Flow monitoring and valve control system for preventing excessive filling

Abstract

The invention discloses a flow monitoring and valve control system for preventing excessive filling, and relates to the technical field of control, and the flow monitoring and valve control system comprises the following steps: carrying out weighted average on instantaneous flow data through a weighted average fusion algorithm to obtain corrected instantaneous flow data; the initial filling flow is calculated by accumulating the corrected instantaneous flow data, and the corrected filling flow is calculated according to the comprehensive correction coefficient and the initial filling flow; the flow change rate is obtained through the instantaneous flow; a flow dynamic threshold value is obtained through a sliding window technology; when the flow change rate is smaller than the flow dynamic threshold value, the current valve opening degree is calculated, and the smooth valve opening degree is calculated through an exponential weighted moving average algorithm; the difference value between the filling flow and the standard filling flow is corrected, control is achieved by adjusting the voltage through a proportional control algorithm, when the difference value is reduced to zero, the valve is closed, excessive filling is avoided, and a solid technical guarantee is provided for filling of the liquid natural gas.

Description

[0001] The invention relates to the field of control technology, in particular to a flow monitoring and valve control system for preventing overfilling. Background Art

[0002] The patent application publication number CN105142176A discloses a method and device for warning of excessive base station mounting. The present invention relates to the field of communications, and is used to warn in advance that the base station is overmounted, thereby preventing the mobile device mounted to the base station from deteriorating in call quality and Internet speed. The mobile position information of all mobile devices in the predicted area is obtained, and the mobile position information includes the current position of the mobile device, the current position offset, the previous position and the previous two positions; the predicted position of each mobile device at the next moment is predicted according to the mobile position information of each mobile device; the predicted mounting amount of each base station in the predicted area at the next moment is calculated according to the predicted position of each mobile device at the next moment; when the predicted mounting amount of the first base station at the next moment exceeds the maximum mounting amount of the first base station, a warning message is sent to the BSS system of the first base station. The embodiment of the present invention is applied to the communication system.

[0003] Although early warning can be given, the accuracy of the prediction is easily affected by the uncertainty of the moving trajectory. Secondly, the reaction lag is large. It only issues a warning when it predicts that the mount volume at the next moment will exceed the threshold. The actual adjustment still requires a certain amount of time. Secondly, the core function is early warning. Although it prevents overload in advance, it will not intervene in the change of mount volume in real time. Finally, the data dimension processed is relatively simple and only relies on the location information of the mobile device. These shortcomings make the system lack sufficient flexibility and real-time adjustment capabilities in a complex and dynamically changing environment, and it is difficult to effectively prevent performance degradation caused by overload.

[0004] In view of this, the present invention proposes a flow monitoring and valve control system for preventing overfilling. Summary of the invention

[0005] In order to overcome the above defects of the prior art and to achieve the above objectives, the present invention provides the following technical solutions: A flow monitoring and valve control system for preventing overfilling, comprising: obtaining instantaneous flow data, basic parameters, valve voltage parameters, signal state parameters and time-effect parameters;

[0006] Correction filling flow calculation module: weighted average the instantaneous flow data through the weighted average fusion algorithm to obtain the corrected instantaneous flow data; calculate the initial filling flow by accumulating the corrected instantaneous flow data, and calculate the corrected filling flow according to the comprehensive correction coefficient and the initial filling flow;

[0007] Standard filling flow calculation module: calculates the standard filling flow according to basic parameters;

[0008] Valve control module: obtain the flow change rate through instantaneous flow; then obtain the flow dynamic threshold through sliding window technology; when the flow change rate is less than the flow dynamic threshold, calculate the current valve opening, and use the exponentially weighted moving average algorithm to calculate the smooth valve opening; correct the difference between the filling flow and the standard filling flow, and adjust the voltage through the proportional control algorithm to control. When the difference becomes smaller until it is 0, the valve is closed and the filling is completed.

[0009] Furthermore, the basic parameters include liquid natural gas temperature, pressure, density and pressure data on both sides of the valve:

[0010] The valve voltage parameters include the current valve voltage, the minimum valve voltage, and the maximum valve voltage;

[0011] The signal status parameters include link quality indication, signal transmission distance, electromagnetic interference signal, and ambient temperature data;

[0012] The timeliness parameters include sending timestamp, receiving timestamp, packet loss rate and retransmission number data.

[0013] Furthermore, the calculation method of correcting the instantaneous flow data includes:

[0014] The collected instantaneous flow data is weighted averaged by the weighted average fusion algorithm to obtain the corrected instantaneous flow data F avg , the formula is: Among them, i is the sampling time number, n is the number of samples, F i is the instantaneous flow data obtained at time i, w i is the weight coefficient at the i-th time; w i Defined according to the corrected signal strength and transmission delay, the formula is: Among them, S i is the corrected signal strength S obtained at the i-th time, τ i is the corrected transmission delay τ obtained at time i;

[0015] The corrected signal strength S and transmission delay τ are calculated by the gated cyclic unit algorithm. The specific acquisition method is:

[0016] The signal state parameters and time parameters are standardized and used as input features, and are organized into a feature vector X-re==[LQI, d, E, H, Ts, Tr, B, R], where LQI is the link quality indicator, d is the signal transmission distance, E is the electromagnetic interference signal, H is the ambient temperature data, Ts is the sending timestamp, Tr is the receiving timestamp, B is the packet loss rate, and R is the number of retransmissions;

[0017] Construct a gated recurrent unit algorithm to calculate the update gate z at time tt , the formula is: t =σ(W z ·[h t-1 ,X_re t ]); where σ is the S igmoid function, W z To update the gate z t The weight matrix of h is obtained by gradient descent method. t-1 is the implicit state at time t-1, X_re t is the input feature vector at time t;

[0018] The reset gate r at time t t , the formula is: t =σ(W r ·[h t-1 ,X_re t ]); where W r To reset the gate r t The weight matrix of

[0019] Current memory state at time t The formula is: Where W h Current memory status The weight matrix, tanh is the hyperbolic tangent function, which generates a nonlinear transformation of the current memory state;

[0020] The final hidden state h at time t t , the formula is:

[0021] After iteration, the gated cyclic unit algorithm outputs the corrected signal strength data and the corrected transmission delay data. The corrected signal strength calculation formula is: S = W o1 ·h t +b1; The corrected transmission delay calculation formula is: τ=W o2 ·h t +b2; where W o1 and W o2 is the output weight, b1 and b2 are biases.

[0022] Furthermore, the initial filling flow rate includes:

[0023] By accumulating and correcting the instantaneous flow data, the initial filling flow data F is calculated. in , the formula is: Among them, t0 and t1 represent the start time and end time of filling respectively, and the range of t is from t0 to t1.

[0024] Furthermore, the method of correcting the filling flow rate includes:

[0025] According to the liquid natural gas temperature and pressure data in the basic parameters, a comprehensive correction coefficient is established to correct the initial filling flow rate and obtain the corrected filling flow rate. The formula of the comprehensive correction coefficient C is: C = (1 + β T (TT std ))(1+β P (PP std )); where β T is the temperature coefficient, β P is the pressure coefficient, T is the acquired temperature data, P is the acquired pressure data, T std is the standard temperature, P std is standard pressure;

[0026] Combined with the comprehensive correction coefficient, the initial filling flow rate is corrected to obtain the corrected filling flow rate. The formula is: F act =F in *C.

[0027] Furthermore, the calculation method of the standard filling flow rate includes:

[0028] The standard filling flow F is calculated based on the density of the liquid natural gas, the pressure data on both sides of the valve and the diameter of the pipeline in the basic parameters. sta , the formula is: in, is the valve flow coefficient, ΔP is the pressure difference on both sides of the valve, A is the cross-sectional area of ​​the pipeline, and the formula is: D is the diameter of the pipe, ρ act is the actual density of liquid natural gas;

[0029] Correction is made according to the comprehensive correction coefficient C, the formula is: ρ act =ρ sta ·C; where ρ sta It is the density of liquid natural gas measured by the sensor.

[0030] Furthermore, the method of obtaining the flow rate change rate includes:

[0031] The filling rate of the flow is calculated by the instantaneous flow rate. The calculation formula of the filling rate is: Where m is the filling rate, Δi is the time interval, F i-Δi is the instantaneous flow rate at time i-Δi;

[0032] The flow rate change rate is calculated by continuous filling rate changes, the formula is: Wherein, a is the flow rate change rate, m(i) is the filling rate m at time i, and m(i-Δi) is the filling rate m at time i-Δi.

[0033] Furthermore, the method of obtaining the dynamic flow threshold by using the sliding window technology includes:

[0034] The average value and standard deviation of the traffic change rate of the last m times are calculated by sliding window technology. The formula of the average value is: in, is the average value of the flow rate change rate, m is the number of recent calculations, a(j) is the flow rate change rate a at the jth moment, and j is the moment m times forward from the i moment;

[0035] The formula for standard deviation is: The flow dynamic threshold is calculated by the mean value and standard deviation of the flow change rate. The formula is: Where k is the adjustment coefficient.

[0036] Furthermore, the method of using the exponentially weighted moving average algorithm to calculate the smoothed valve opening includes:

[0037] When the flow change rate a is less than the flow dynamic threshold a std When the control module is triggered, the current valve opening is calculated. The current valve opening is calculated by the valve voltage parameter. The formula is: Where t' is the current time, O(t') is the current valve opening, V(t') is the valve voltage at t', V min is the minimum valve voltage, V max is the maximum valve voltage;

[0038] The valve opening is filtered using an exponentially weighted moving average algorithm, and the calculation formula is as follows: S(t') = α·O(t')+(1-α)S(t'-1); where S(t') is the current smoothed valve opening, S(t'-1) is the smoothed valve opening at the previous moment, and α is the smoothing coefficient, which ranges from 0 to 1.

[0039] Furthermore, the voltage adjustment method of the proportional control algorithm includes:

[0040] Calculate the standard filling flow F sta and the current corrected filling flow F act The difference is: ΔF = F sta -F act ;

[0041] The current voltage of the control valve is adjusted by the proportional control algorithm. The formula is: V(t') = k·ΔF; k is the proportional coefficient, which controls the response speed of the control voltage to the flow difference. When ΔF is 0, V(t') is 0, then S(t') is 0, the valve is closed, and the filling is ended.

[0042] The technical effects and advantages of the flow monitoring and valve control system for preventing overfilling of the present invention are as follows:

[0043] The present invention can obtain instantaneous flow data, basic parameters, valve voltage parameters, signal state parameters and time-effect parameters; perform weighted averaging on the instantaneous flow data through a weighted average fusion algorithm to obtain corrected instantaneous flow data; calculate the initial filling flow by accumulating the corrected instantaneous flow data, and calculate the corrected filling flow based on the comprehensive correction coefficient and the initial filling flow; calculate the standard filling flow based on the basic parameters; obtain the flow change rate through the instantaneous flow; and then obtain the flow dynamic threshold through the sliding window technology; when the flow change rate is less than the flow dynamic threshold, calculate the current valve opening, and use the exponential weighted moving average algorithm to calculate the smoothed valve opening; adjust the voltage of the control valve through a proportional control algorithm, and when the difference between the standard filling flow and the corrected filling flow reaches 0, the valve is closed and the filling is completed; it has high precision, real-time and adaptability, significantly improves the filling efficiency, and is suitable for precise flow control under complex working conditions, ensuring the stability and safety of equipment operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A schematic diagram of a flow monitoring and valve control system for preventing overfilling according to the present invention;

[0045] Figure 2 It is a schematic diagram of a flow monitoring and valve control method for preventing overfilling according to the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] Example 1

[0048] See also Figure 1 As shown, the present invention provides a flow monitoring and valve control system for preventing overfilling, comprising:

[0049] Data acquisition module: used to obtain instantaneous flow data, basic parameters, valve voltage parameters, signal status parameters and time parameters;

[0050] Correction filling flow calculation module: weighted average the instantaneous flow data through the weighted average fusion algorithm to obtain the corrected instantaneous flow data; calculate the initial filling flow by accumulating the corrected instantaneous flow data, and calculate the corrected filling flow according to the comprehensive correction coefficient and the initial filling flow;

[0051] Standard filling flow calculation module: calculates the standard filling flow according to basic parameters;

[0052] Valve control module: obtain the flow change rate through instantaneous flow; then obtain the flow dynamic threshold through sliding window technology; when the flow change rate is less than the flow dynamic threshold, calculate the current valve opening, and use the exponentially weighted moving average algorithm to calculate the smooth valve opening; correct the difference between the filling flow and the standard filling flow, and adjust the voltage through the proportional control algorithm to control. When the difference becomes smaller until it is 0, the valve is closed and the filling is completed.

[0053] The basic parameters include liquid natural gas temperature, pressure, density and pressure data on both sides of the valve: the valve voltage parameters include the current valve voltage, the minimum valve voltage, and the maximum valve voltage; the signal status parameters include link quality indication, signal transmission distance, electromagnetic interference signal, and ambient temperature data; the timeliness parameters include sending timestamp, receiving timestamp, packet loss rate, and retransmission number data.

[0054] Instantaneous flow sensors are installed in the downstream area of ​​the filling pump and the pipe bending area in the filling pipeline to collect instantaneous flow data. Temperature sensors, pressure sensors and density sensors are installed in the liquefied natural gas storage area to collect temperature, pressure and density data of liquefied natural gas. Pressure sensors are installed on both sides of the valve to collect pressure data on both sides of the valve. Potentiometers are installed on the motors that control the valves to collect valve voltage data.

[0055] Zigbee technology is used to transmit data, and the link quality index is provided by Zigbee. The signal transmission distance is obtained by laser, the electromagnetic interference signal is obtained by electromagnetic interference detection, the ambient temperature is obtained by a temperature sensor, the sending timestamp and the receiving timestamp are obtained by a microcontroller, and the packet loss rate is calculated by comparing the number of sent and received data packets between the sending device and the receiving device. The number of retransmissions is obtained through Zigbee, and Zigbee has a retransmission mechanism.

[0056] The calculation method for correcting instantaneous flow data includes:

[0057] The collected instantaneous flow data is weighted averaged by the weighted average fusion algorithm to obtain the corrected instantaneous flow data F avg , the formula is: Among them, For weighted instantaneous flow calculation, is the weighted sum, and the two are divided for normalization to ensure that the calculated correction result is the weighted average of all collected data. i is the sampling time sequence number, i = 1, 2, 3, ..., n, n is the total amount collected, F i is the instantaneous flow data obtained for the i-th time, w iis the weight coefficient of the ith time, w i Defined according to the corrected signal strength and transmission delay, the formula is: Among them, S i is the corrected signal strength S obtained at the i-th time, τ i is the corrected transmission delay τ obtained at time i;

[0058] The corrected signal strength S and transmission delay τ are calculated by the gated cyclic unit algorithm. The specific acquisition method is:

[0059] After the sensor collects the instantaneous flow, it is transmitted to the collection terminal through Zigbee technology. During the transmission process, the signal will attenuate and the time will be delayed. The signal state parameters and time parameters are standardized and used as input features to form a feature vector x t =[LQI, d, E, H, Ts, Tr, B, R], LQI is link quality indicator, d is signal transmission distance, E is electromagnetic interference signal, H is ambient temperature data, Ts is sending timestamp, Tr is receiving timestamp, B is packet loss rate and R is retransmission number data;

[0060] Construct a gated recurrent unit algorithm to calculate the update gate z at time t t , the formula is: t =σ(W z ·[h t-1 ,X_re t ]); where σ is the S igmoid function, W z To update the gate z t The weight matrix of h is obtained by gradient descent method. t-1 is the implicit state at time t-1, X_re t is the input feature vector at time t;

[0061] The reset gate r at time t t , the formula is: t =σ(W r ·[h t-1 ,X_re t ]); where W r To reset the gate r t The weight matrix of

[0062] Current memory state at time t The formula is: Where W h Current memory status The weight matrix, tanh is the hyperbolic tangent function, which generates a nonlinear transformation of the current memory state;

[0063] The final hidden state h at time tt , the formula is:

[0064] After iteration, the gated cyclic unit algorithm outputs the corrected signal strength data and the corrected transmission delay data. The corrected signal strength calculation formula is: S = W o1 ·h t +b1; The corrected transmission delay calculation formula is: τ=W o2 ·h t +b2; where W o1 and W o2 is the output weight, b1 and b2 are biases.

[0065] The corrected signal strength is an important indicator of data quality during transmission. The stronger the signal, the less interference there is during data transmission and the higher the data reliability. The corrected transmission delay refers to the time delay experienced during data transmission. If the delay is large, it means that the network condition is poor or congestion or retransmission occurs during data transmission, which will reduce the reliability of the data. The corrected signal strength S i and transmission delay τ i Combined, the weight coefficient w is calculated i , reflects the reliability and transmission quality of each data collection, and ensures the final weighted average result F avg More accurate.

[0066] The initial filling flow methods include:

[0067] By accumulating and correcting the instantaneous flow data, the initial filling flow data F is calculated. in , the formula is: Among them, t0 and t1 represent the start time and end time of filling respectively.

[0068] Ways to calibrate the filling flow include:

[0069] According to the liquid natural gas temperature and pressure data in the basic parameters, a comprehensive correction coefficient is established to correct the initial filling flow rate and obtain the corrected filling flow rate. The formula of the comprehensive correction coefficient C is: C = (1 + β T (TT std ))(1+β P (PP std )); where β T is the temperature coefficient, β P is the pressure coefficient. The temperature coefficient and pressure coefficient are obtained through the technical specifications. T is the obtained temperature data, P is the obtained pressure data, and T std is the standard temperature, P std is standard pressure;

[0070] Combined with the comprehensive correction coefficient, the initial filling flow rate is corrected to obtain the corrected filling flow rate. The formula is: F act =F in *C.

[0071] Among them, F is the instantaneous flow rate, which is collected by the flow sensor, F avg is the corrected instantaneous flow rate, F in It is the initial filling flow obtained by accumulating and correcting the instantaneous flow, but the initial filling flow will be affected by the temperature and pressure of the liquid natural gas. The comprehensive correction coefficient C is established by the temperature and pressure of the liquid natural gas to obtain the corrected filling flow.

[0072] The calculation method of standard filling flow includes:

[0073] The standard filling flow F is calculated based on the density of the liquid natural gas, the pressure data on both sides of the valve and the diameter of the pipeline in the basic parameters. sta , the formula is: in, is the valve flow coefficient, which can be obtained from the manual. ΔP is the pressure difference on both sides of the valve, and A is the cross-sectional area of ​​the pipeline. The formula is: D is the diameter of the pipe, ρ act is the actual density of liquid natural gas;

[0074] Correction is made according to the comprehensive correction coefficient C, the formula is: ρ act =ρ sta ·C; where ρ sta It is the density of liquid natural gas measured by the sensor.

[0075] Methods for obtaining the flow rate change rate include:

[0076] The filling rate of the flow is calculated by the instantaneous flow rate. The calculation formula of the filling rate is: Where m is the filling rate, Δi is the time interval, F i-Δi is the instantaneous flow rate at time i-Δi;

[0077] The flow rate change rate is calculated by continuous filling rate changes, the formula is: Wherein, a is the flow rate change rate, m(i) is the filling rate at time i, and m(i-Δi) is the filling rate at time i-Δi.

[0078] The methods of obtaining dynamic traffic thresholds through sliding window technology include:

[0079] The average value and standard deviation of the traffic change rate of the last m times are calculated by sliding window technology. The formula of the average value is: in, is the average value of the flow rate change rate, m is the number of recent calculations, a(j) is the flow rate change rate a at the jth moment, and j is the moment m times forward from the i moment;

[0080] The formula for standard deviation is: The flow dynamic threshold is calculated by the mean value and standard deviation of the flow change rate. The formula is: Where k is the adjustment coefficient;

[0081] Sliding window technology is an algorithmic technology used to process dynamic data streams or sequence data. It maintains a fixed-size window and slides over time to analyze the latest subset of a piece of data without processing the entire data set.

[0082] Ways to calculate smoothed valve opening using the exponentially weighted moving average algorithm include:

[0083] When the flow change rate a is less than the flow dynamic threshold a std When the control module is triggered, the current valve opening is calculated. The current valve opening is calculated by the valve voltage parameter. The calculation formula is as follows: Where t' is the current time, O(t') is the current valve opening, V(t') is the valve voltage at t', V min is the minimum valve voltage, V max is the maximum valve voltage;

[0084] The valve opening is filtered using an exponentially weighted moving average algorithm. The calculation formula is as follows: S(t') = α·O(t')+(1-α)S(t'-1); where S(t') is the current smoothed valve opening, S(t'-1) is the smoothed valve opening at the previous moment, and α is the smoothing coefficient. The range of α is 0 to 1, and the smoothing coefficient is set by empirical rules.

[0085] The proportional control algorithm adjusts the voltage in the following ways:

[0086] Calculate the standard filling flow F sta and the current corrected filling flow F act The difference is: ΔF = F sta -F act ;

[0087] The current voltage of the control valve is adjusted by the proportional control algorithm, and the formula is: V(t') = k·ΔF; k is the proportional coefficient, which is determined by the existing reference standard and controls the response speed of the voltage to the flow difference. When ΔF is 0, V(t') is 0, then S(t') is 0, the valve is closed, and the filling is ended.

[0088] The present embodiment can realize efficient and safe filling management of liquefied natural gas. The system includes a data acquisition module, a correction filling flow calculation module, a standard filling flow calculation module and a valve control module; instantaneous flow data, basic parameters, valve voltage parameters, signal state parameters and time parameters are obtained; the instantaneous flow data is weighted averaged by a weighted average fusion algorithm to obtain the corrected instantaneous flow data; the initial filling flow is calculated by accumulating the corrected instantaneous flow data, and the corrected filling flow is calculated by the comprehensive correction coefficient and the initial filling flow; the standard filling flow is calculated according to the basic parameters; the flow change rate is obtained by the instantaneous flow; and the flow dynamic threshold is obtained by the sliding window technology; when the flow change rate is less than the flow dynamic threshold, the current valve opening is calculated, and the smoothed valve opening is calculated using the exponential weighted moving average algorithm; the voltage of the control valve is adjusted by the proportional control algorithm, and when the difference between the standard filling flow and the corrected filling flow reaches 0, the valve is closed and the filling is completed; it has high precision, real-time and adaptability, significantly improves the filling efficiency, and is suitable for precise flow control under complex working conditions to ensure the stability and safety of equipment operation.

[0089] Example 2

[0090] See also Figure 2 As shown, the part not described in detail in this embodiment is described in Example 1, which provides a flow monitoring and valve control method to prevent overfilling, including:

[0091] S1. Obtain instantaneous flow data, basic parameters, valve voltage parameters, signal status parameters and time-effect parameters;

[0092] S2. Perform weighted averaging on the instantaneous flow data by a weighted average fusion algorithm to obtain corrected instantaneous flow data; calculate the initial filling flow by accumulating the corrected instantaneous flow data, and calculate the corrected filling flow based on the comprehensive correction coefficient and the initial filling flow;

[0093] S3. Calculate the standard filling flow rate according to the basic parameters;

[0094] S4. Obtain the flow rate change rate through instantaneous flow; then obtain the flow dynamic threshold through sliding window technology; when the flow rate change rate is less than the flow dynamic threshold, calculate the current valve opening, and use the exponentially weighted moving average algorithm to calculate the smoothed valve opening; correct the difference between the filling flow and the standard filling flow, and adjust the voltage through the proportional control algorithm to control, when the difference becomes smaller until it is 0, the valve is closed and the filling is completed.

[0095] Example 3

[0096] This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the operation mode of the above-mentioned off-site payment security protection method is implemented.

[0097] Since the electronic device introduced in this embodiment is an electronic device used to implement a remote payment security protection method in the embodiment of this application, based on the remote payment security protection method introduced in the embodiment of this application, the technical personnel of this field can understand the specific implementation of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application is not described in detail here. As long as the technical personnel of this field implement the electronic device used in the remote payment security protection method in the embodiment of this application, it belongs to the scope of protection to be protected by this application.

[0098] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0099] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical users in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A flow monitoring and valve control system for preventing overfilling, characterized in that: include: Data acquisition module: used to obtain instantaneous flow data, basic parameters, valve voltage parameters, signal status parameters and time parameters; Correction filling flow calculation module: weighted average the instantaneous flow data through the weighted average fusion algorithm to obtain the corrected instantaneous flow data; calculate the initial filling flow by accumulating the corrected instantaneous flow data, and calculate the corrected filling flow according to the comprehensive correction coefficient and the initial filling flow; Standard filling flow calculation module: calculates the standard filling flow according to basic parameters; Valve control module: obtain the flow rate change rate through instantaneous flow; Then, the dynamic threshold of the flow rate is obtained through the sliding window technology. When the flow rate change rate is less than the dynamic threshold of the flow rate, the current valve opening is calculated, and the exponentially weighted moving average algorithm is used to calculate the smoothed valve opening. The difference between the filling flow rate and the standard filling flow rate is corrected, and the voltage is adjusted through the proportional control algorithm to control it. When the difference becomes smaller until it reaches 0, the valve is closed and the filling is completed.

2. A flow monitoring and valve control system for preventing overfilling according to claim 1, characterized in that: The basic parameters include liquid natural gas temperature, pressure, density and pressure data on both sides of the valve: The valve voltage parameters include the current valve voltage, the minimum valve voltage, and the maximum valve voltage; The signal status parameters include link quality indication, signal transmission distance, electromagnetic interference signal, and ambient temperature data; The timeliness parameters include sending timestamp, receiving timestamp, packet loss rate and retransmission number data.

3. A flow monitoring and valve control system for preventing overfilling according to claim 1, characterized in that: The calculation method of correcting the instantaneous flow data includes: The collected instantaneous flow data is weighted averaged by the weighted average fusion algorithm to obtain the corrected instantaneous flow data F avg , the formula is: Among them, i is the sampling time number, n is the number of samples, F i is the instantaneous flow data obtained at time i, w i is the weight coefficient at the i-th time; w i Defined according to the corrected signal strength and transmission delay, the formula is: Among them, S i is the corrected signal strength S obtained at the i-th time, τ i is the corrected transmission delay τ obtained at time i; The corrected signal strength S and transmission delay τ are calculated by the gated cyclic unit algorithm. The specific acquisition method is: The signal state parameters and time parameters are standardized and used as input features, and are organized into a feature vector X_re==[LQI, d, E, H, Ts, Tr, B, R], where LQI is the link quality indicator, d is the signal transmission distance, E is the electromagnetic interference signal, H is the ambient temperature data, Ts is the sending timestamp, Tr is the receiving timestamp, B is the packet loss rate, and R is the number of retransmissions; Construct a gated recurrent unit algorithm to calculate the update gate z at time t t , the formula is: t =σ(W z ·[h t-1 ,X_re t ]); where σ is the Sigmoid function, W z To update the gate z t The weight matrix of h is obtained by gradient descent method. t-1 is the implicit state at time t-1, X_re t is the input feature vector at time t; The reset gate r at time t t , the formula is: t =σ(W r ·[h t-1 ,X_re t ]); where W r To reset the gate r t The weight matrix of Current memory state at time t The formula is: Where W h Current memory status The weight matrix, tanh is the hyperbolic tangent function, which generates a nonlinear transformation of the current memory state; The final hidden state h at time t t , the formula is: After iteration, the gated cyclic unit algorithm outputs the corrected signal strength data and the corrected transmission delay data. The corrected signal strength calculation formula is: S = W o1 ·h t +b1; The corrected transmission delay calculation formula is: τ=W o2 ·h t +b2; where W o1 and W o2 is the output weight, b1 and b2 are biases.

4. A flow monitoring and valve control system for preventing overfilling according to claim 3, characterized in that: The initial filling flow rate includes: By accumulating and correcting the instantaneous flow data, the initial filling flow data F is calculated. in , the formula is: Among them, t0 and t1 represent the start time and end time of filling respectively, and the range of t is from t0 to t1.

5. A flow monitoring and valve control system for preventing overfilling according to claim 4, characterized in that: The method of correcting the filling flow rate includes: The comprehensive correction coefficient is established according to the liquid natural gas temperature and pressure data in the basic parameters to correct the initial filling flow rate and obtain the corrected filling flow rate. The formula of the comprehensive correction coefficient C is: C=(1+β T (TT std ))(1+β P (PP std )); where β T is the temperature coefficient, β P is the pressure coefficient, T is the acquired temperature data, P is the acquired pressure data, T std is the standard temperature, P std is standard pressure; Combined with the comprehensive correction coefficient, the initial filling flow rate is corrected to obtain the corrected filling flow rate. The formula is: F act =F in *C.

6. A flow monitoring and valve control system for preventing overfilling according to claim 5, characterized in that: The calculation method of the standard filling flow rate includes: The standard filling flow F is calculated based on the density of the liquid natural gas, the pressure data on both sides of the valve and the diameter of the pipeline in the basic parameters. sta , the formula is: in, is the valve flow coefficient, ΔP is the pressure difference on both sides of the valve, A is the cross-sectional area of ​​the pipeline, and the formula is: D is the diameter of the pipe, ρ act is the actual density of liquid natural gas; Correction is made according to the comprehensive correction coefficient C, the formula is: ρ act =ρ sta ·C; where ρ sta It is the density of liquid natural gas measured by the sensor.

7. A flow monitoring and valve control system for preventing overfilling according to claim 6, characterized in that: The method of obtaining the flow rate change rate includes: The filling rate of the flow is calculated by the instantaneous flow rate. The calculation formula of the filling rate is: Where m is the filling rate, Δi is the time interval, F i-Δi is the instantaneous flow rate at time i-Δi; The flow rate change rate is calculated by continuous filling rate changes, the formula is: Wherein, a is the flow rate change rate, m(i) is the filling rate m at time i, and m(i-Δi) is the filling rate m at time i-Δi.

8. A flow monitoring and valve control system for preventing overfilling according to claim 7, characterized in that: The method of obtaining the dynamic flow threshold by using the sliding window technology includes: The average value and standard deviation of the traffic change rate of the last m times are calculated by sliding window technology. The formula of the average value is: in, is the average value of the flow rate change rate, m is the number of recent calculations, a(j) is the flow rate change rate a at the jth moment, and j is the moment m times forward from the i moment; The formula for standard deviation is: The flow dynamic threshold is calculated by the mean value and standard deviation of the flow change rate. The formula is: Where k is the adjustment coefficient.

9. A flow monitoring and valve control system for preventing overfilling according to claim 8, characterized in that: The method of using the exponentially weighted moving average algorithm to calculate the smoothed valve opening includes: When the flow change rate a is less than the flow dynamic threshold a std When the control module is triggered, the current valve opening is calculated. The current valve opening is calculated by the valve voltage parameter. The formula is: Among them, t ‘ is the current moment, O(t ‘ ) is the current valve opening, V(t ‘ ) is t ‘ The valve voltage at min is the minimum valve voltage, V max is the maximum valve voltage; The valve opening is filtered using the exponentially weighted moving average algorithm, and the formula is: S(t ‘ )=α·O(t ‘ )+(1-α)S(t ‘ -1); where S(t ‘ ) is the current smooth valve opening, S(t ‘ -1) is the smoothed valve opening at the previous moment, α is the smoothing coefficient, and α ranges from 0 to 1.

10. A flow monitoring and valve control system for preventing overfilling according to claim 9, characterized in that: The proportional control algorithm adjusts the voltage in the following ways: Calculate the standard filling flow F sta and the current corrected filling flow F act The difference ΔF is: ΔF = F sta -F act ; The current voltage of the control valve is adjusted through the proportional control algorithm. The formula is: V(t ‘ ) = k·ΔF; k is the proportionality coefficient, which is determined by the existing reference standard and controls the response speed of the voltage to the flow difference. When ΔF is 0, V(t ‘ ) is 0, then S(t ‘ ) is 0, the valve is closed and filling is finished.

Images