period, T dec takes a value of 5 - 10 s, is the average value of the historical 10 - 30 groups of feature sequences.

[0026] Preferably, in step S3: for the position feature, low(P un ) = 0.3, upp(P un ) = 0.6; for the speed feature, low(V un ) = 0.4, upp(V un ) = 0.8; for the acceleration feature, low(A un ) = 0.5, upp(A un ) = 1.0; for the load feature, low(L un ) = 1.0, upp(L un ) = 3.0; for the center of gravity offset feature, low(G un ) = 0.2, upp(G un ) = 0.5; for the current feature, low(I un ) = 0.8, upp(I un ) = 1.5.

[0027] Preferably, the t react (t) in step S5 is dynamically adjusted according to the operating conditions of the device: when V(t) ≤ 1 m / s, t react (t) = 0.15 s; when 1 m / s < V(t) ≤ 3 m / s, t react (t) = 0.2 s; when V(t) > 3 m / s, t react (t) = 0.25 s.

[0028] Preferably, the normalization formula described in step S7 is: where R fea-min = 0, R fea-max = 1, R dis-min = 0, R dis-max = 1.2.

[0029] Preferably, the progressive speed control strategy described in step S8 is specifically as follows:

[0030] If R sys < Sσ, Sσ is the R sys minimum value, and Sσ takes a value of 0.2 - 0.3: The device gives an early warning, maintains the current speed, and prompts an abnormal status;

[0031] If Sσ < R sys ≤ Eσ and Eσ is the R sys maximum value, and Eσ takes a value of 0.7 - 0.8: The device decelerates gently, and V goal is set to

[0032] If Sσ < R sys [[ID=???]]≤ Eσ and D real (t) < D safe (t): The device decelerates strongly, and V goal is set to 0 and stops gradually;

[0033] If R sys > Eσ and lim(D safe (t) → 0): The device stops suddenly, and the mechanical locking mechanism is started.

[0034] Preferably, it further includes step S9: The acquisition data, calculation results, and control instructions of each step are recorded in real time. After every 50 - 100 groups of data are accumulated, the parameter values of λ, T dec , k sys are optimized based on the least squares method to achieve model iterative update.

[0035] Preferably, when there are multiple devices operating in coordination on the stage, the safety distance D inter between devices needs to be added in step S5. At this time, D safe (t) = D react (t) + D per (t) + D buff (t) + D inter , D inter is calculated according to V(t) and A(t) of adjacent devices according to D inter = 0.5 * (V(t) + V adj (t)) * t react [[ID=8l]](t), V adj It seems there is a typo in the original text at line 38 where "≤ Eσ and D" should probably be something else for the correct logical flow. Please check and correct if needed.(t) represents the current speed of the adjacent device;

[0036] and / or

[0037] In step S1, the ambient temperature T(t) of the equipment is also collected, and in step S2, T(t) is included in the feature set M(t), along with the corresponding outlier parameters. For the dynamic design value of temperature, ΔT max =5℃, and the low(T) of the temperature characteristic in step S3 un =0.2, upp(T) un = 0.5.

[0038] The present invention has the following significant beneficial effects:

[0039] This invention presents a progressive stage equipment speed control method incorporating safety constraints. By acquiring the motion characteristics of the equipment in the current environment, assessing the risk of anomalies, and constructing segmented safety thresholds based on dynamic safety position sensitivity, a progressive speed with safety constraints is finally obtained. This solves the safety hazards caused by control system failures in existing methods. The invention achieves lower costs through a lightweight algorithm design, significantly reducing hardware investment and deployment costs. Furthermore, by combining dynamic safety distance updates with characteristic anomaly risk assessment, it greatly improves the resilience and fault tolerance of the control system, providing a high-precision, highly robust safety control solution for large stage machinery (such as stage wagons and rotating stages). Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the progressive stage equipment speed control method integrating safety constraints of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0042] like Figure 1 The present invention is shown in detail below:

[0043] 1. Multi-source feature acquisition (step S1)

[0044] A multi-source sensing system, consisting of a laser displacement sensor, photoelectric encoder, piezoelectric force sensor, gravity sensor, Hall current sensor, and temperature sensor, is used to collect real-time motion and environmental status data of stage equipment, generating a feature set M(t) = [P(t), V(t), A(t), L(t), G(t), I(t), T(t)]. T ;

[0045] in:

[0046] P(t): The current position of the device acquired by the laser displacement sensor, with an accuracy of ±0.1mm. ；

[0047] V(t), A(t): The current speed and acceleration of the device acquired by the photoelectric encoder, with a speed accuracy of ±0.01 m / s and an acceleration accuracy of ±0.05 m / s². 2 ；

[0048] L(t): The sum of load data collected by six distributed piezoelectric force sensors. Accuracy ±0.5kg;

[0049] G(t): The offset of the center of gravity calculated based on the coordinates of the force sensor and the load, G(t) = |Δx(t)| + |Δy(t)|, combined with the correction of the gravity sensor, with an accuracy of ±0.5mm;

[0050] I(t): Device drive current acquired by Hall current sensor, with an accuracy of ±0.1A;

[0051] T(t): The ambient temperature of the equipment's operation, collected by the temperature sensor, with an accuracy of ±0.5℃, used to correct the mechanical friction coefficient.

[0052] 2. Calculation of dynamic outliers (step S2)

[0053] To avoid the problem that fixed thresholds cannot adapt to dynamic operating conditions, dynamic design index values ​​are introduced. The calculation formula is:

[0054]

[0055] in:

[0056] Characteristic initial design values ​​(such as initial velocity design value V) init =2m / s);

[0057] λ: Attenuation coefficient (value ranges from 0.1 to 0.3; the larger the load, the larger λ becomes to accelerate the update speed of design values);

[0058] T dec Attenuation period (value 5-10s, adjusted according to equipment stability, larger value for stable operation);

[0059] The average value of 10-30 historical characteristic sequences is used to avoid fluctuations in design values ​​caused by instantaneous interference.

[0060] based on The outlier parameters for each feature are calculated using the following formula:

[0061]

[0062] Where: ΔM max The maximum permissible deviation for each feature (e.g., position ΔP) max =5mm, speed ΔV max =0.3m / s), M i un The larger the value, the more severe the abnormality of the corresponding feature.

[0063] 3. Anomaly Level Classification (Step S3)

[0064] To achieve quantitative risk classification, based on extensive experimental data and equipment safety standards, classification thresholds for each characteristic outlier were determined, as shown in Table 1 below:

[0065] Table 1. Thresholds for classifying outliers for each feature.

[0066]

[0067] 4. Calculation of anomaly risk (Step S4)

[0068] Hierarchical clustering is used to fuse the anomaly levels of each feature in order to quantify the overall feature risk.

[0069] The anomaly levels (including temperature) of the seven features are used as initial nodes, and the weight of each node is its anomaly parameter (e.g., the weight of a level 1 anomaly is 0.2).

[0070] Repeatedly merge the two nodes with the smallest weights, and the weight of the new node is the sum of the weights of the two child nodes (e.g., merge the temperature node with a weight of 0.1 and the location node with a weight of 0.2, and the weight of the new node is 0.3).

[0071] Until only one root node remains, calculate the sum of the weights W of the "non-cumulative nodes" on the path from the child node to the root node. sum The sum of the weights of the initial feature nodes (i.e., the total weights of the initial feature nodes) and the sum of the weights of the nodes "including the cumulative nodes" (W) max (That is, the sum of the weights of all nodes);

[0072] Characteristic Anomaly Risk Parameter R fea =W sum / W max The value ranges from 0 to 1, with a larger value indicating a higher risk at the feature level.

[0073] 5. Dynamic safety distance calculation (step S5)

[0074] Breaking away from existing fixed safety distance designs, a dynamic safety distance D is calculated by combining equipment status and environmental parameters.safe (t), the formula is:

[0075] D safe (t)=D react (t)+D per (t)+D buff (t)+D inter ;

[0076] The calculation logic for each component is as follows:

[0077] Response distance D react (t): The distance the equipment moves from detecting an anomaly to initiating control, D react (t)=V(t)*t react (t), where t react (t) dynamically adjusted according to velocity (t when velocity ≤ 1m / s) react =0.15s, 0.2s when 1-3m / s, 0.25s when >3m / s);

[0078] Execution distance D per (t): The distance the equipment travels from start-up control to stop. Where a eff (t) represents the maximum safe acceleration (adjusted according to the load; the larger the load, the greater the acceleration). eff The smaller the value, such as when the load is ≤500kg, a eff =2m / s 2 When >500kg, a eff =1.5m / s 2 ), The system correction coefficient (η = 0.8) k is the importance coefficient. lod k fri k agi These are the correction factors for load, friction, and aging, respectively, and the sum of the three is 1.

[0079] Error margin D buff (t): Compensation for positioning and offset errors, D buff (t)=max(ΔD err ΔD off ), where ΔD err For the positioning error (±0.5mm), ΔD off The lateral offset error is ±0.3mm;

[0080] Distance between equipment D inter Security redundancy during multi-device collaborative operation, D inter =0.5*(V(t)+V adj(t))*t react (t)(V adj (t) represents the speed of adjacent devices), and D is the speed of a single device during operation. inter =0.

[0081] 6. Scope Risk Calculation (Step S6)

[0082] Based on the current location of the equipment and the total operating distance, calculate the actual distance D from the equipment to the wall (or target location). real (t)=D sum -P(t), (D sum Define the total operating distance of the equipment (e.g., 50m) and define the range risk factor R. dis =D real (t) / D safe (t):

[0083] If R dis >1: The actual distance is greater than the safe distance, and the risk within the area is low;

[0084] If 0.5 <R dis ≤1: The actual distance is close to the safe distance, and the risk within the range is moderate;

[0085] If R dis ≤0.5: The actual distance is less than the safe distance, indicating a high risk within the area.

[0086] 7. System Comprehensive Risk Calculation (Step S7)

[0087] For R fea With R fea Normalization is performed to eliminate dimensional differences:

[0088] Characteristic risk normalization: Among them, R fea-min =0, R fea-max =1;

[0089] Scope risk normalization: Where R dis-min =0, R dis-max =1.2 (covering R) dis >1 (Scenario).

[0090] Based on the principle of "prioritizing scope risk", a weighting coefficient γ is set. dis >1 / 2 (e.g., γ) dis =0.6, γ fea =0.4), calculate the overall risk of the system:

[0091] R sys =γ fea ·R nor-fea +γ dis ·Rnor-dis ;

[0092] R sys The value ranges from 0 to 1, and Sσ = 0.25 (minimum risk) and Eσ = 0.75 (maximum risk) are defined for subsequent control strategy determination.

[0093] 8. Progressive speed control (step S8)

[0094] According to R sys Based on the equipment status, a graded and gradual speed control strategy is formulated, as shown in Table 2 below:

[0095] Table 2. Progressive speed control strategy

[0096]

[0097]

[0098] The smooth transition of the target velocity is achieved through an exponential function, as shown in the formula:

[0099]

[0100] in:

[0101] Ta=(V(t)-V goal ) / a eff (t): Shift time (e.g., from 2m / s to 1m / s, a) eff =1m / s 2 If , then Ta = 1s);

[0102] μ = 8 / Ta: The velocity curve variation coefficient ensures that the curve reaches the inflection point at t = Ta / 2, achieving a smooth transition of "slow first, then fast" or "fast first, then slow".

[0103] 9. Iterative Model Optimization (Step S9)

[0104] To improve the long-term adaptability of the method, the collected data, calculation results, and control commands are recorded in real time. After accumulating 50 sets of valid data, the key parameters are optimized based on the least squares method.

[0105] Optimize λ, T dec To minimize the deviation between M(t) and the actual working condition;

[0106] Optimize k sys : Make D per The calculated value of (t) deviates from the actual stopping distance by less than 5%;

[0107] Optimize γ dis Adjust weights based on historical risk events; for example, increase γ when there is an increase in anomalies due to scope risk.dis Up to 0.7.

[0108] The present invention will be further described in detail below with reference to specific embodiments, the implementation object being a seating platform for a cultural and tourism resident performance (maximum load 800kg, maximum speed 3m / s, total running distance D). sum =50m, the distance between the target location and the wall is 0.8m).

[0109] 1. Implementation Preparation

[0110] Sensor configuration: Laser displacement sensor (model KeyenceIL-300), photoelectric encoder (model OmronE6B2-CWZ6C), piezoelectric force sensor (model HBM U9C), gravity sensor (model Bosch BMA250), Hall current sensor (model LEM LA55-P), temperature sensor (model PT100);

[0111] Initial parameter settings: λ=0.2, T dec =8s, a eff =2m / s 2 (Load ≤ 500 kg), γ dis =0.6, Sσ=0.25, Eσ=0.75.

[0112] 2. Implementation Steps

[0113] Step 1: Feature Acquisition

[0114] The characteristic data collected at a certain moment are: P(t) = 48m, V(t) = 2.5m / s, A(t) = 0.8m / s 2 , L(t)=600kg, G(t)=0.3mm, I(t)=15A,

[0115] Step 2: Outlier Calculation

[0116] Dynamic design index values

[0117] ( (The average of 15 historical data sets);

[0118] Speed ​​anomalies

[0119] (ΔV max =0.3m / s);

[0120] Other outlier characteristics: P un =0.2, A un =0.6, L un =1.2, Gun =0.3, I un =0.9, T un =0.6.

[0121] Step 3: Anomaly Level Classification

[0122] According to Table 1, the feature levels are: P (level 1), V (level 3), A (level 2), L (level 2), G (level 2), I (level 2), and T (level 2).

[0123] Step 4: Feature Risk Calculation

[0124] Initial node weights: 0.2 (P), 2.0 (V), 0.6 (A), 1.2 (L), 0.3 (G), 0.9 (I), 0.6 (T);

[0125] Merging process: First merge 0.2(P) and 0.3(G) → 0.5, then merge 0.5 and 0.6(A) → 1.1, then merge 1.1 and 0.6(T) → 1.7, then merge 1.7 and 0.9(I) → 2.6, then merge 2.6 and 1.2(L) → 3.8, and finally merge 3.8 and 2.0(V) → 5.8;

[0126] W sum =0.2+2.0+0.6+1.2+0.3+0.9+0.6=5.8,

[0127] W max =5.8+0.5+1.1+1.7+2.6+3.8=15.5, R fea =5.8 / 15.5≈0.37\).

[0128] Step 5: Calculate the safe distance

[0129] Response time t react =0.25s (V = 2.5m / s > 3m / s), D react =2.5 × 0.25 = 0.625 m;

[0130] If the load is 600kg > 500kg, a eff =1.5m / s 2 k lod =0.4, k fri =0.3, k agi =0.3, k sys =0.8×0.4+1.0×0.3+1.2×0.3=0.88,

[0131] D buff=max(0.0005,0.0003)=0.0005m, single device operation D inter =0;

[0132] D safe =0.625+1.83+0.0005≈2.455m.

[0133] Step 6: Scope Risk Calculation

[0134] Actual distance D real =50-48=2m;

[0135] R dis =2 / 2.455≈0.815.

[0136] Step 7: System Risk Calculation

[0137]

[0138] R sys =0.4×0.37+0.6×0.679≈0.56 (between Sσ=0.25 and Eσ=0.75).

[0139] Step 8: Speed ​​Control

[0140] Device status: R sys =0.56 and Determined to be of "low risk," a slight reduction in speed will be implemented.

[0141] Target speed V goal =0.8 × 1.9 = 1.52 m / s;

[0142] The shift time Ta = (2.5 - 1.52) / 1.5 ≈ 0.65 s, μ = 8 / 0.65 ≈ 12.3;

[0143] Speed ​​transition curve: It achieves a smooth decrease from 2.5 m / s to 1.52 m / s within 0.65 s.

[0144] Step 9: Model Optimization

[0145] After accumulating 50 sets of data, the least squares method was used to optimize λ = 0.22 (originally 0.2), making... The deviation from the actual speed was reduced from 0.6 m / s to 0.4 m / s, improving the accuracy of dynamic design indicators.

[0146] 3. Implementation Results

[0147] In this embodiment, when the speed of the seat platform exceeds the tolerance and the distance is close to the safety threshold, it achieves smooth control through gradual deceleration, without mechanical impact and without obvious discomfort to the audience; the system risk identification is accurate, with no false triggering or missed triggering, and no additional hardware cost, which meets the actual needs of the performance scenario.

[0148] The following is a specific embodiment of the present invention:

[0149] 1. Real-time acquisition and storage of the device's state values ​​during movement via multi-source sensors, generating a feature set M(t) = [P(t), V(t), A(t), L(t), G(t), I(t)...] T Where P(t) represents the current position; V(t) is the current velocity; A(t) is the current acceleration; and the current total load is... F represents the force sensor value; the center of gravity offset G(t) = |Δx(t)| + |Δy(t)|, where Δx and Δy are the offsets of the load's center of gravity relative to the equipment's center, and x i y i The installation coordinates of the i-th sensor are represented; I(t) represents the current of the device.

[0150] 2. Based on the equipment data acquisition value M with the above characteristics, dynamically design the index value. Determine if the current data is abnormal. Let M... un This is represented as an outlier parameter, indicating whether the currently collected data is considered abnormal. ΔM max M represents the maximum allowable deviation for each eigenvalue. i un The larger the value, the more severe the anomaly. i∈(1…N + () refers to the nth feature in the feature set, such as i=1 representing the positional feature. i=2 represents the velocity characteristic, i.e. i=3 represents acceleration characteristics, i=6 represents current characteristics, etc., N + Represents the set of positive integers. The initial design value is given, where λ is the attenuation coefficient and T is the initial design value. dec decay period, It is the average value of the characteristic sequence.

[0151] 3. Regarding the outlier parameter M calculated above... i unThe system is divided into severity levels: {0 (No abnormality)...3 (Severe abnormality)}, and categories exceeding the abnormal range are labeled as {0,1,2,3}. The classification rules are as follows: In the formula, low and upp represent the upper and lower limits of the values ​​of each abnormal feature parameter.

[0152] 4. Each feature has a corresponding label level. There are a total of 6 features, 4 levels, and 24 data nodes. LAB_M i un =[1(P 1 un ), 1(V 2 un ), 1(A 3 un ), 3 (L 4 un ), 0(G 5 un ), 2(I 6 un )] T This indicates that displacement, velocity, and acceleration are slightly abnormal, load is severely abnormal, center of gravity offset is normal, and current is moderately abnormal. Based on this label and the abnormal parameter value, the two abnormal value parameters with the smallest values ​​are repeatedly merged into a new value, that is, the abnormal value parameter value of the new node is equal to the sum of the abnormal value parameter values ​​of the two child nodes, until only one root node remains. For example, the abnormal value of displacement level 1 is (0.5), represented as P1(0.5); the abnormal value of velocity level 1 is (0.8), represented as V1(0.8); acceleration level 1 (1), A1(1); load level 3 (4), represented as L3(4); center of gravity offset level 0 (0), represented as G0(0); current level 2 (3.5), represented as I2(3.5); the 6 characteristic levels are used as the initial nodes, denoted as: G0(0), P1(0.5), V1(0.8), A1(1), I2(3.5), L3(4). The first merge has minimum outliers of 0 and 0.5, merging G0(0) and P1(0.5) to generate a new node N1 with weight = 0 + 0.5 = 0.5. The remaining nodes are N1(0.5), V1(0.8), A1(1), I2(3.5), and L3(4). The second merge has minimum values ​​of N1(0.5) and V1(0.8), generating N2(1.3). The remaining nodes are N2(1.3), A1(1), I2(3.5), and L3(4), continuing until the end. The sum of the weights and path lengths of each node is: G0(0): 0 × 5; P1(0.5): 0.5 × 5; V1(0.8): 0.8 × 4, etc.

[0153] 5. Following the steps above, calculate the sum of the weights of all nodes on the path from the child node to the root node that do not include the cumulative nodes (such as N1 and N2), and label it as W. sum Then calculate the sum of the weights of all nodes, including the cumulative node, denoted as W. max So, what is the risk parameter R for obtaining outlier features? fea =W sum / W max .

[0154] 6. For the application scenario, the safe distance for the computing device when it is close to a wall is: D safe (t)=D react (t)+D per (t)+D buff (t),D react =V(t)*t react V(t) is the response distance, V(t) is the speed of the device, and t react It is response time. It is the execution distance, a eff(t) It is the maximum safe acceleration, k sys It is the system correction factor for the equipment. Where k lod +k fri +k agi =1,k lod This refers to load correction, k fri It is a correction for the coefficient of friction, k agi This is equipment aging correction, η = 0.8. D is the importance coefficient. buff (t)=max(ΔD err ΔD off ) is the error margin, which depends on the maximum value of the error, ΔD err It is the positioning error of the equipment, ΔD off It is the lateral offset error of the equipment.

[0155] 7. Based on the current location of the device, the actual location of the device can be obtained as D. real (t)=D sum -P(t),D sum This is the total distance the equipment travels. The ratio of the actual distance to the safe distance is defined as the safe range risk coefficient R. dis =D real / D safe .

[0156] 8. Regarding the characteristic risk parameter R fea With the range risk coefficient R dis By performing comprehensive risk management, the safety constraints of the equipment, namely R, can be obtained. sys =γfea ·R nor-fea +γ dis ·R nor-dis ,in R fea-min Minimum risk value of the feature, R fea-max Similarly, the maximum risk value of the feature can be obtained as R. nor-dis γ represents the proportion of the two risk parameters, and its value ranges from {(γ...}. fea =1-γ dis )|(γ dis >1 / 2)}.

[0157] 9. Integration Risk R sys Based on key features, a progressive speed control strategy with safety constraints is generated, as shown in Table 3. Where, Sσ=min(R sys Eσ=amx(R) sys ) represents the minimum and maximum values ​​of the overall risk.

[0158] Table 3. Equipment Status and Speed ​​Shift Strategy Corresponding to Constraints (Table 3)

[0159]

[0160]

[0161] 10. Based on the direction of motion and the progressive control strategy implemented in the previous step, obtain the target velocity V for progressive control. goal To adjust the current movement speed of the device, among which In the formula, the speed change time is Ta = (VV) goal ) / a eff(t) The coefficient of variation of the μ velocity curve is generally μ = 8 / Ta.

[0162] 11. By quantifying dynamic safety distance and integrating characteristic risks as supporting conditions for speed changes, a safety control strategy for equipment under different conditions is ultimately achieved, reducing the risk of equipment collisions. Simultaneously, data from each iteration is recorded and saved, allowing for continuous optimization and model updates, thus improving the long-term adaptability and stability of the method.

[0163] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A progressive stage equipment speed control method incorporating safety constraints, characterized in that, Includes the following steps: S1: Real-time acquisition of state values ​​of stage equipment during movement via multi-source sensors to generate a motion feature set. ,in: The current location of the device. At the current speed, For the current acceleration, This represents the current total load. Force sensor readings; center of gravity offset ,in , It is the offset of the load's center of gravity relative to the equipment's center, and , , , Indicates the first The installation coordinates of each sensor; Indicates the current current of the device; S2: Based on the motion feature set Determine dynamic design index values Calculate the outlier parameters for each feature. , This represents the maximum permissible deviation for the corresponding feature. , These are positive integers, corresponding to position, velocity, acceleration, load, center of gravity offset, and current characteristics, respectively. S3: Regarding the outlier parameter The classification is based on the following rules: If Marked as level 0, no abnormality; if Marked as Level 1, mild abnormality; if Marked as level 2, moderately abnormal; if Marked as level 3, severe abnormality, among which , These are the lower and upper threshold values ​​for the corresponding feature outliers, respectively. S4: Using the anomaly level of each feature as the initial node, repeatedly merge the two nodes with the smallest outlier parameters until only one root node remains. The outlier value of the new node is the sum of the outlier values ​​of its two child nodes. Calculate the sum of the weights of the child nodes on the path to the root node that do not include the accumulated nodes. The sum of weights including the accumulated nodes To obtain characteristic anomaly risk parameters ; S5: Dynamic safety distance when the front end of the computing device is close to a wall. ,in , For response time, , For maximum safe acceleration, , , , These are respectively the load correction factor, friction correction factor, and equipment aging correction factor. , This is the importance coefficient. , For positioning error, This refers to the lateral offset error; S6: Based on total equipment operating distance Calculate the actual distance To obtain the range risk coefficient ; S7: Yes , After normalization, the following results were obtained: , , and according to Comprehensive risk of computing systems ,and ; S8: According to Determine a progressive speed control strategy based on equipment status, and follow... Adjust the target speed. For variable speed time, This is the coefficient of variation for the velocity curve.

2. The progressive stage equipment speed control method incorporating safety constraints according to claim 1, characterized in that, The multi-source sensors mentioned in step S1 include: a laser displacement sensor, a photoelectric encoder, a piezoelectric force sensor, a gravity sensor, and a Hall current sensor; the laser displacement sensor collects data... The photoelectric encoder collects data. , The piezoelectric force sensor collects data. To calculate The gravity sensor assists in correction. The Hall current sensor collects... .

3. The progressive stage equipment speed control method incorporating safety constraints according to claim 1, characterized in that, The dynamic design index value mentioned in step S2 ,in For the initial design values ​​of the features, The attenuation coefficient, The value range is 0.1-0.

3. For the decay period, Value range: 5-10 seconds It is the average value of 10-30 historical characteristic sequences.

4. The progressive stage equipment speed control method incorporating safety constraints according to claim 1, characterized in that, In step S3: for the position feature, low(P un )=0.3, upp(P un )=0.6; for the velocity feature, low(V un )=0.4, upp(V un )=0.8; for the Acceleration characteristics, low(A) un )=0.5、upp(A un )=1.0; for Load characteristics, low(L) un =1.0, upp(L un )=3.0; For the centroid offset feature, low(G un )=0.2、upp(G un )=0.5; for Current characteristics, low(I) un =0.8, upp(I un =1.

5.

5. The progressive stage equipment speed control method incorporating safety constraints according to claim 1, characterized in that, The steps described in step S5 Dynamically adjust according to equipment operating conditions: When hour, ;when hour, ;when hour, .

6. The progressive stage equipment speed control method incorporating safety constraints according to claim 1, characterized in that, The normalization formula mentioned in step S7 is: , ,in , , , .

7. The progressive stage equipment speed control method incorporating safety constraints according to claim 1, characterized in that, The progressive speed control strategy described in step S8 is specifically as follows: like , for Minimum value, Value 0.2-0.3: Equipment warning, maintains current speed and indicates abnormal status; like and , for Maximum value A value of 0.7-0.8 indicates slight deceleration of the equipment. Set as ; like and The equipment is slowed down significantly. Set it to 0 and gradually stop; like and In case of emergency stop of equipment, activate the mechanical locking mechanism.

8. The progressive stage equipment speed control method incorporating safety constraints according to claim 1, characterized in that, It also includes step S9: real-time recording of the collected data, calculation results, and control commands from each step; after accumulating 50-100 sets of data, optimization is performed based on the least squares method. , , The parameter values ​​are used to achieve iterative updates of the model.

9. The progressive stage equipment speed control method incorporating safety constraints according to claim 1, characterized in that, When multiple devices are operating simultaneously on the stage, step S5 also needs to include the safe distance between the devices. ,at this time , According to adjacent devices according to calculate, The current speed of adjacent devices; and / or In step S1, the ambient temperature of the equipment operating environment is also collected. And in step S2, Incorporation into feature set The corresponding outlier parameters , For dynamic temperature design values, And in step S3, the temperature characteristic low(T) un =0.2, upp(T) un =0.5.

Images