Dynamic and Agile Adjustment System and Method for Chain Tension of Scraper Conveyor

Abstract

A dynamic and agile adjustment system for scraper conveyor chain tension includes: a pin stress sensor mounted on an auxiliary support for detecting the load applied to the sprocket shaft assembly by the two chains; an impact force sensor located on the upper edge of the transition groove on both sides of the tail section for detecting the force of the scraper colliding with the upper edge due to changes in chain tension; a pressure sensor located in the telescopic cylinder for collecting the actual value of the lower chamber pressure; a displacement sensor located in the telescopic cylinder for collecting the stroke of the telescopic cylinder; a current sensor located at the drive motor for collecting drive motor current data; an information acquisition unit for acquiring the collected data; an analysis and calculation unit for processing and integrating the data acquired by the information acquisition unit to obtain the real-time state and value of the chain tension; and a control output unit for acquiring the chain tension state and value and controlling the telescopic cylinder to perform corresponding extension and retraction. A method for dynamic and agile adjustment of scraper conveyor chain tension is also provided.

Description

Technical Field

[0001] This invention relates to the field of scraper conveyor technology, and in particular to a dynamic and agile adjustment system and method for chain tension of scraper conveyors. Background Technology

[0002] During the full-height mining of ultra-thick coal seams, the significantly increased coal drop at the working face due to factors such as spalling leads to a greater probability of chain drive system tension fluctuations and malfunctions. The chain is a crucial component of the scraper conveyor. During coal transport, the power unit drives the chain in a reciprocating motion on the conveyor. Simultaneously, the chain undergoes elastic deformation under load, resulting in either excessively loose or excessively tight states. If the chain tension is not quickly and effectively adjusted, it cannot be ensured that it remains within a reasonable range, potentially leading to problems such as chain stacking, jamming, and breakage.

[0003] In current applications of chain tension adjustment methods for scraper conveyors, the controller compares the actual chain tension value with the preset value to determine if the chain tension is normal. If an abnormality occurs, the controller activates the telescopic cylinder to adjust the chain tension back to the normal range. However, due to the complex underground working environment, monitoring results are easily affected by uncontrollable factors. For example, when the telescopic cylinder pulls the tailstock, it may encounter rock blockage, leading to increased friction on the tailstock. Additionally, sensor damage may cause data distortion. Furthermore, when using machine vision technology, the camera lens is easily affected by high levels of dust and water mist underground. Ultimately, this results in poor chain tension control. Summary of the Invention

[0004] In order to solve the technical problems existing in the above-mentioned technologies, it is necessary to provide a dynamic and agile adjustment system for the chain tension of a scraper conveyor.

[0005] A dynamic and agile adjustment system for chain tension of a scraper conveyor includes a pin stress sensor, an impact force sensor, a pressure sensor, a displacement sensor, a current sensor, an information acquisition unit, an analysis and calculation unit, a control output unit, and a hydraulic drive unit.

[0006] The hydraulic drive unit includes a telescopic cylinder for adjusting the tension of the scraper conveyor chain, and a hydraulic oil supply assembly for supplying hydraulic oil to the telescopic cylinder.

[0007] The pin stress sensor is installed on the auxiliary bracket at the tail of the scraper conveyor telescopic machine and is used to detect the load applied by the two chains to the sprocket shaft assembly.

[0008] The impact force sensor is installed on the upper edge of the transition groove at the tail of the scraper conveyor and is used to detect the force that causes the scraper to collide with the upper edge due to the change in chain tension.

[0009] The pressure sensor is installed in the telescopic cylinder and is used to collect the actual pressure value of the lower chamber of the telescopic cylinder.

[0010] The displacement sensor is installed in the telescopic cylinder and is used to collect the real-time stroke of the piston rod of the telescopic cylinder.

[0011] The current sensor is installed at the drive motor of the scraper conveyor to collect the current data of the drive motor.

[0012] The information acquisition unit is used to acquire data collected by the pin stress sensor, impact force sensor, and current sensor.

[0013] The analysis and calculation unit is used to organize and integrate the data acquired by the information acquisition unit to obtain the real-time status of the chain tension and the corresponding chain tension value;

[0014] The control output unit is used to acquire the tension state of the chain and the corresponding chain tension value, and control the telescopic cylinder to extend and retract accordingly, so as to dynamically control and adjust the chain tension of the scraper conveyor.

[0015] Preferably, the analysis and calculation unit includes a fuzzy expert system and a probabilistic neural network.

[0016] Preferably, the liquid supply assembly adopts a high-flow-rate four-loop agile liquid supply system.

[0017] It is also necessary to provide a method for dynamic and agile adjustment of chain tension in scraper conveyors.

[0018] A method for dynamically and agilely adjusting the chain tension of a scraper conveyor, utilizing the aforementioned dynamic and agile adjustment system for scraper conveyor chain tension, includes the following steps:

[0019] S1. Using a fuzzy expert system, establish a fuzzy reasoning logic for the scraper conveyor state based on pre-stored historical data and expert experience. Use the fuzzy reasoning logic for the scraper conveyor state as a training sample for a probabilistic neural network to perform fusion judgment, obtain the set of the scraper conveyor chain tension state space, and save it to the probabilistic neural network.

[0020] S2. Obtain the load data applied by the chain to the sprocket shaft assembly, the force data of the scraper impact or upper edge, and the current data of the drive motor;

[0021] S3. Using a probabilistic neural network, the load data of the chain applied to the sprocket shaft group, the force data of the scraper collision or upper edge, and the current data of the drive motor are fused and processed, and then fused with the state space set of the scraper conveyor chain tension to obtain the real-time state of the scraper conveyor chain tension and the chain tension value.

[0022] S4. Based on the obtained chain tension state and chain tension value, the real-time stroke of the telescopic cylinder piston rod is obtained according to the actual value of the telescopic cylinder pressure corresponding to the chain tension value, so as to control the telescopic cylinder to perform corresponding extension and retraction, thereby achieving dynamic adjustment of chain tension.

[0023] Preferably, in step S1, the set of tension state spaces of the scraper conveyor chain is obtained in the following manner:

[0024] (1) Establish fuzzy reasoning logic for the load current state of the drive motors at the head and tail of the scraper conveyor, and use the fuzzy reasoning logic for the load current state of the drive motors at the head and tail of the scraper conveyor as training samples to train and save it to a probabilistic neural network. By performing calculations in the probabilistic neural network, the load current state of the drive motors at the head and tail of the scraper conveyor is fused and judged to obtain the following set of load current state space of the drive motors at the head and tail of the conveyor: {very stable, stable, relatively stable, fluctuating, abnormal}.

[0025] (2) Establish a fuzzy reasoning logic for the load state of the sprocket shaft group pins of the scraper conveyor head and tail, and use the fuzzy reasoning logic for the load state of the sprocket shaft group pins of the scraper conveyor head and tail as training samples to train and save it to a probabilistic neural network. Through calculation in the probabilistic neural network, the load state of the sprocket shaft group pins of the scraper conveyor head and tail is fused and judged to obtain the following set of load state space of the sprocket shaft group pins of the scraper conveyor head and tail: {normal, alarm, fault};

[0026] (3) Establish fuzzy reasoning logic for the force state of the power components (sprocket shaft assembly) at the head and tail of the scraper conveyor, and use the fuzzy reasoning logic for the force state of the power components (sprocket shaft assembly) at the head and tail of the scraper conveyor as training samples to train and save it to a probabilistic neural network. By performing calculations in the probabilistic neural network, the force state of the power components (sprocket shaft assembly) at the head and tail of the scraper conveyor is fused and judged, and the state space set of the power components (sprocket shaft assembly) at the head and tail of the scraper conveyor is obtained as follows: {normal, small force, large force}; similarly, the state space set of the telescopic tail component of the scraper conveyor is obtained as follows: {normal, small force, large force};

[0027] (4) After obtaining the state space set of the forces on the power unit (sprocket shaft group) and the tail unit of the telescopic conveyor, a fuzzy reasoning logic for the tension state of the scraper conveyor chain is established. The fuzzy reasoning logic for the tension state of the scraper conveyor chain is used as a training sample to train and save it to the probabilistic neural network. Through calculation in the probabilistic neural network, the state space set of the tension of the scraper conveyor chain is obtained as follows: {normal, slightly small, slightly large}.

[0028] Compared with the prior art, the scraper conveyor chain tension dynamic and agile adjustment system and method provided by the present invention have the following advantages:

[0029] 1. Multiple sensors, such as pin stress sensors, impact force sensors, pressure sensors, and displacement sensors, are used to detect chain tension-related data through multiple channels. Fuzzy logic-probabilistic neural networks are used to synthesize the data from each sensor to obtain a consistent interpretation of the extension and retraction of the automatic telescopic tail cylinder, resulting in high control accuracy.

[0030] 2. The impact force sensor is placed in the concave surface of the upper edge on both sides of the transition groove at the tail of the machine to detect the force of the scraper hitting the upper edge due to the change in chain tension. A curved square plate is connected in front of the impact force sensor to help the impact force sensor collect the force of the scraper impact.

[0031] 3. Replace the original positioning pin between the sprocket auxiliary bracket and the sprocket shaft assembly with a pair of pin stress sensors. The left and right stress pin sensors work independently to directly measure the load applied to the sprocket shaft assembly by the two chains.

[0032] 4. A high-flow-rate four-circuit pipeline liquid supply unit is used to drive the telescopic cylinder, improving the response speed of the chain tension adjustment system. Attached Figure Description

[0033] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 This is a schematic diagram of the control principle of the present invention.

[0035] Figure 2 This is a schematic diagram of the structure of a probabilistic neural network.

[0036] Figure 3 This is a schematic diagram of the installation and connection of the impact force sensor of the present invention.

[0037] Figure 4 This is a schematic diagram showing the location of the pin stress sensor of the present invention.

[0038] In the diagram: Pin stress sensor 01, impact force sensor 02, pressure sensor 03, displacement sensor 04, current sensor 05, information acquisition unit 06, analysis and calculation unit 07, fuzzy expert system 71, probabilistic neural network 72, control output unit 08, hydraulic drive unit 09, telescopic cylinder 91, fluid supply assembly 92, auxiliary support 10, movable upper edge 20, arc-shaped square plate 30, sprocket shaft assembly 40. Detailed Implementation

[0039] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0040] In the description of this invention, it should be understood that the terms "upper", "middle", "outer", "inner", "lower", etc., which indicate orientation or positional relationship, are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the components or elements referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting this invention.

[0041] Please refer to Figures 1 to 4 In one embodiment, the present invention provides a dynamic and agile adjustment system for the chain tension of a scraper conveyor, including a pin stress sensor 01, an impact force sensor 02, a pressure sensor 03, a displacement sensor 04, a current sensor 05, an information acquisition unit 06, an analysis and calculation unit 07, a control output unit 08, and a hydraulic drive unit 09.

[0042] The hydraulic drive unit 09 includes a telescopic cylinder 91 for adjusting the tension of the scraper conveyor chain, and a hydraulic supply assembly 92 for supplying hydraulic oil to the telescopic cylinder 91.

[0043] The pin stress sensor 01 is installed on the auxiliary bracket 10 at the tail of the scraper conveyor telescopic machine to detect the load applied by the two chains to the sprocket shaft assembly 40.

[0044] The impact force sensor 02 is installed in the upper edge 20 on both sides of the transition groove at the tail of the scraper conveyor, and a curved square plate 30 is connected in front of the impact force sensor 02 to help the impact force sensor 02 collect the force of the scraper impact. The impact force sensor 02 is used to detect the force of the scraper hitting the upper edge 20 due to the change of chain tension.

[0045] Pressure sensor 03 is installed in telescopic cylinder 91 to collect the actual pressure value of the lower chamber of telescopic cylinder 91;

[0046] Displacement sensor 04 is installed in telescopic cylinder 91 to collect the real-time stroke of piston rod of telescopic cylinder 91.

[0047] The current sensor 05 is installed at the drive motor of the scraper conveyor to collect the current data of the drive motor.

[0048] Information acquisition unit 06 is used to acquire data collected by pin stress sensor 01, impact force sensor 02, and current sensor 05;

[0049] The analysis and calculation unit 07 is used to organize and integrate the data acquired by the information acquisition unit 06 to obtain the real-time status of the chain tension and the corresponding chain tension value;

[0050] The control output unit 08 is used to obtain the tension status of the chain and the corresponding chain tension value, and to control the telescopic cylinder 91 to extend and retract accordingly, so as to dynamically control and adjust the chain tension of the scraper conveyor.

[0051] Specifically, the analysis and calculation unit 07 includes a fuzzy expert system 71 and a probabilistic neural network 72. The fuzzy expert system 71 can describe and process incomplete or uncertain information, and, combined with computer technology, achieve automated control. The fuzzy expert system 71 is stored in the knowledge base in the form of "If [condition 1] and [condition 2] and... and [condition n], then [conclusion]", which can be simplified as follows: IF xis A and y is B and...., THEN z is C. The probabilistic neural network 72 is a function approximation method based on Bayesian decision theory. It integrates density function estimation theory with radial basis function neural networks. Under the given conditions, pattern classification is achieved by continuously approximating the optimal Bayesian decision surface. The following explanation uses binary classification as an example:

[0052] First, let the prior probabilities of mode 1 and mode 2 be p1 and p2, respectively. Then we have:

[0053] p1 = p(c1), p2 = p(c2), and p1 + p2 = 1 (Formula 1)

[0054] Given a vector The classification result of this vector based on prior knowledge is expressed as follows:

[0055]

[0056] In formula 2, Indicates when vector The posterior probability of event c1 occurring when it occurs can be derived from Bayesian theory:

[0057]

[0058] To avoid misclassification, the classification rules need further optimization. Assume the classification process θ... i In this context, the input vector x1 is denoted as class c1, and ε is... ij In the process θ i The loss caused by misjudgment in the process, then θi The expected risk is expressed as:

[0059]

[0060] Assuming the risk of perfectly correct classification is 0, the risk of classifying input variable c1 can be expressed as:

[0061]

[0062] After optimizing Formula 5, the Bayesian decision rule is expressed as follows:

[0063]

[0064] In Formula 6, f i Let c1 be the probability density function for category c1.

[0065] like Figure 2 As shown, the probabilistic neural network 72 consists of an input layer, hidden layers, a summation layer, and an output layer. The input-output calculation relationship of the j-th neuron of the i-th class in the hidden layer is represented as follows:

[0066]

[0067] In Formula 7, i = 1, 2, ..., N, where N represents the total number of classes in the training samples, d represents the dimension of the training samples, and x ij This represents the j-th center within the i-th category. The summation layer performs a weighted average of all output values ​​from hidden layer neurons belonging to the same category, calculated as follows:

[0068]

[0069] In Formula 8, v i Let L represent the output of the i-th class; L represents the total number of neurons in the i-th class, which is equal to the total number of classes N in the training samples. The summation layer calculates all neurons sequentially, selects the value with the highest probability, and transmits it to the output layer. The node with the highest probability is used as the output node. The highest probability result is represented as follows:

[0070] y = arg max(v i ) Formula 9.

[0071] Specifically, the liquid supply component 92 uses a high-flow four-circuit agile liquid supply to drive the telescopic cylinder 91, thereby improving the response speed of the chain tension adjustment system.

[0072] In one embodiment, the present invention provides a method for dynamic and agile adjustment of chain tension in a scraper conveyor, the method comprising the following steps:

[0073] S1. Using the fuzzy expert system 71, a fuzzy reasoning logic for the state of the scraper conveyor is established based on pre-stored historical data and expert experience. The fuzzy reasoning logic for the state of the scraper conveyor is used as a training sample for the probabilistic neural network 72 for fusion judgment to obtain the set of the tension state space of the scraper conveyor chain and save it to the probabilistic neural network 72.

[0074] S2. Obtain the load data applied by the chain to the sprocket shaft assembly 40, the force data of the scraper impact or upper edge 20, and the current data of the drive motor.

[0075] S3. Using a probabilistic neural network 72, the load data of the chain applied to the sprocket shaft group 40, the force data of the scraper collision or upper edge 20, and the current data of the drive motor are fused and processed, and then fused with the scraper conveyor chain tension state space set to obtain the real-time state of the scraper conveyor chain tension and the chain tension value.

[0076] S4. Based on the obtained chain tension state and chain tension value, the real-time stroke of the piston rod of the telescopic cylinder 91 is obtained according to the actual pressure value of the telescopic cylinder 91 corresponding to the chain tension value, so as to control the telescopic cylinder 91 to perform corresponding extension and retraction, thereby achieving dynamic adjustment of chain tension.

[0077] In step S1, the set of tension state spaces of the scraper conveyor chain is obtained in the following manner:

[0078] (1) Establish the fuzzy reasoning logic for the load current state of the drive motors at the head and tail of the scraper conveyor, as follows:

[0079] IF M is A_MS, S is A_SN, F is A_FS and C is A_CN, THEN L is A_LM;

[0080] IF M is A_MS, S is A_SN, F is A_FS and C is A_CD, THEN L is A_LB;

[0081] IF M is A_MS, S is A_SN, F is A_FN and C is A_CN, THEN L is A_LV;

[0082] IF M is A_MS, S is A_SN, F is A_FN and C is A_CD, THEN L is A_LM;

[0083] IF M is A_MS, S is A_SN, F is A_FM and C is A_CN, THEN L is A_LM;

[0084] IF M is A_MS,S is A_SN,F is A_FM and C is A_CD,THEN L is A_LB;

[0085] IF M is A_MS,S is A_SM,F is A_FS and C is A_CN,THEN L is A_LM;

[0086] IF M is A_MS,S is A_SM,F is A_FS and C is A_CD,THEN L is A_LB;

[0087] IF M is A_MS,S is A_SM,F is A_FN and C is A_CN,THEN L is A_LM;

[0088] IF M is A_MS,S is A_SM,F is A_FN and C is A_CD,THEN L is A_LB;

[0089] IF M is A_MS,S is A_SM,F is A_FM and C is A_CN,THEN L is A_LB;

[0090] IF M is A_MS,S is A_SM,F is A_FM and C is A_CD,THEN L is A_LB;

[0091] IF M is A_MN,S is A_SN,F is A_FS and C is A_CN,THEN L is A_LS;

[0092] IF M is A_MN,S is A_SN,F is A_FS and C is A_CD,THEN L is A_LM;

[0093] IF M is A_MN,S is A_SN,F is A_FN and C is A_CN,THEN L is A_LV;

[0094] IF M is A_MN,S is A_SN,F is A_FN and C is A_CD,THEN L is A_LS;

[0095] IF M is A_MN,S is A_SN,F is A_FM and C is A_CN,THEN L is A_LM;

[0096] IF M is A_MN,S is A_SN,F is A_FM and C is A_CD,THEN L is A_LB;

[0097] IF M is A_MN,S is A_SM,F is A_FS and C is A_CN,THEN L is A_LM;

[0098] IF M is A_MN,S is A_SM,F is A_FS and C is A_CD,THEN L is A_LB;

[0099] IF M is A_MN,S is A_SM,F is A_FN and C is A_CN,THEN L is A_LM;

[0100] IF M is A_MN,S is A_SM,F is A_FN and C is A_CD,THEN L is A_LB;

[0101] IF M is A_MN,S is A_SM,F is A_FM and C is A_CN,THEN L is A_LB;

[0102] IF M is A_MN,S is A_SM,F is A_FM and C is A_CD,THEN L is A_LB;

[0103] IF M is A_MM,S is A_SN,F is A_FS and C is A_CN,THEN L is A_LM;

[0104] IF M is A_MM,S is A_SN,F is A_FS and C is A_CD,THEN L is A_LB;

[0105] IF M is A_MM,S is A_SN,F is A_FN and C is A_CN,THEN L is A_LS;

[0106] IF M is A_MM,S is A_SN,F is A_FN and C is A_CD,THEN L is A_LM;

[0107] IF M is A_MM,S is A_SN,F is A_FM and C is A_CN,THEN L is A_LM;

[0108] IF M is A_MM,S is A_SN,F is A_FM and C is A_CD,THEN L is A_LB;

[0109] IF M is A_MM,S is A_SM,F is A_FS and C is A_CN,THEN L is A_LB;

[0110] IF M is A_MM,S is A_SM,F is A_FS and C is A_CD,THEN L is A_LD;

[0111] IF M is A_MM,S is A_SM,F is A_FN and C is A_CN,THEN L is A_LM;

[0112] IF M is A_MM,S is A_SM,F is A_FN and C is A_CD,THEN L is A_LB;

[0113] IF M is A_MM,S is A_SM,F is A_FM and C is A_CN,THEN L is A_LB;

[0114] IF M is A_MM,S is A_SM,F is A_FM and C is A_CD,THEN L is A_LD;

[0115] IF M is A_ML,S is A_SN,F is A_FS and C is A_CN,THEN L is A_LD;

[0116] IF M is A_ML,S is A_SN,F is A_FS and C is A_CD,THEN L is A_LD;

[0117] IF M is A_ML,S is A_SN,F is A_FN and C is A_CN,THEN L is A_LD;

[0118] IF M is A_ML,S is A_SN,F is A_FN and C is A_CD,THEN L is A_LD;

[0119] IF M is A_ML,S is A_SN,F is A_FM and C is A_CN,THEN L is A_LD;

[0120] IF M is A_ML,S is A_SN,F is A_FM and C is A_CD,THEN L is A_LD;

[0121] IF M is A_ML,S is A_SM,F is A_FS and C is A_CN,THEN L is A_LD;

[0122] IF M is A_ML,S is A_SM,F is A_FS and C is A_CD,THEN L is A_LD;

[0123] IF M is A_ML,S is A_SM,F is A_FN and C is A_CN,THEN L is A_LD;

[0124] IF M is A_ML,S is A_SM,F is A_FN and C is A_CD,THEN L is A_LD;

[0125] IF M is A_ML,S is A_SM,F is A_FM and C is A_CN,THEN L is A_LD;

[0126] IF M is A_ML,S is A_SM,F is A_FM and C is A_CD,THEN L is A_LD;

[0127] The symbols in the above reasoning logic are respectively expressed as follows: A_MS—small mean, A_MN—normal mean, A_MM—large mean, A_ML—large mean, A_SN—normal standard deviation, A_SM—large standard deviation, A_FS—small peak factor, A_FN—normal peak factor, A_FM—large peak factor, A_CN—normal frequency center, A_CD—abnormal frequency center, A_LV—very stable load current, A_LS—stable load current, A_LM—relatively stable load current, A_LB—fluctuating load current, A_LD—abnormal load current.

[0128] The established fuzzy reasoning logic for the load current state of the drive motors at the head and tail of the scraper conveyor is used as a training sample, trained and saved to the probabilistic neural network 72. Through calculation in the probabilistic neural network 72, the load current state of the drive motors at the head and tail of the scraper conveyor is fused and judged, and the load current state space set of the drive motors at the head and tail of the conveyor is obtained as follows: {very stable, stable, relatively stable, fluctuating, abnormal}.

[0129] (2) Establish the fuzzy reasoning logic for the stress load state of the sprocket shaft group pins at the head and tail of the scraper conveyor, as follows:

[0130] IF M is S_MN,S is S_SN and F is S_FN,THEN P is S_PN

[0131] IF M is S_MN,S is S_SN and F is S_FD,THEN P is S_PN

[0132] IF M is S_MN,S is S_SM and F is S_FN,THEN P is S_PN

[0133] IF M is S_MN,S is S_SM and F is S_FD,THEN P is S_PA

[0134] IF M is S_MA,S is S_SN and F is S_FN,THEN P is S_PA

[0135] IF M is S_MA,S is S_SN and F is S_FD,THEN P is S_PA

[0136] IF M is S_MA, S is S_SM and F is S_FN, THEN P is S_PA

[0137] IF M is S_MA, S is S_SM and F is S_FD, THEN P is S_PF

[0138] IF M is S_MF,S is S_SN and F is S_FN,THEN P is S_PF

[0139] IF M is S_MF,S is S_SN and F is S_FD,THEN P is S_PF

[0140] IF M is S_MF,S is S_SM and F is S_FN,THEN P is S_PF

[0141] IF M is S_MF,S is S_SM and F is S_FD,THEN P is S_PF

[0142] The symbols in the above reasoning logic are expressed as follows: S_MN—normal mean, S_MA—small mean, S_MF—large mean, S_SN—normal standard deviation, S_SM—abnormal standard deviation, S_FN—normal frequency center, S_FD—abnormal frequency center, S_PN—normal load condition, S_PA—small load, S_PF—large load. The established fuzzy reasoning logic for the load condition of the conveyor head and tail sprocket shaft pins is used as training samples to train and store the probabilistic neural network 72.

[0143] The established fuzzy reasoning logic for the load state of the sprocket shaft group pins at the head and tail of the scraper conveyor is used as a training sample, trained and saved to the probabilistic neural network 72. Through calculation in the probabilistic neural network 72, the load state of the sprocket shaft group pins at the head and tail of the scraper conveyor is fused and judged, and the load state space set of the sprocket shaft group pins at the head and tail of the scraper conveyor is obtained as follows: {normal, alarm, fault}.

[0144] (3) Establish fuzzy reasoning logic for the force state of the power components (sprocket shaft assembly) at the head and tail of the scraper conveyor, as follows:

[0145] IF L is D_LV,P is D_PN,THEN D is D_DN

[0146] IF L is D_LV,P is D_PA,THEN D is D_DA

[0147] IF L is D_LV,P is D_PF,THEN D is D_DF

[0148] IF L is D_LS,P is D_PN,THEN D is D_DN

[0149] IF L is D_LS,P is D_PA,THEN D is D_DA

[0150] IF L is D_LS,P is D_PF,THEN D is D_DF

[0151] IF L is D_LM,P is D_PN,THEN D is D_DN

[0152] IF L is D_LM,P is D_PA,THEN D is D_DA

[0153] IF L is D_LM,P is D_PF,THEN D is D_DF

[0154] IF L is D_LB,P is D_PN,THEN D is D_DA

[0155] IF L is D_LB,P is D_PA,THEN D is D_DA

[0156] IF L is D_LB,P is D_PF,THEN D is D_DF

[0157] IF L is D_LD,P is D_PN,THEN D is D_DF

[0158] IF L is D_LD,P is D_PA,THEN D is D_DF

[0159] IF L is D_LD,P is D_PF,THEN D is D_DF

[0160] The symbols in the above reasoning logic are expressed as follows: D_LV—load current is very stable, D_LS—load current is stable, D_LM—load current is relatively stable, D_LB—load current fluctuates, D_LD—load current is abnormal, D_PN—load is normal, D_PA—load is small, D_PF—load is large, D_DN—component is in normal condition, D_DA—component is under less stress, D_DF—component is under more stress.

[0161] The established fuzzy reasoning logic for the force state of the head and tail power components (sprocket shaft assembly) of the scraper conveyor is used as training samples, trained, and saved to the probabilistic neural network 72. Through calculation in the probabilistic neural network 72, the force state of the head and tail power components (sprocket shaft assembly) of the scraper conveyor is fused and judged, resulting in the following state space set for the head and tail power components (sprocket shaft assembly): {normal, low force, high force}. Similarly, after reasonable evaluation by the impact force sensor 02 and the hydraulic cylinder pressure sensor 03, a fusion judgment is made, resulting in the following state space set for the telescopic tail component of the scraper conveyor: {normal, low force, high force}.

[0162] (4) After obtaining the state space set of the forces on the power unit (sprocket shaft assembly) and the tail assembly of the telescopic conveyor, the fuzzy reasoning logic of the chain tension state of the scraper conveyor is established as follows:

[0163] IF D is T_DN,E is T_EN,THEN T is T_TN

[0164] IF D is T_DN,E is T_EA,THEN T is T_TA

[0165] IF D is T_DN,E is T_EF,THEN T is T_TF

[0166] IF D is T_DA,E is T_EN,THEN T is T_TA

[0167] IF D is T_DA,E is T_EA,THEN T is T_TA

[0168] IF D is T_DA,E is T_EF,THEN T is T_TF

[0169] IF D is T_DF,E is T_EN,THEN T is T_TF

[0170] IF D is T_DF,E is T_EA,THEN T is T_TF

[0171] IF D is T_DF,E is T_EF,THEN T is T_TF

[0172] Among them, T_DN—the conveyor power unit (sprocket shaft assembly) is under normal stress, T_DA—the conveyor power unit (sprocket shaft assembly) is understressed, T_DF—the conveyor power unit (sprocket shaft assembly) is understressed, T_EN—the telescopic conveyor tail component is understressed, T_EA—the telescopic conveyor tail component is understressed, T_EF—the telescopic conveyor tail component is understressed, T_TN—the conveyor chain tension is normal, T_TA—the conveyor chain tension is understressed, and T_TF—the conveyor chain tension is overstressed.

[0173] The established fuzzy inference logic of the tension state of the scraper conveyor chain is used as a training sample, trained and saved to the probabilistic neural network 72. Through calculation in the probabilistic neural network 72, the tension state space set of the scraper conveyor chain is obtained as follows: {normal, slightly small, slightly large}.

[0174] The above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the invention. Those skilled in the art will understand that implementing all or part of the above-described embodiments and making equivalent changes in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A dynamic and agile tension adjustment system for scraper conveyor chains, characterized in that: It includes a pin stress sensor, impact force sensor, pressure sensor, displacement sensor, current sensor, information acquisition unit, analysis and calculation unit, control output unit, and hydraulic drive unit; The hydraulic drive unit includes a telescopic cylinder for adjusting the tension of the scraper conveyor chain, and a hydraulic oil supply assembly for supplying hydraulic oil to the telescopic cylinder. The pin stress sensor is installed on the auxiliary bracket at the tail of the scraper conveyor telescopic machine and is used to detect the load applied by the two chains to the sprocket shaft assembly. The impact force sensor is installed on the upper edge of the transition groove at the tail of the scraper conveyor and is used to detect the force that causes the scraper to collide with the upper edge due to the change in chain tension. The pressure sensor is installed in the telescopic cylinder and is used to collect the actual pressure value of the lower chamber of the telescopic cylinder. The displacement sensor is installed in the telescopic cylinder and is used to collect the real-time stroke of the piston rod of the telescopic cylinder. The current sensor is installed at the drive motor of the scraper conveyor to collect the current data of the drive motor. The information acquisition unit is used to acquire data collected by the pin stress sensor, impact force sensor, and current sensor. The analysis and calculation unit is used to organize and integrate the data acquired by the information acquisition unit to obtain the real-time status of the chain tension and the corresponding chain tension value; The control output unit is used to acquire the tension state of the chain and the corresponding chain tension value, and control the telescopic cylinder to extend and retract accordingly, so as to dynamically control and adjust the chain tension of the scraper conveyor.

2. The scraper conveyor chain tension dynamic and agile adjustment system according to claim 1, characterized in that: The analysis and calculation unit includes a fuzzy expert system and a probabilistic neural network.

3. The scraper conveyor chain tension dynamic and agile adjustment system according to claim 2, characterized in that: The liquid supply assembly adopts a high-flow, four-loop, agile liquid supply system.

4. A method for dynamically and quickly adjusting the chain tension of a scraper conveyor, comprising adjusting the chain tension using the dynamic and quick adjustment system for scraper conveyor described in any one of claims 2 to 3, characterized in that: Includes the following steps, S1. Using a fuzzy expert system, establish a fuzzy reasoning logic for the scraper conveyor state based on pre-stored historical data and expert experience. Use the fuzzy reasoning logic for the scraper conveyor state as a training sample for a probabilistic neural network to perform fusion judgment, obtain the set of the scraper conveyor chain tension state space, and save it to the probabilistic neural network. S2. Obtain the load data applied by the chain to the sprocket shaft assembly, the force data of the scraper impact or upper edge, and the current data of the drive motor; S3. Using a probabilistic neural network, the load data of the chain applied to the sprocket shaft group, the force data of the scraper collision or upper edge, and the current data of the drive motor are fused and processed, and then fused with the state space set of the scraper conveyor chain tension to obtain the real-time state of the scraper conveyor chain tension and the chain tension value. S4. Based on the obtained chain tension state and chain tension value, the real-time stroke of the telescopic cylinder piston rod is obtained according to the actual value of the telescopic cylinder pressure corresponding to the chain tension value, so as to control the telescopic cylinder to perform corresponding extension and retraction, thereby achieving dynamic adjustment of chain tension.

5. The method for dynamic and agile adjustment of chain tension in a scraper conveyor according to claim 4, characterized in that: In step S1, the set of tension state spaces of the scraper conveyor chain is obtained in the following manner: (1) Establish fuzzy reasoning logic for the load current state of the drive motors at the head and tail of the scraper conveyor, and use the fuzzy reasoning logic for the load current state of the drive motors at the head and tail of the scraper conveyor as training samples to train and save it to a probabilistic neural network. By performing calculations in the probabilistic neural network, the load current state of the drive motors at the head and tail of the scraper conveyor is fused and judged to obtain the following set of load current state space of the drive motors at the head and tail of the conveyor: {very stable, stable, relatively stable, fluctuating, abnormal}. (2) Establish fuzzy reasoning logic for the load state of the sprocket shaft group pins of the scraper conveyor head and tail, and use the fuzzy reasoning logic for the load state of the sprocket shaft group pins of the scraper conveyor head and tail as training samples to train and save to the probabilistic neural network. Through calculation in the probabilistic neural network, the load state of the sprocket shaft group pins of the scraper conveyor head and tail is fused and judged to obtain the load state space set of the sprocket shaft group pins of the scraper conveyor head and tail as follows: {normal, alarm, fault}; (3) Establish fuzzy reasoning logic for the force state of the head and tail power components of the scraper conveyor, and use the fuzzy reasoning logic for the force state of the head and tail power components of the scraper conveyor as training samples to train and save it to the probabilistic neural network. By performing calculations in the probabilistic neural network, the force state of the head and tail power components of the scraper conveyor is fused and judged, and the state space set of the head and tail power components of the scraper conveyor is obtained as follows: {normal, small force, large force}; similarly, the state space set of the telescopic tail component of the scraper conveyor is obtained as follows: {normal, small force, large force}. Among them, the power components of the scraper conveyor head and tail are the sprocket shaft assemblies of the head and tail; (4) After obtaining the state space set of the forces on the power unit of the conveyor and the tail unit of the telescopic conveyor, a fuzzy reasoning logic of the tension state of the scraper conveyor chain is established. The fuzzy reasoning logic of the tension state of the scraper conveyor chain is used as a training sample to train and save it to the probabilistic neural network. Through calculation in the probabilistic neural network, the state space set of the tension of the scraper conveyor chain is obtained as follows: {normal, slightly small, slightly large}.

Images