Novel hopper with automatic material cleaning function

Abstract

The novel hopper with the automatic material cleaning function comprises a hopper body, a flow sensor and a motor, a supporting frame is fixedly arranged at the bottom of the hopper body, a stirring rod is arranged in the middle of the inner bottom of the hopper body, material stirring pieces are arranged on the stirring rod, and the motor is arranged at the lower end of the outer bottom of the hopper body; a feeding port is formed in the top of the hopper, a discharging port is formed in the side face of the hopper, the discharging port is connected with a discharging pipe, a pipe cover is arranged at the end of the discharging pipe, a flow sensor is arranged at the top of the discharging pipe, and the flow sensor is used for monitoring the material flow in the discharging pipe in real time and detecting whether the blocking phenomenon occurs or not. The material flow can be detected in real time in the material dumping or processing process, and corresponding automatic adjusting measures are taken to achieve the automatic cleaning function so as to prevent blockage, accumulation or mixing of materials.

Description

Technical Field

[0001] The invention relates to the technical field of automation equipment or industrial equipment, and more specifically to a new type of hopper with an automatic material clearing function. Background Art

[0002] Traditional hoppers have complex structures and single functions, which are prone to material accumulation or blockage and cannot be judged. Especially when dealing with materials with high viscosity, high humidity or uneven particles, cleaning work usually requires manual intervention, which is time-consuming and unhygienic. Especially in industries with strict hygiene requirements such as food, pharmaceuticals, and chemicals, traditional cleaning methods are difficult to ensure thorough cleaning and may cause cross contamination or safety hazards.

[0003] The existing hoppers have low unloading efficiency, incomplete unloading, and difficulty in determining whether material accumulation or blockage has occurred during the unloading process. In addition, the traditional hopper cleaning process usually relies on manual intervention, which is not only time-consuming but also difficult to ensure cleanliness, thus affecting production efficiency and the smoothness of material transportation. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a new type of hopper with automatic cleaning function, which can not only detect the blockage of the hopper in real time, but also automatically take adjustment measures to effectively realize automatic cleaning of residual materials on the inner wall, thereby improving unloading efficiency, reducing material accumulation and ensuring the cleanliness of the equipment.

[0005] The technical solution adopted by the present invention to solve its technical problems is: constructing a new type of hopper with automatic material clearing function, including a hopper, a flow sensor and a motor, a support frame is fixedly arranged at the bottom of the hopper, a stirring rod is arranged in the middle of the inner bottom of the hopper, a stirring piece is arranged on the stirring rod, and a motor is arranged at the lower end part of the outer bottom of the hopper; a feed inlet is arranged at the top of the hopper, a discharge port is arranged on the side of the hopper, the discharge port is connected with a discharge pipe, a pipe cover is arranged at the end of the discharge pipe, and a flow sensor is arranged at the top of the discharge pipe, and the flow sensor is used to monitor the material flow in the discharge pipe in real time and detect whether blockage occurs.

[0006] According to the above scheme, a bearing seat is arranged in the middle of the bottom of the hopper, the motor is fixedly arranged on the rotating shaft at the bottom end of the hopper, the bottom end of the rotating shaft is transmission-connected with the output end of the motor, the rotating shaft is pivotally connected with the bearing seat and passes through the bearing seat, the sleeve is fixedly arranged on the top of the rotating shaft, a plurality of fixed slots are opened inside the sleeve along the circumferential direction, and an inclined plate is arranged at the bottom of the hopper.

[0007] According to the above scheme, the stirring rod is connected to the shaft sleeve in the fixed groove through the fixed clamping block, so that the stirring rod is connected to the rotating shaft, and a plurality of stirring pieces are evenly arranged on the side of the stirring rod;

[0008] A threaded hole groove is provided on the end of the stirring piece away from the stirring rod, and a scraper piece is clamped on the end of the stirring piece away from the stirring rod. A groove is provided at the end of the scraper piece. The end of the stirring piece away from the stirring rod is clamped in the groove at the end of the scraper piece. A nut is provided on the top of the scraper piece. The nut is connected to the fastening bolt and cooperates with the threaded hole groove opened on the top of the stirring piece.

[0009] According to the above scheme, the flow sensor monitors the flow at the discharge pipe in real time by detecting the vibration changes caused by the material flow; when the material flow is abnormal, the flow sensor compares the real-time material flow with the reference flow when the hopper is working normally, so as to determine whether blockage occurs, and then take corresponding measures to ensure the normal operation of the hopper.

[0010] According to the above scheme, the flow sensor is equipped with a control system, and the control system adopts a fuzzy adaptive PID control algorithm. The flow sensor is used to dynamically adjust the speed of the motor and the material input amount of the feed port according to the deviation between the real-time detected flow and the preset reference flow.

[0011] According to the above scheme, the fuzzy adaptive PID control algorithm adopted by the control system includes the following steps:

[0012] S1. Obtain the material flow signal Q(t) detected by the flow sensor in real time, calculate and preset the reference flow Q ref The deviation of e(t)=Q ref -Q(t), taking the flow deviation e(t) and the deviation change rate de / dt as input variables, the correction values ​​of the fuzzy adaptive PID control algorithm parameters (Kp, Ki, Kd) are generated through fuzzy inference rules;

[0013] S2, according to the modified fuzzy adaptive PID control algorithm parameter proportional coefficient K p0 , integral coefficient K i0 , differential coefficient K d0 ;

[0014] Calculate the motor speed adjustment amount and convert it into the motor speed adjustment instruction and the feed valve opening instruction to achieve speed adaptive adjustment. The specific calculation formula is as follows:

[0015]

[0016] S3. When it is detected that the flow rate is continuously lower than the threshold, the high-speed rotation mode of the scraper is triggered, and the feed inlet valve is linked to reduce the material input.

[0017] According to the above solution, the method for generating the correction amounts of the fuzzy adaptive PID control algorithm parameters (Kp, Ki, Kd) in step S1 includes the following steps:

[0018] S101. Define fuzzy sets and membership functions

[0019] Set the physical range of the flow deviation e(t) according to the range of the flow sensor, and set the physical range of the deviation change rate de / dt according to the dynamic characteristics of the system. Then divide the deviation e(t) and the deviation change rate de / dt into 5 fuzzy linguistic variables;

[0020] S102. Design the fuzzy rule base

[0021] The adjustment of the fuzzy adaptive PID control algorithm parameters needs to be based on the dynamic characteristics of the system. The typical rule logic is as follows:

[0022] Proportional gain Kp: When the deviation becomes large, greatly increase Kp to quickly reduce the deviation of the chip. When the deviation is small, reduce Kp to avoid oscillation;

[0023] Integral gain Ki: Increase Ki to eliminate the static error, and reduce Ki to prevent integral saturation,

[0024] Differential gain Kd: When the deviation change rate is large, increase Kd to suppress overshoot, and reduce Kd to suppress noise sensitivity;

[0025] S103. Fuzzy inference and defuzzification

[0026] First, calculate the membership degrees of the current e(t) and de / dt to each fuzzy set, and then perform rule activation. Rule activation is to match the fuzzy input with the conditions in the rule base, calculate the triggering strength of each rule, calculate the minimum operator and product operator through the membership degrees of the input, and then perform rule logic matching. Determine the corresponding fuzzy set through the membership degrees of the input, and then trigger all the rules containing this fuzzy set; furthermore, trim the membership function of the output variable according to the rule strength, and at the same time aggregate the outputs of all rules;

[0027] After completing the fuzzy inference, perform defuzzification. Defuzzification converts the fuzzy output into an exact control quantity, and usually uses the centroid method to convert the fuzzy output into an exact value;

[0028] S104. Dynamically update the parameters of the fuzzy adaptive PID control algorithm

[0029] The parameter correction formula is as follows:

[0030] K p = K p0 + ΔK p × α p

[0031] K i = K i0 + ΔK i × α i

[0032] K d = K d0 + ΔK d × α d

[0033] K p0 、K i0 、K d0 are initial parameters, and α p 、α i 、α d are correction scaling factors used to match the actual parameter range.

[0034] According to the above solution, the control system integrates an LSTM prediction model, trains the LSTM neural network model with historical flow data, predicts the flow trend in the next 5 - 10 seconds, and adjusts the motor speed in advance to prevent blockage.

[0035] According to the above solution, the input parameters of the LSTM prediction model include the historical flow data sequence {Q(1), Q(2),..., Q(n)}, ambient temperature T, and ambient humidity H, and the output is the flow prediction value Q in the next 5 - 10 seconds. pred , when Q pred < 0.8Q ref is satisfied, the system triggers the anti - block operation, and the specific process includes the following steps:

[0036] S1. Construction of the LSTM network model

[0037] The input parameters are the historical flow data sequence {Q(1), Q(2),..., Q(n)}, ambient temperature T, and ambient humidity. Then, the LSTM network architecture is constructed. The input layer includes three neurons for flow, temperature, and humidity. The hidden layer consists of two stacked LSTMs, with thirty - two neurons in each layer. The activation function uses the ReLU function to accelerate convergence. The output layer is a single neuron, which is the flow prediction value in the next 5 - 10 seconds, and the activation function is a linear activation.

[0038] S2. Model training

[0039] (1) Dataset construction

[0040] Select the flow data of the past 60 seconds with a sampling interval of 1 second to capture the periodic characteristics of the material flow and obtain the historical flow sequence {Q(1), Q(2),..., Q(n)}; collect the ambient temperature (T) and humidity (H) in real time through the temperature and humidity sensors to correct the influence of material viscosity on fluidity; obtain a large number of data sets and divide them into training sets and validation sets according to an appropriate ratio, including data of normal working conditions and blockage events.

[0041] (2) Training parameters

[0042] When training the model, the training cycle (epoch) is initially selected as 300 times to fully learn the time series dependence relationship. To balance memory occupancy and gradient stability, the batch size is selected as 16, and the Adam optimizer is used with an initial learning rate of 0.001 to automatically adjust the parameter update step size. The loss function is the mean squared error, and the formula is:

[0043]

[0044] S3. Prediction trigger conditions

[0045] Blockage determination condition: When the predicted flow rate Q pred <0.8Q ref , it is determined that there is a potential blockage risk; the blockage probability formula:

[0046]

[0047] Among them, Q ref is the set safety reference flow rate value, and Q pred is the predicted flow rate value;

[0048] That is, when P block > 20%, it is determined that there is a potential blockage risk.

[0049] According to the above scheme, when it is determined that there is a potential blockage risk, an anti-blockage response strategy is adopted;

[0050] The anti-blockage response strategy: After it is determined as a blockage risk, the motor performs a speed regulation operation, and the speed is increased to 120% of the rated value through the frequency converter to enhance the material fluidity; the scraping plate starts a high-frequency reciprocating motion to peel off the material attached to the inner wall; at this time, the fuzzy adaptive PID control algorithm continues to adjust, and the parameters of the fuzzy adaptive PID control algorithm are dynamically adjusted according to the prediction deviation e(t) and the deviation change rate de / dt. The proportional coefficient continues to increase as the deviation increases, the integral coefficient gradually decays during continuous deviation to prevent integral saturation, and the differential coefficient is adaptively adjusted according to the deviation change rate to complete the control closed-loop.

[0051] Implementing the new hopper with an automatic material cleaning function of the present invention has the following beneficial effects:

[0052] 1. The present invention not only achieves the effects of automatic material clearing, saving material costs, and preventing hopper blockage, but also adds a flow sensor to monitor the feeding flow rate in real time and an intelligent feedback control system. Through the stirring rod and stirring blades, the effects of smoothly feeding the material by stirring the material during feeding to prevent hopper blockage are achieved; through the scraping blade, the effects of quickly clearing the material adhering to the inner wall of the hopper are achieved, reducing the influence of material adhesion on the side wall of the hopper on feeding and material waste, and achieving the effect of automatic material clearing; through the nut and fastening bolt, the effect of being able to regularly remove the scraping blade to clean the residual material adhering to the scraping blade is achieved; through the flow monitoring of the material sensor and the intelligent control feedback system, the feeding flow rate of the material is detected in real time and automatic adjustment measures are taken to achieve the effects of controlling the feeding amount and the motor speed;

[0053] 2. The present invention can detect the material flow rate in real time during the material dumping or processing process and make corresponding automatic adjustment measures to realize the automatic cleaning function, so as to prevent material blockage, accumulation or mixing, aiming to improve production efficiency, ensure the smoothness of material transportation, and meet the requirements of modern industry for hygiene, automation and high-efficiency production;

[0054] 3. By introducing a solid material sensor, mechanical scraping, etc., the automatic material clearing system of the present invention can detect the situation of unsmooth material flow in real time and start the cleaning function to ensure the hopper is unblocked and improve production efficiency. The automatic material clearing system monitors the flow state of the material at the hopper discharge port through an intelligent sensor, judges in time whether cleaning is needed, reduces manual intervention, reduces production downtime, and improves the automation level. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The present invention will be further described below in conjunction with the drawings and embodiments. In the drawings:

[0056] Figure 1 is a schematic structural diagram of the first embodiment of the novel hopper with an automatic material clearing function of the present invention;

[0057] Figure 2 is a schematic structural diagram of the second embodiment of the novel hopper with an automatic material clearing function of the present invention;

[0058] Figure 3 is an enlarged schematic diagram of the cover of the novel hopper with an automatic material clearing function of the present invention;

[0059] Figure 4 is a schematic structural diagram of the third embodiment of the novel hopper with an automatic material clearing function of the present invention;

[0060] Figure 5 is an enlarged schematic diagram of the material clearing mechanism of the novel hopper with an automatic material clearing function of the present invention;

[0061] Figure 6 Structural schematic diagram of the fourth embodiment of the novel hopper with an automatic material cleaning function according to the present invention;

[0062] Figure 7 Flow chart of the control system of the novel hopper with an automatic material cleaning function according to the present invention;

[0063] In the figure: 1. Hopper, 2. Flow sensor, 3. Fixed rod, 4. Fixed groove, 5. Pipe cover, 6. Discharge pipe, 7. Support frame, 8. Motor, 9. Rotating shaft, 10. Bearing seat, 11. Bush, 12. Fixed clamping block, 13. Fixed clamping groove, 14. Stirring rod, 15. Stirring blade, 16. Scraping blade, 17. Tightening bolt, 18. Threaded hole groove, 19. Nut, 20. Inclined plate, 21. Feed inlet. Detailed implementation manners

[0064] For a clearer understanding of the technical features, purposes and effects of the present invention, the detailed implementation manners of the present invention will now be described in detail with reference to the accompanying drawings.

[0065] As Figures 1-7 shown, the novel hopper with an automatic material cleaning function of the present invention includes a hopper 1, a flow sensor 2 and a motor 8. A support frame 7 is fixedly arranged at the bottom of the hopper 1. A stirring rod is arranged in the middle of the inner bottom of the hopper 1, and stirring blades 15 are arranged on the stirring rod. A motor 8 is arranged at the lower end of the outer bottom of the hopper 1. A feed inlet 21 is arranged at the top of the hopper 1. A discharge port is arranged on the side of the hopper 1. The discharge port is connected to a discharge pipe 6. A pipe cover 5 is arranged at the end of the discharge pipe 6. A flow sensor 2 is arranged at the top of the discharge pipe 6.

[0066] A bearing seat 10 is arranged in the middle of the bottom of the hopper 1. The motor 8 is fixedly arranged on the rotating shaft 9 at the bottom end of the hopper 1. The bottom end of the rotating shaft 9 is in transmission connection with the output end of the motor 8. The rotating shaft 9 is pivotally connected to the bearing seat 10 and penetrates through the bearing seat 10. A bush 11 is fixedly arranged at the top of the rotating shaft 9. A plurality of fixed clamping grooves 13 are arranged along the circumferential direction inside the bush 11. An inclined plate 20 is arranged at the inner bottom of the hopper 1.

[0067] Threaded hole grooves 18 are arranged at the end of the stirring blade 15 far from the stirring rod 14. A scraping blade 16 is clamped at the end of the stirring blade 15 far from the stirring rod 14. A fixed groove 4 is arranged at the end of the scraping blade 16. The end of the stirring blade 15 far from the stirring rod 14 is clamped in the fixed groove 4 at the end of the scraping blade 16. A nut 19 is arranged at the top of the scraping blade 16. The nut 19 is connected to a tightening bolt 17 and is in fit connection with the threaded hole groove 18 opened at the top of the stirring blade 15. When it is necessary to clean the residual materials on the inner wall of the hopper 1, the scraping blade 16 is installed on the stirring blade 15 through the fixed groove 4 at the end of the scraping blade 16 and the nut 19.

[0068] The flow sensor 2 is a solid material flow sensor 2 based on mass flow detection. The flow sensor 2 utilizes the vibration principle and is installed at the outlet pipe 6 of the hopper 1. It monitors the flow rate at the outlet pipe 6 in real time by detecting the vibration changes caused by the material flow. When the material flow rate is abnormal, the flow sensor 2 compares the real-time material flow rate with the reference flow rate during the normal operation of the hopper 1 to determine whether there is a blockage. Then corresponding measures are taken to ensure the normal operation of the hopper 1.

[0069] The control system of the flow sensor 2 incorporates a fuzzy PID control algorithm to dynamically adjust the rotational speed of the motor 8 and the material input amount at the feed inlet 21 according to the deviation between the real-time flow rate detected by the flow sensor 2 and the preset reference flow rate. The control system adopts a fuzzy adaptive PID control algorithm, which includes the following steps:

[0070] S1. Obtain the material flow rate signal Q(t) detected by the flow sensor 2 in real time, and calculate the deviation e(t) = Q ref -Q(t) with the preset reference flow rate Q ref . Take the flow deviation e(t) and the deviation change rate de / dt as input variables, and generate the correction amounts of the PID parameters (Kp, Ki, Kd) through fuzzy inference rules. The specific steps are as follows:

[0071] S101. Define the fuzzy sets and membership functions

[0072] According to the sensor range, set the physical range of the flow deviation e(t), for example, the range is [-100, 100] (unit: L / min). According to the system dynamic characteristics, set the physical range of the deviation change rate de / dt to [-50, 50] (unit: L / min). Then divide the deviation e(t) and the deviation change rate de / dt into 5 fuzzy linguistic variables. The membership function is triangular, as shown in the specific division tables 1 and 2:

[0073] Table 1 Division of the membership function of the deviation

[0074]

[0075] Table 2 Division of the membership function of the deviation change rate

[0076]

[0077] S102. Design the fuzzy rule base

[0078] The adjustment of the PID parameters needs to be based on the system dynamic characteristics. The typical rule logic is as follows:

[0079] Proportional gain Kp: When the deviation becomes larger, significantly increase Kp to quickly reduce the deviation. When the deviation is small, reduce Kp to avoid oscillation.

[0080] Integral gain Ki: Increasing Ki is used to eliminate static errors, while reducing Ki can cause integral saturation.

[0081] Differential gain Kd: When the deviation change rate is large, increase Kd to suppress overshoot, and reduce Kd to suppress noise sensitivity.

[0082] In (1), the deviation and the deviation change rate have been divided into membership functions, and 25 controls can be obtained. Specific regulation example: When the deviation e(t) = PB and the deviation change rate de / dt = PB, the deviation is extremely large and continues to increase. At this time, it is necessary to make the maximum proportional gain respond quickly (ΔKp = PB), turn off the integral to avoid overshoot (ΔKi = NB), and enhance the differential prediction trend (ΔKd = PB).

[0083] S103, Fuzzy reasoning and defuzzification

[0084] First, the membership of the current e(t) and de / dt to each fuzzy set is calculated, and then the rule activation is performed. Rule activation is to match the fuzzy input with the conditions in the rule base, calculate the trigger strength of each rule, calculate the minimum operator and product operator through the input membership, and then perform rule logic matching. The corresponding fuzzy set is determined by the input membership, and then all rules containing the fuzzy set are triggered. Then, the membership function of the output variable is tailored according to the rule strength, and the output of all rules is aggregated at the same time.

[0085] After completing fuzzy reasoning, defuzzification is performed to convert the fuzzy output into an accurate control quantity. The center of gravity method is usually used to convert the fuzzy output into an accurate value. For example, if the center of gravity of the synthesized ΔKp membership function in the corresponding interval is 50, the output accurate value ΔKp=50.

[0086] S104, PID parameters dynamic update

[0087] The parameter correction formula is as follows:

[0088] K p =K p0 +ΔK p ×α p

[0089] K i =K i0 +ΔK i ×α i

[0090] K d =K d0 +ΔK d ×α d

[0091] K p0 , Ki0 , K d0 is the initial parameter

[0092] αii p , α i , α d are the correction scaling factors used to match the actual parameter range

[0093] S2. According to the corrected PID parameter proportional coefficient K p0 , integral coefficient K i0 , differential coefficient K d0

[0094] Calculate the motor speed adjustment amount and convert it into the speed adjustment instruction of motor 8 and the opening instruction of the feed valve to achieve speed adaptive adjustment. The specific calculation formula is as follows:

[0095]

[0096] S3. When it is detected that the flow rate is continuously lower than the threshold, trigger the high-speed rotation mode of the scraping blade 16, and at the same time link the valve at the feed port 21 to reduce the material input amount.

[0097] The control system integrates an LSTM prediction model, trains the LSTM neural network model through historical flow rate data, predicts the flow rate trend in the next 5 - 10 seconds, and adjusts the motor speed in advance to prevent blockage. The input parameters of the LSTM model include the historical flow rate data sequence {Q(1), Q(2),..., Q(n)}, ambient temperature T, and ambient humidity H. The output is the flow rate prediction value Q of the next 5 - 10 seconds pred , when Q pred <0.8Q ref is satisfied, the system triggers the anti-blocking operation. The specific process is as follows:

[0098] S1. Construction of the LSTM network model

[0099] The input parameters are the historical flow rate data sequence {Q(1), Q(2),..., Q(n)}, ambient temperature T, and ambient humidity, and then construct the LSTM network architecture. The input layer consists of 3 neurons, corresponding to the three features of flow rate, temperature, and humidity respectively. The hidden layer consists of two stacked LSTMs, with 32 neurons in each layer, and the activation function uses the ReLU function to accelerate convergence. The output layer is 1 neuron, that is, the flow rate prediction value in the next 5 - 10 seconds, and the activation function is a linear activation.

[0100] S2. Model training

[0101] S201. Construction of the data set

[0102] Select the flow rate data of the past 60 seconds (sampling interval of 1 second) to capture the periodic characteristics of the material flow. Through this method, the historical flow rate sequence {Q(1), Q(2),..., Q(n)} is obtained. The ambient temperature (T) and humidity (H) are collected in real time by temperature and humidity sensors to correct the influence of material viscosity on fluidity. A large amount of data sets are obtained and divided into training sets and validation sets according to an appropriate ratio, including data under normal working conditions and blockage events.

[0103] S202. Training parameters

[0104] When training the model, the training epoch is initially selected as 300 times to fully learn the time series dependence relationship. To balance memory occupancy and gradient stability, the batch size is selected as 16. And the Adam optimizer is selected, with an initial learning rate of 0.001, automatically adjusting the parameter update step size. The loss function is the mean squared error, and the formula is:

[0105]

[0106] S203. Prediction trigger conditions

[0107] Blockage determination condition: When the predicted flow rate Q pred <0.8Q ref , it is determined that there is a potential blockage risk. The blockage probability formula:

[0108]

[0109] Where Q ref is the set safety reference flow rate value, and Q pred is the predicted flow rate value. That is, when P block > 20%, it is determined that there is a potential blockage risk.

[0110] Anti-blocking response strategy: After it is determined as a blockage risk, the motor performs a speed regulation operation, and the speed is increased to 120% of the rated value through the frequency converter to enhance the material fluidity. The scraping plate starts a high-frequency reciprocating motion to peel off the material attached to the inner wall. At this time, the fuzzy PID algorithm continues to adjust, dynamically adjusting the PID parameters according to the prediction deviation e(t) and the deviation change rate de / dt. The proportional coefficient continues to increase as the deviation increases, the integral coefficient gradually decays during continuous deviation to prevent integral saturation, and the differential coefficient is adaptively adjusted according to the deviation change rate to complete the control closed-loop.

[0111] The working principle of the present invention:

[0112] When it is necessary to use the hopper 1 for discharging materials, first start the motor 8 to drive the sleeve 11 to rotate through the rotating shaft 9. There is a fixed card slot 13 inside the sleeve 11, and a fixed card block 12 is provided at the bottom of the stirring rod 14. The fixed card block 12 and the fixed card slot 13 are clamped and connected together to drive the stirring rod 14, the stirring blade 15 and the scraping blade 16 to rotate together. Then pour the materials into the hopper 1 through the material inlet 21 to mix and discharge the materials. During the process of mixing and discharging materials, the opening of the pipe cover 5 is closed through the fixed rod 3 and the fixed groove 4 to prevent the materials from leaking when being processed in the hopper 1. After the discharging is completed, open the pipe cover 5 at the discharge pipe 6, and the mixed materials in the hopper 1 will finally pour out from the discharge pipe 6 on the lower right side to avoid incomplete mixing of the materials. When all the raw materials in the hopper 1 are unloaded, close the opening of the pipe cover 5 through the fixed rod 3 and the fixed groove 4 to prevent the materials from leaking when being mixed and processed in the hopper 1.

[0113] During the process of discharging materials from hopper 1, the flow sensor 2 located at the top of the discharge pipe 6 of hopper 1 is used to monitor the material flow in real time, and it has a material flow monitoring and prediction system. The flow sensor 2 is used to monitor the material flow in the discharge pipe 6 in real time and detect whether there is a blockage. The prediction mechanism uses machine learning algorithms to analyze historical data and real-time sensor data to predict the risks of abnormal flow and material accumulation. The flow sensor 2 used is a mass flow sensor based on the vibration principle to measure the material flow rate. The flow sensor 2 measures the tiny vibrations caused by the material flow at the discharge pipe 6, detects the material flow through the change in vibration frequency, and accurately detects the material flow at the discharge pipe 6. When the material flows through the sensor 2, its vibration characteristics will change. The flow sensor 2 calculates the material flow rate in real time based on these changes, and compares the real-time material flow with the material flow during the normal operation of hopper 1 to judge the blockage situation at the discharge pipe 6 of hopper 1 in real time. The flow sensor 2 collects data such as flow, humidity, and temperature in real time, sends them to the central control system for processing and analysis, and makes corresponding automatic adjustment measures. When it is detected that the material flow at the discharge pipe 6 exceeds the preset normal range, the flow sensor 2 will immediately send an alarm signal, and the system starts the automatic material cleaning mechanism. At this time, the system avoids the blockage of the discharge pipe 6 caused by excessive material accumulation by reducing the material input and increasing the rotation speed of the stirring rod 14 to ensure that the material can pass through the discharge pipe 6 smoothly. When the flow returns to the normal range, the flow sensor 2 sends a normal signal, and the system adjusts the material feeding amount in a timely manner and appropriately reduces the motor 8 speed to maintain the normal operation of hopper 1. In the prediction mechanism system, machine learning algorithms are used to establish a material flow model, predict future flow changes based on the real-time collected data, material type, and environmental conditions, and make adjustment measures before a possible blockage occurs to improve the discharging efficiency. Through the flow monitoring and feedback function and the prediction mechanism, it can effectively judge whether there is a blockage in hopper 1, and automatically take adjustment measures when there is a blockage risk, effectively avoiding interruptions or equipment damage during the production process. At the same time, the prediction mechanism of the system can make anti-blockage measures before a possible blockage occurs in hopper 1 by using machine learning algorithms to analyze historical data and real-time sensor data, saving the discharging time, improving the efficiency of material transportation, and ensuring the continuity and stability of the discharging work. The precise detection and automatic feedback function of the flow sensor 2 and the prediction mechanism make the working state of hopper 1 more intelligent, greatly reducing the risks of manual intervention and misoperation, thereby improving the overall operation safety and work efficiency.

[0114] The intelligent feedback control system of the new hopper 1 with an automatic material cleaning function in the present invention has the functions of automatically adjusting the flow rate, stirring speed, and intelligently adjusting the material processing mode. Using the fuzzy adaptive PID control algorithm, it can achieve the automatic adjustment of the motor 8 and the stirring speed. When the flow sensor 2 detects that the material flow rate is abnormal or normal, the intelligent feedback system will automatically adjust the rotation speed of the motor 8, slowing down or accelerating the stirring speed to ensure that the material can smoothly pass through the discharge pipe 6. In addition, according to the characteristics of the material, the flow detection data, and the production requirements, the intelligent feedback system can automatically switch the operation mode. For example, it switches to the energy-saving mode when the material is less, and switches to the high-load mode when it detects that the material has poor fluidity.

[0115] During the feeding process, when too much material adheres to the inner wall of the hopper 1 due to long-term friction or high humidity of the material. The rotation of the stirring rod 14 drives the scraping blade 16 to rotate on the inner wall of the hopper 1, thus effectively scraping off the residual material adhering to the inner wall. The residual material falls off along the inner wall and flows towards the discharge pipe 6 together with most of the material. In this way, the adhesion of the material on the side wall of the hopper 1 can be reduced, preventing the accumulation and blockage of the material, and achieving the function of automatic material cleaning. During the stirring and scraping process, the function of the scraping blade 16 is to remove the material that is difficult to naturally fall off or flow from the inner wall of the hopper 1, so as to ensure that the residual material and most of the material can flow smoothly towards the discharge pipe 6 together, avoiding problems such as insufficient flow rate or uneven feeding caused by material adhesion. This not only achieves automatic material cleaning but also greatly improves the feeding efficiency and reduces the decline in the production efficiency of the hopper 1 caused by material accumulation.

[0116] During the material cleaning process, there is an efficient automatic cleaning system. When too much material adheres to the inner wall of the hopper 1, the feeding flow rate of the material will decrease accordingly. The flow sensor 2 can detect and send out a signal. After the signal is sent out, the automatic cleaning system will adjust the cleaning intensity according to the type and accumulation degree of the material. For example, the system will increase or decrease the rotation speed and pressure of the scraping blade 16 according to the viscosity of the material. And by using historical flow data and machine learning analysis, the cleaning cycle is optimized. The system dynamically adjusts the cleaning frequency according to the accumulation situation and flow change of the material. In addition, this automatic cleaning process also effectively extends the service life of the equipment and reduces the need for manual operation. The operator can rely on the automated system for timely monitoring and adjustment to ensure the stability and continuity of the entire material conveying and feeding process. Through the intelligent scraping mechanism, the operation of the hopper 1 is more efficient and clean, which helps to improve the overall efficiency of production.

[0117] Such as Figures 4-5As shown, after the blanking work is completed and the materials on the inner wall of the hopper 1 are completely cleaned, first loosen the nut 19, and then remove the fastening bolt 17. Then, the scraping blade 16 can be pulled out from the connection part to clean the residual materials attached to the scraping blade 16, which is convenient for reuse next time. The scraping blade 16 adopts a detachable connection design, ensuring convenient disassembly and cleaning during operation. This design not only optimizes the cleaning and maintenance process of the scraping blade 16, avoiding material residue or contamination, but also ensures that the scraping blade 16 always maintains a good working state during use, extending the service life of the equipment.

[0118] When cleaning and repairing the stirring blade 15, thanks to the inclined angle design of the feed inlet 21 of the hopper 1, this design effectively increases the area of the feed inlet 21 and improves the convenience of operation. Through this design, the staff can easily take out the stirring rod 14 and the stirring blade 15 from the hopper 1 without complex disassembly procedures, thus simplifying the cleaning and repair process. The inclined angle design not only improves the maintenance efficiency of the equipment, reduces the difficulties that may be encountered during the repair process, ensures the quick inspection and maintenance of the stirring components, but also increases the blanking area and improves the blanking efficiency of the hopper 1 under normal working conditions.

[0119] As Figure 6 shown, the bottom of the automatic material cleaning hopper 1 is equipped with an inclined plate 20. This design optimizes the material flow path and promotes the smoother flow of materials to the discharge pipe 6. The role of the inclined plate 20 significantly improves the discharge speed, reduces the situation of material retention, and thus effectively improves the overall working efficiency.

[0120] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the purpose of the present invention and the scope protected by the claims. These all fall within the protection scope of the present invention.

Claims

1. A new type of hopper with automatic material clearing function, characterized in that: It includes a hopper, a flow sensor and a motor. A support frame is fixedly arranged at the bottom of the hopper, a stirring rod is arranged in the middle of the inner bottom of the hopper, a stirring piece is arranged on the stirring rod, and a motor is arranged at the lower end part of the outer bottom of the hopper; a feed port is arranged at the top of the hopper, a discharge port is arranged at the side of the hopper, the discharge port is connected with a discharge pipe, a pipe cover is arranged at the end of the discharge pipe, and a flow sensor is arranged at the top of the discharge pipe, and the flow sensor is used for real-time monitoring of the material flow in the discharge pipe and detecting whether blockage occurs.

2. The novel hopper with automatic material clearing function according to claim 1 is characterized in that: A bearing seat is arranged in the middle of the bottom of the hopper, the motor is fixedly arranged on the rotating shaft at the bottom end of the hopper, the bottom end of the rotating shaft is transmission-connected with the output end of the motor, the rotating shaft is pivotally connected with the bearing seat and passes through the bearing seat, the shaft sleeve is fixedly arranged on the top of the rotating shaft, a plurality of fixed slots are opened inside the shaft sleeve along the circumferential direction, and an inclined plate is arranged at the bottom of the hopper.

3. The novel hopper with automatic material clearing function according to claim 2 is characterized in that: The stirring rod is connected to the shaft sleeve in the fixed groove through a fixed clamp block, so that the stirring rod is connected to the rotating shaft, and a plurality of stirring pieces are evenly arranged on the side of the stirring rod; A threaded hole groove is provided on the end of the stirring piece away from the stirring rod, and a scraper piece is clamped on the end of the stirring piece away from the stirring rod. A groove is provided at the end of the scraper piece. The end of the stirring piece away from the stirring rod is clamped in the groove at the end of the scraper piece. A nut is provided on the top of the scraper piece. The nut is connected to the fastening bolt and cooperates with the threaded hole groove opened on the top of the stirring piece.

4. The novel hopper with automatic material clearing function according to claim 1 is characterized in that: The flow sensor monitors the flow at the discharge pipe in real time by detecting the vibration changes caused by the material flow; when the material flow is abnormal, the flow sensor compares the real-time material flow with the reference flow when the hopper is working normally, so as to determine whether blockage occurs, and then take corresponding measures to ensure the normal operation of the hopper.

5. The novel hopper with automatic material clearing function according to claim 4 is characterized in that: The flow sensor is equipped with a control system, and a fuzzy adaptive PID control algorithm is adopted in the control system. The flow sensor is used to dynamically adjust the speed of the motor and the material input amount of the feed port according to the deviation between the real-time detected flow and the preset reference flow.

6. The novel hopper with automatic material clearing function according to claim 5 is characterized in that: The fuzzy adaptive PID control algorithm adopted by the control system includes the following steps: S1. Obtain the material flow signal Q(t) detected by the flow sensor in real time, calculate and preset the reference flow Q ref The deviation of e(t)=Q ref -Q(t), taking the flow deviation e(t) and the deviation change rate de / dt as input variables, the correction values ​​of the fuzzy adaptive PID control algorithm parameters Kp, Ki, and Kd are generated through fuzzy inference rules; S2, according to the modified fuzzy adaptive PID control algorithm parameter proportional coefficient K p0 , integral coefficient K i0 , differential coefficient K d0 ; Calculate the motor speed adjustment amount and convert it into the motor speed adjustment instruction and the feed valve opening instruction to achieve speed adaptive adjustment. The specific calculation formula is as follows: S3. When it is detected that the flow rate is continuously lower than the threshold, the high-speed rotation mode of the scraper is triggered, and the feed inlet valve is linked to reduce the material input.

7. The novel hopper with automatic material clearing function according to claim 6 is characterized in that: The method for generating the correction amount of the fuzzy adaptive PID control algorithm parameter by fuzzy inference rules in step S1 comprises the following steps: S101. Define fuzzy sets and membership functions The physical range of the flow deviation e(t) is set according to the flow sensor range, and the physical range of the deviation change rate de / dt is set according to the system dynamic characteristics. Then the deviation e(t) and the deviation change rate de / dt are divided into 5 fuzzy linguistic variables. S102. Design of fuzzy rule base The adjustment of the fuzzy adaptive PID control algorithm parameters needs to be based on the dynamic characteristics of the system. The typical rule logic is as follows: Proportional gain Kp: When the deviation becomes larger, Kp is greatly increased to quickly reduce the deviation. When the deviation is small, Kp is reduced to avoid oscillation. Integral gain Ki: Increasing Ki is used to eliminate static errors, while reducing Ki can cause integral saturation. Differential gain Kd: When the deviation change rate is large, increase Kd to suppress overshoot, and reduce Kd to suppress noise sensitivity; S103, Fuzzy reasoning and defuzzification First, the membership of the current e(t) and de / dt to each fuzzy set is calculated, and then the rule activation is performed. The rule activation is to match the fuzzy input with the conditions in the rule base, calculate the trigger strength of each rule, calculate the minimum operator and product operator through the input membership, and then perform rule logic matching. The corresponding fuzzy set is determined by the input membership, and then all rules containing the fuzzy set are triggered; then the membership function of the output variable is tailored according to the rule strength, and the output of all rules is aggregated at the same time; After the fuzzy reasoning is completed, defuzzification is performed to convert the fuzzy output into an accurate control quantity. The centroid method is usually used to convert the fuzzy output into an accurate value. S104, dynamic update of fuzzy adaptive PID control algorithm parameters The parameter correction formula is as follows: K P =K p0 +ΔK p ×α p K i =K i0 +ΔK i ×α i K d =K d0 +ΔK d ×α d K p0 , K i0 , K d0 is the initial parameter, α p , α i , α d To correct the scaling factor, it is used to match the actual parameter range.

8. The novel hopper with automatic material clearing function according to claim 5 is characterized in that: The control system integrates an LSTM prediction model, trains the LSTM neural network model through historical flow data, predicts the flow trend in the next 5-10 seconds, and adjusts the motor speed in advance to prevent blockage.

9. The novel hopper with automatic material clearing function according to claim 8 is characterized in that: The input parameters of the LSTM prediction model include the historical traffic data sequence {Q(1), Q(2), ..., Q(n)}, ambient temperature T, ambient humidity H, and the output is the traffic prediction value Q for the next 5-10 seconds. pred , when Q is satisfied pred <0.8Q ref When the system triggers the anti-blocking operation, the specific process includes the following steps: S1. LSTM network model construction The input parameters are the historical traffic data sequence {Q(1), Q(2), ..., Q(n)}, ambient temperature T, and ambient humidity. Then the LSTM network architecture is constructed. The input layer includes three neurons: traffic, temperature, and humidity. The hidden layer consists of two layers of stacked LSTM, with 32 neurons in each layer. The activation function uses the ReLU function to accelerate convergence. The output layer is a neuron, which is the predicted traffic value for the next 5-10 seconds. The activation function is linear activation. S2. Model training (1) Dataset construction Select the flow data of the past 60 seconds with a sampling interval of 1 second to capture the periodic characteristics of material flow and obtain the historical flow sequence {Q(1), Q(2), ..., Q(n)}; collect the ambient temperature (T) and humidity (H) in real time through the temperature and humidity sensor to correct the influence of material viscosity on fluidity; obtain a large number of data sets, divide them into training sets and validation sets according to appropriate proportions, including data of normal working conditions and blockage events; (2) Training parameters When training the model, the initial training cycle is selected as 300 times to fully learn the temporal dependency. In order to balance memory usage and gradient stability, the batch size is selected as 16, and the Adam optimizer is selected, the initial learning rate is 0.001, the parameter update step size is automatically adjusted, and the loss function is the mean square error, the formula is: S3. Prediction trigger conditions Blockage judgment condition: When the predicted flow rate Q pred <0.8Q ref When it is determined to be a potential congestion risk; the congestion probability formula is: Among them, Q ref is the set safety reference flow value, Q pred To predict the flow value; That is, when P block When it is >20%, it is judged as a potential blockage risk.

10. The novel hopper with automatic material clearing function according to claim 9 is characterized in that: When a potential blockage risk is identified, an anti-blockage response strategy is adopted; The anti-blocking response strategy is as follows: after determining that there is a risk of blockage, the motor performs speed regulation operation, and the speed is increased to 120% of the rated value through the frequency converter to enhance the fluidity of the material; the scraper starts high-frequency reciprocating motion to peel off the material attached to the inner wall; at this time, the fuzzy adaptive PID control algorithm continues to adjust, and the fuzzy adaptive PID control algorithm parameters are dynamically adjusted according to the predicted deviation e(t) and the deviation change rate de / dt. The proportional coefficient continues to increase as the deviation increases, and the integral coefficient gradually decays when the deviation continues to deviate to prevent integral saturation. The differential coefficient is adaptively adjusted according to the deviation change rate to complete the control closed loop.

Images