A proportional flow control system and device

Abstract

The invention relates to the technical field of proportional flow detection and control, and discloses a proportional flow control system and device, including a data module for collecting flow information, preprocessing the collected flow information, and extracting a feature vector of the processed flow information, a processing module for monitoring and predicting flow information characteristics and evaluating flow prediction model performance indicators, a control module for judging flow information, reflecting instantaneous changes in flow queues, reflecting long-term changes in flow queues, monitoring flow buffer queue information, and adjusting flow bandwidth packet loss rate, and a feedback module for feeding back flow information, flow arrival rate, flow transmission link status, and flow buffer status, and controlling flow transmission pipeline valves, thereby realizing intelligent flow detection and intelligent flow control.

Description

Technical Field

[0001] The invention relates to the technical field of proportional flow detection and control, and discloses a proportional flow control system and a device. Background Art

[0002] With the growing development of the electronic materials industry and changes in technology, there are more and more types of manufacturing processes, which are mainly used in the electronic materials equipment end. During the manufacturing process, continuous variable liquid supply is required. At present, the process end cannot achieve continuous variable liquid supply, and can only supply liquid or cut off liquid, but cannot achieve continuous variable liquid supply. The purpose of this solution is to meet the needs of the electronic materials equipment end and to be able to provide continuous variable liquid supply to the equipment end.

[0003] For example, the existing Chinese patent with the authorization publication number CN110462269B discloses a flow control device that can be used continuously when used in a continuously operating device and a flow control method for the flow control device. The fluid is flow-controlled through one of the control valves (5) respectively arranged in the branch roads (P1, P2, P3). When it is determined based on the signal from the monitoring unit (5a) of the control valve (5) that the control valve (5) needs to be replaced, the control valve (5) that is performing flow control is closed, and the control valve (5) set in other branch roads (P1, P2, P3) is opened and the flow control is performed in the same manner as before the replacement. In addition, the on-off valves (12, 13) arranged on the upstream and downstream sides of the closed control valve (5) are also closed.

[0004] However, the above patents have the following problems: there is no built-in learning mechanism or traffic prediction function, it cannot prevent problems in advance or optimize traffic distribution, it can only control the traffic on a single path at a time, it cannot fine-tune multiple paths at the same time, and there is no built-in performance evaluation unit, making it difficult to determine whether the control effect is optimal. Summary of the invention

[0005] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0006] In order to solve the above technical problems, the main purpose of the present invention is to provide a proportional flow control system, comprising:

[0007] The data module includes a data collection unit for collecting flow information, a data processing unit for preprocessing the collected flow information, and a feature extraction unit for extracting feature vectors of the processed flow information;

[0008] A processing module, a flow prediction model for monitoring and predicting flow information characteristics, and a performance evaluation unit for evaluating performance indicators of the flow prediction model;

[0009] The control module includes a determination unit for determining flow information, a transient unit for reflecting instantaneous changes in flow queues, a steady-state unit for reflecting long-term changes in flow queues, a monitoring unit for monitoring flow buffer queue information, and an adjustment unit for adjusting flow bandwidth packet loss rate;

[0010] The feedback module is used to feed back flow information, flow rate, flow transmission link status and flow buffer status, and control the flow transmission pipeline valve.

[0011] As a preferred solution of a proportional flow control system of the present invention, wherein:

[0012] The preprocessing includes filtering, denoising, calibration and normalization;

[0013] The feature extraction is used to extract features from the preprocessed flow information.

[0014] As a preferred solution of a proportional flow control system of the present invention, wherein:

[0015] Taking the historical traffic data of the past n time periods as basic data, the basic data is divided into a training set and a test set, wherein the basic data includes traffic transmission links, link numbers and traffic values, and the traffic prediction model initialized by training the training set is used to map all traffic transmission links through the trained traffic prediction model;

[0016] The basic data obtains m hidden traffic information through the traffic integration model, and the m hidden traffic information is input into the traffic detection model. The traffic detection model sets a multi-layer traffic perception layer, and calculates the m hidden information weights of multiple traffic transmission links through the multi-layer traffic perception layer. The traffic detection model predicts the traffic change information vector of each traffic transmission link by iterating the traffic changes of multiple traffic transmission links, and finally integrates them into the global traffic change to output the traffic prediction value.

[0017] As a preferred solution of a proportional flow control system of the present invention, wherein:

[0018] Taking the characteristic vector of the traffic as input, obtaining the hidden information of the traffic through the traffic integration model;

[0019] The flow integration model includes flow information substitution coefficients and flow information writing values;

[0020] The flow information substitution coefficient is used to control the degree coefficient of the flow information of the previous timestamp being substituted into the flow information at the current moment, and the calculation expression is as follows:

[0021]

[0022] in, is the degree coefficient of the flow information of the previous timestamp being substituted into the flow information of the current moment, is the activation function, Write the coefficient weight coefficient for the flow information, Enter the flow information for time t, is the traffic transmission link at time t, is the flow information of the hth flow transmission link at time t, Substitute the coefficient bias value for the flow information;

[0023] The flow information write value is used to control the value of the flow information at the previous moment written into the current flow information, further capturing the dependency of the flow time series on the flow at the current moment. The calculation expression is as follows:

[0024]

[0025] in, is the coefficient value of the flow information at the previous moment being written into the current flow information, Write value weight coefficient for traffic information, is the flow information of the hth flow transmission link at time t, The offset value for writing the flow information;

[0026] The flow detection model calculation expression is as follows:

[0027]

[0028] in, is the weight coefficient of the i-th traffic perception layer, is the weight coefficient of the first layer traffic perception layer, is the weight coefficient of the kth traffic perception layer, Input traffic information to the traffic perception layer at the first moment, Enter the flow information for the t-1th moment, is the bias value of the first layer traffic perception layer, is the bias value of the i-th traffic perception layer, is the bias value of the kth traffic perception layer, e is the exponential constant, n is the current traffic perception layer calculation, k is the total number of traffic perception layers, The model operator indicates that the traffic detection model is running to calculate the traffic information.

[0029] As a preferred solution of a proportional flow control system of the present invention, wherein:

[0030] The performance evaluation unit evaluates the prediction capability of the flow detection model by calculating the accuracy of the flow information prediction value output by the flow detection model. The calculation expression is as follows:

[0031]

[0032] in, The accuracy of the traffic information prediction value output by the traffic detection model, is the real-time flow value, Output the traffic information prediction value for the traffic detection model, t is the time, T is the total time of traffic calculation;

[0033] The accuracy of the traffic information prediction value output by the traffic detection model ranges from 0 to 1. The greater the accuracy, the more accurate the traffic information prediction value output by the traffic detection model is.

[0034] As a preferred solution of a proportional flow control system of the present invention, wherein:

[0035] The transient unit obtains the arrival rate of the traffic at the next moment by receiving the traffic information change rate of the traffic transmission link at the current moment and the traffic information prediction value output by the traffic detection model, and pre-calculates the change rate of the traffic transmitted by the traffic transmission link at the next moment. The higher the arrival rate, the faster the traffic transmission speed, and the lower the arrival rate, the slower the traffic transmission speed. If the arrival rate is 0, the traffic transmission speed is moderate.

[0036] The steady-state unit is used to obtain average flow information of the flow transmission link.

[0037] As a preferred solution of a proportional flow control system of the present invention, wherein:

[0038] The determination unit receives the traffic information prediction value output by the traffic detection model, determines the traffic information of the traffic transmission link, and introduces a small amount of traffic into the traffic buffer queue if the traffic information of the traffic transmission link is less than the average traffic information and the traffic information tends to increase with traffic bursts;

[0039] If the flow information of the traffic delivery link is less than the average flow information and the flow information decreases, it is a flow attenuation suppression state, and all flows are allowed to enter the flow buffer queue;

[0040] If the flow information of the traffic transport link is greater than the average flow information and the amount of flow information increases, the control flow enters the flow buffer queue;

[0041] The adjusting unit receives the judgment result of the flow transmission link flow information output by the judgment unit and the arrival rate output by the transient unit. If the flow load is greater than the maximum threshold, the flow load is released, that is, all flow data packets are discarded. If the flow load is not greater than the maximum threshold, the flow data packets are discarded in a dispersed manner.

[0042] The flow data packet is used to distribute and adjust the flow transmission link through the transmission pipeline and the automatic control ball valve;

[0043] The monitoring unit detects the flow buffer by monitoring the flow arrival rate and flow information of the flow buffer.

[0044] As a preferred solution of a proportional flow control device of the present invention, wherein:

[0045] A proportional flow control device comprises a base portion 1, a liquid level tank 2 and a delivery pipeline 3;

[0046] One end of the delivery pipeline 3 is connected to the liquid input device, and the other end is connected to the liquid level tank 2;

[0047] The liquid level tank 2 is located at the top of the base part 1;

[0048] The base part 1 is used to support a liquid level tank 2 .

[0049] As a preferred solution of a proportional flow control device of the present invention, wherein:

[0050] The base portion 1 includes a base bracket 11 and a base support leg 12;

[0051] The delivery pipeline 3 includes a delivery pipe 31, a first manual ball valve 32, a proportional control valve 33, a second manual ball valve 34, a pressure regulating valve 35, a third manual ball valve 36 and a signal line pipe 37;

[0052] One end of the first manual ball valve 32 is connected to the delivery pipe 31, and the other end is connected to the proportional regulating valve 33;

[0053] The other end of the proportional control valve 33 is connected to the second manual ball valve 34, and one side of the proportional control valve is connected to the signal line pipe 37, and the pressure regulating valve 35 and the proportional control valve 33 are connected through the signal line pipe 37;

[0054] The other end of the proportional regulating valve 33 is connected to the second manual ball valve 34 , and the other end of the second manual ball valve 34 is connected to the pressure regulating valve 35 ;

[0055] The other end of the pressure regulating valve 35 is connected to a third manual ball valve 36;

[0056] A signal line pipe 37 is disposed on one side of the pressure regulating valve 35 , and the pressure regulating valve 35 is connected to the proportional control valve 33 via the signal line pipe 37 .

[0057] As a preferred solution of a proportional flow control device of the present invention, wherein:

[0058] The liquid level tank 2 comprises a tank body 22, and a plurality of spring legs 21 are arranged on the side wall of the tank body 22, and the distances between the plurality of spring legs 21 are equal and symmetrical, so as to support and fix the liquid level tank 2 on the base part 1;

[0059] A liquid level gauge 23 is provided on the side wall of the tank body 22, and the liquid level gauge 23 is perpendicular to the horizontal ground and is used to measure the liquid level in the liquid level tank 2;

[0060] A liquid discharge port 24 is disposed at one end of the tank body 22 for discharging liquid inside the tank body 22 .

[0061] As a preferred solution of a proportional flow control device of the present invention, wherein:

[0062] The top of the liquid level tank 2 is respectively provided with a fourth spare port 25 , a reflux port 26 , a pure water port 27 , a nitrogen port 28 , a first spare port 29 , a second spare port 210 and a third spare port 211 .

[0063] Beneficial effects of the present invention:

[0064] This application is trained based on historical data and uses a multi-layer traffic perception layer to calculate the weight of hidden information to predict traffic changes, so that the system can prepare in advance to deal with possible traffic fluctuations, ensure that the system's predictive ability always remains at a high level, and adjust in time to adapt to new traffic trends. The judgment unit, transient unit, steady-state unit and adjustment unit work together to respond quickly according to the actual state of the traffic and maintain efficient operation of the network. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0066] Figure 1 This is a system composition diagram of a proportional flow control system of the present invention;

[0067] Figure 2 It is an overall schematic diagram of a proportional flow control device of the present invention;

[0068] Figure 3This is a structural diagram of a base of a proportional flow control device of the present invention;

[0069] Figure 4 This is a bottom structural diagram of a liquid level tank of a proportional flow control device of the present invention;

[0070] Figure 5 This is a top structural diagram of a liquid level tank of a proportional flow control device of the present invention;

[0071] Figure 6 This is a structural diagram of a delivery pipeline of a proportional flow control device of the present invention;

[0072] Figure numerals: 1. base part; 2. liquid level tank; 3. delivery pipeline; 11. base bracket; 12. base support leg; 21. spring support foot; 22. tank body; 23. liquid level gauge; 24. drain port; 25. fourth spare interface; 26. reflux port; 27. pure water port; 28. nitrogen port; 29. ​​first spare port; 210. second spare port; 211. third spare port; 31. delivery pipe; 32. first manual ball valve; 33. proportional control valve; 34. second manual ball valve; 35. pressure regulating valve; 36. third manual ball valve; 37. signal line pipe. DETAILED DESCRIPTION

[0073] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0074] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0075] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0076] Embodiment 1: like Figure 1 As shown, a proportional flow control system comprises:

[0077] A processing module, a flow prediction model for monitoring and predicting flow information characteristics, and a performance evaluation unit for evaluating performance indicators of the flow prediction model;

[0078] Among them, preprocessing includes filtering, denoising, calibration and normalization;

[0079] Feature extraction is used to extract features from preprocessed traffic information;

[0080] A processing module, a flow prediction model for monitoring and predicting flow information characteristics, and a performance evaluation unit for evaluating performance indicators of the flow prediction model;

[0081] The historical traffic data of the past n time periods is used as the basic data, and the basic data is divided into a training set and a test set. The basic data includes traffic transmission links, link numbers, and traffic values. The traffic prediction model initialized by the training set is trained, and all traffic transmission links are mapped by the trained traffic prediction model.

[0082] The basic data obtains m hidden traffic information through the traffic integration model, and inputs the m hidden traffic information into the traffic detection model. The traffic detection model sets a multi-layer traffic perception layer, and calculates the m hidden information weights of multiple traffic transmission links through the multi-layer traffic perception layer. The traffic detection model predicts the traffic change information vector of each traffic transmission link by iterating the traffic changes of multiple traffic transmission links, and finally integrates them into the global traffic change, and outputs the traffic prediction value;

[0083] Taking the characteristic vector of traffic as input, the hidden information of traffic is obtained through the traffic integration model;

[0084] Furthermore, traffic hidden information includes traffic information, traffic feature information, etc., which can cover the parameters of traffic characteristics. The traffic information is first converted into feature vectors, which contain important information about the traffic, such as traffic size, direction, rate, etc. These feature vectors are provided as input to the traffic integration model.

[0085] The traffic integration model includes traffic information substitution coefficients and traffic information writing values;

[0086] The traffic integration model consists of two key components: traffic information substitution coefficients and traffic information writing values, which work together to capture the trend of traffic changes over time.

[0087] The flow information substitution coefficient controls the degree of influence of the flow information of the previous timestamp on the flow information at the current moment. This allows the system to take into account past information while adjusting the weights according to the current situation. The activation function σ ensures that the output value is within a reasonable range, and the flow information write value determines how the flow information of the previous moment affects the specific value of the current flow information, thereby better capturing the dependencies in the time series;

[0088] The flow information substitution coefficient is used to control the degree to which the flow information of the previous timestamp is substituted into the flow information of the current moment. The calculation expression is as follows:

[0089]

[0090] in, is the degree coefficient of the flow information of the previous timestamp being substituted into the flow information of the current moment, is the activation function, Write the coefficient weight coefficient for the flow information, Enter the flow information for time t, is the traffic transmission link at time t, is the flow information of the hth flow transmission link at time t, Substitute the coefficient bias value for the flow information;

[0091] The flow information write value is used to control the value of the flow information at the previous moment written into the current flow information, further capturing the dependency of the flow time series on the flow at the current moment. The calculation expression is as follows:

[0092]

[0093] in, is the coefficient value of the flow information at the previous moment being written into the current flow information, Write value weight coefficient for traffic information, is the flow information of the hth flow transmission link at time t, The offset value for writing the flow information;

[0094] The traffic detection model uses a multi-layer perceptron structure, and each layer has its own weight coefficient and bias value. These parameters are continuously adjusted through the training process to optimize the model performance. The model can calculate the hidden information weights of different links and iteratively predict the traffic changes on each link, and finally integrate them into the global traffic changes and output the predicted value;

[0095] The traffic detection model calculation expression is as follows:

[0096]

[0097] in, is the weight coefficient of the i-th traffic perception layer, is the weight coefficient of the first layer traffic perception layer, is the weight coefficient of the kth traffic perception layer, Input traffic information to the traffic perception layer at the first moment, Enter the flow information for the t-1th moment, is the bias value of the first layer traffic perception layer, is the bias value of the i-th traffic perception layer, is the bias value of the kth traffic perception layer, e is the exponential constant, n is the current traffic perception layer calculation, k is the total number of traffic perception layers, The model operator represents that the traffic detection model is running to calculate the traffic information;

[0098] Using exponential functions in traffic detection models can introduce nonlinear elements, which helps the model adapt to complex data distribution more flexibly;

[0099] By combining time series characteristics and machine learning algorithms, the system can make more accurate traffic predictions and reduce errors. The system can quickly respond to changes in traffic and adjust strategies in a timely manner to ensure the stable operation of the network or fluid transmission system. Over time, the system will continue to learn new traffic patterns and automatically adjust parameters to cope with unknown conditions, improving long-term stability. Better traffic predictions mean that bandwidth or other resources can be allocated more efficiently to avoid waste or shortages. Sophisticated traffic analysis helps identify abnormal traffic patterns, warn of potential problems in advance, and enhance system security. Automated traffic management and prediction reduce the need for manual intervention and reduce maintenance costs.

[0100] The calculation method of accuracy is actually a quantification of the prediction error at each moment. Ideally, if the predicted value is exactly equal to the actual value, the accuracy is 1; conversely, if the prediction is completely wrong, the accuracy is 0. Any value within this range represents a different degree of match.

[0101] The performance evaluation unit evaluates the prediction capability of the traffic detection model by calculating the accuracy of the traffic information prediction value output by the traffic detection model. The calculation expression is as follows:

[0102]

[0103] in, The accuracy of the traffic information prediction value output by the traffic detection model, is the real-time flow value, Output the traffic information prediction value for the traffic detection model, t is the time, T is the total time of traffic calculation;

[0104] The accuracy rate is a quantitative indicator that can be used to directly and objectively evaluate the prediction capability of the traffic detection model and help understand whether the model's performance meets expectations. Regular evaluation of the accuracy rate can identify the changing trend of the model's performance over time and promptly discover potential problems or signs of degradation, so that measures can be taken to optimize or retrain the model.

[0105] It provides an intuitive and effective means to evaluate the performance of the traffic detection model and ensure that it can reliably serve the actual application needs. By regularly monitoring the accuracy, it can not only maintain the optimal performance of the model, but also lay a solid foundation for the long-term stable operation of the system;

[0106] The accuracy of the traffic information prediction value output by the traffic detection model ranges from 0 to 1. The greater the accuracy, the more accurate the traffic information prediction value output by the traffic detection model is.

[0107] The control module includes a determination unit for determining flow information, a transient unit for reflecting instantaneous changes in flow queues, a steady-state unit for reflecting long-term changes in flow queues, a monitoring unit for monitoring flow buffer queue information, and an adjustment unit for adjusting flow bandwidth packet loss rate;

[0108] The transient unit obtains the arrival rate of the traffic at the next moment by receiving the traffic information change rate of the traffic transmission link at the current moment and the traffic information prediction value output by the traffic detection model, and pre-calculates the change rate of the traffic transmission link at the next moment. The higher the arrival rate, the faster the traffic transmission speed, and the lower the arrival rate, the slower the traffic transmission speed. If the arrival rate is 0, the traffic transmission speed is moderate.

[0109] The steady-state unit is used to obtain the average flow information of the flow transmission link;

[0110] Further, the determination unit receives the traffic information prediction value output by the traffic detection model, and determines the traffic information of the traffic transmission link. If the traffic information of the traffic transmission link is less than the average traffic information, and the traffic information tends to increase with traffic bursts, a small amount of traffic is introduced into the traffic buffer queue;

[0111] If the flow information of the traffic delivery link is less than the average flow information and the flow information decreases, it is a flow attenuation suppression state, and all flows are allowed to enter the flow buffer queue;

[0112] If the flow information of the traffic transport link is greater than the average flow information and the amount of flow information increases, the control flow enters the flow buffer queue;

[0113] The adjustment unit receives the judgment result of the flow transmission link flow information output by the judgment unit and the arrival rate output by the transient unit. If the flow load is greater than the maximum threshold, the flow load is released, that is, all flow data packets are discarded. If the flow load is not greater than the maximum threshold, the flow data packets are discarded in a dispersed manner.

[0114] The flow data packet is used to distribute and adjust the flow transmission link through the transmission pipeline and the automatic control ball valve;

[0115] The monitoring unit detects the flow buffer by monitoring the flow arrival rate and flow information of the flow buffer;

[0116] The principle includes that the transient unit detects the transient change of the pipeline flow queue, the steady-state unit detects the average change of the pipeline queue, the determination unit receives the transient change and the average change to determine the pipeline flow load situation, and releases the flow into the flow buffer according to the situation, and then detects the flow buffer through the monitoring unit to determine the amount of flow load released into the flow buffer, and the adjustment unit is used to adjust the flow load release amount;

[0117] Through the collaborative work of multiple units, refined management and real-time regulation of traffic are achieved, and the response speed and accuracy of the system are improved. The long-term traffic information provided by the steady-state unit is used as a reference to help the system maintain stable operation when facing sudden changes in traffic, reducing the risk of service interruption caused by instantaneous traffic peaks. The adjustment unit dynamically adjusts the traffic bandwidth and packet loss rate according to actual needs, optimizes the use efficiency of network resources, and reduces operating costs. The predictive ability of the transient unit allows the system to be prepared before traffic changes, thereby managing traffic more actively rather than passively reacting. The design of the control module allows the system to self-adjust according to different traffic patterns, enhancing its adaptability to various application scenarios;

[0118] A feedback module is used to feed back flow information, flow rate, flow transmission link status, and flow buffer status, and to control the flow transmission pipeline valve;

[0119] Choose proportional control valves that can support continuous modulation. These valves can adjust the opening in proportion to the input signal to accurately control the flow rate.

[0120] The introduction of a proportional-integral-derivative (PID) controller calculates the appropriate valve opening adjustment based on the difference between the set point (target flow value) and the actual measured value. The P part of the PID controller directly corresponds to proportional flow control, which adjusts the output based on the proportion of the error.

[0121] The flow delivery pipeline valve is a proportional control valve.

[0122] Embodiment 2: like Figure 2 As shown, a proportional flow control device comprises:

[0123] A proportional flow control device comprises a base portion 1, a liquid level tank 2 and a delivery pipeline 3;

[0124] One end of the delivery pipeline 3 is connected to the liquid input device, and the other end is connected to the liquid level tank 2;

[0125] The liquid level tank 2 is located at the top of the base part 1;

[0126] like Figure 3The base portion 1 shown is used to support the liquid level tank 2, and the base portion 1 includes a base bracket 11 and a base support leg 12;

[0127] like Figure 6 As shown, the delivery pipeline 3 includes a delivery pipe 31, a first manual ball valve 32, a proportional control valve 33, a second manual ball valve 34, a pressure regulating valve 35, a third manual ball valve 36 and a signal line pipe 37;

[0128] One end of the first manual ball valve 32 is connected to the delivery pipe 31, and the other end is connected to the proportional regulating valve 33;

[0129] The other end of the proportional control valve 33 is connected to the second manual ball valve 34, and one side of the proportional control valve is connected to the signal line pipe 37, and the pressure regulating valve 35 and the proportional control valve 33 are connected through the signal line pipe 37;

[0130] The other end of the proportional regulating valve 33 is connected to the second manual ball valve 34 , and the other end of the second manual ball valve 34 is connected to the pressure regulating valve 35 ;

[0131] The other end of the pressure regulating valve 35 is connected to the third manual ball valve 36;

[0132] A signal line pipe 37 is provided on one side of the pressure regulating valve 35, and the pressure regulating valve 35 is connected to the proportional regulating valve 33 through the signal line pipe 37;

[0133] like Figure 4 As shown, the liquid level tank 2 includes a tank body 22, and a plurality of spring legs 21 are arranged on the side wall of the tank body 22, and the distances between the plurality of spring legs 21 are equal and symmetrical, so as to support and fix the liquid level tank 2 on the base part 1;

[0134] A liquid level gauge 23 is provided on the side wall of the tank body 22, and the liquid level gauge 23 is perpendicular to the horizontal ground and is used to measure the liquid level in the liquid level tank 2;

[0135] A drain port 24 is provided at one end of the tank body 22 for draining the liquid inside the tank body 22;

[0136] like Figure 5 As shown, the top of the liquid level tank 2 is respectively provided with a fourth spare port 25, a reflux port 26, a pure water port 27, a nitrogen port 28, a first spare port 29, a second spare port 210 and a third spare port 211;

[0137] A proportional flow control device is specifically implemented in production, and actual production parameters, for example: a proportional flow control device may include an air source switch device, a pressure adjustment device, an air source switch device, a proportional flow adjustment device, an air source switch device, a liquid level display device, and a liquid level sensor signal output device; a first air source switch device is provided on the fixed wire pipe; a pressure adjustment device is placed on the air source switch device, and an air source switch device is placed at the rear end of the pressure adjustment device; a proportional flow adjustment device is placed at the rear end of the air source switch device; an air source switch device is placed at the rear end of the proportional flow adjustment device; a liquid level display device is placed at the rear end of the air source switch device; and a liquid level sensor signal output device is placed at the rear end of the liquid level display device. The liquid passes through the gas source switch device, which is normally open. The pressure adjustment device at the rear end realizes pressure adjustment (when the proportional flow control device adjusts the liquid flow, it is necessary to adjust the pressure through the proportional flow control device). The gas source switch device at the rear end is normally open (when the pressure adjustment device is damaged or needs to be repaired, the gas source switch device is closed for disassembly or repair). The proportional flow control device at the rear end is normally open (when the liquid level sensor signal output device outputs a signal to the proportional adjustment device, the proportional adjustment device adjusts the flow to control the size of the flow). The gas source switch device at the rear end (when the proportional flow control device is damaged or needs to be repaired, the gas source switch device is closed for disassembly or repair). The liquid level display device at the rear end displays the liquid capacity in the storage tank. The liquid level sensor signal output device at the upper end (when the liquid level display device displays that the liquid capacity in the storage tank reaches a certain level, it is linked with the liquid level sensor signal output device, and the signal is output to the proportional flow control device to control the flow). The device can accurately control the flow size, and ensure that the transportation process can be safely supplied to the equipment end under the premise of controlling the flow size.

[0138] It is important to note that the construction and arrangement of the present application shown in a plurality of different exemplary embodiments are only exemplary. Although only two embodiments are described in detail in this disclosure, it should be readily understood by those who refer to this disclosure that many modifications are possible, for example, the size, scale, structure, shape and ratio of various elements, and parameter values ​​(e.g., temperature, pressure, etc.), installation arrangement, use of materials, color, directional changes, etc., without substantially departing from the novel teachings and advantages of the subject matter described in the application. For example, the element shown as integrally formed can be composed of multiple parts or elements, the position of the element can be inverted or otherwise changed, and the nature or number or position of the discrete element can be changed or changed. Therefore, all such modifications are intended to be included in the scope of the present invention. The order or sequence of any process or method steps can be changed or reordered according to alternative embodiments. Any "device plus function" clause is intended to cover the structure of the execution function described in this article, and is not only structurally equivalent but also equivalent structure. Without departing from the scope of the present invention, other replacements, modifications, changes and omissions can be made in the design, operating conditions and arrangement of the exemplary embodiment. Therefore, the invention is not limited to a specific embodiment, but extends to several modifications still falling within the scope of the appended claims.

[0139] Additionally, in order to provide a concise description of exemplary embodiments, all features of an actual embodiment may not be described (ie, those features that are not relevant to the best mode presently contemplated for carrying out the invention or those features that are not relevant to implementing the invention).

[0140] It should be understood that in the development of any actual implementation, as in any engineering or design project, numerous implementation-specific decisions may be made. Such a development effort may be complex and time-consuming, but for those of ordinary skill having the benefit of this disclosure, the development effort will be a routine task of design, fabrication, and production without undue experimentation.

[0141] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A proportional flow control system, comprising: The data module includes a data collection unit for collecting flow information, a data processing unit for preprocessing the collected flow information, and a feature extraction unit for extracting feature vectors of the processed flow information; A processing module, a flow prediction model for monitoring and predicting flow information characteristics, and a performance evaluation unit for evaluating performance indicators of the flow prediction model; A control module, a transient unit for reflecting instantaneous changes in the traffic queue, a steady-state unit for reflecting long-term changes in the traffic queue, including a determination unit for determining traffic information, an adjustment unit for adjusting the traffic bandwidth packet loss rate, and a monitoring unit for monitoring traffic buffer queue information; The feedback module is used to feed back flow information, flow rate, flow transmission link status and flow buffer status, and control the flow transmission pipeline valve.

2. A proportional flow control system according to claim 1, characterized in that: The preprocessing includes filtering, denoising, calibration and normalization; The feature extraction is used to extract features from the pre-processed flow information.

3. The proportional flow control system according to claim 1, characterized in that: Taking the historical traffic data of the past n time periods as basic data, the basic data is divided into a training set and a test set, wherein the basic data includes traffic transmission links, link numbers and traffic values, and the traffic prediction model initialized by training the training set is used to map all traffic transmission links through the trained traffic prediction model; The basic data obtains m hidden traffic information through the traffic integration model, and the m hidden traffic information is input into the traffic detection model. The traffic detection model sets a multi-layer traffic perception layer, and calculates the m hidden information weights of multiple traffic transmission links through the multi-layer traffic perception layer. The traffic detection model predicts the traffic change information vector of each traffic transmission link by iterating the traffic changes of multiple traffic transmission links, and finally integrates them into the global traffic change to output the traffic prediction value.

4. A proportional flow control system according to claim 3, characterized in that: Taking the characteristic vector of the traffic as input, obtaining the hidden information of the traffic through the traffic integration model; The flow integration model includes flow information substitution coefficients and flow information writing values; The flow information substitution coefficient is used to control the degree coefficient of the flow information of the previous timestamp being substituted into the flow information at the current moment, and the calculation expression is as follows: ； in, is the degree coefficient of the flow information of the previous timestamp being substituted into the flow information of the current moment, is the activation function, Write the coefficient weight coefficient for the flow information, Enter the flow information for time t, is the traffic transmission link at time t, is the flow information of the hth flow transmission link at time t, Substitute the coefficient bias value for the flow information; The flow information write value is used to control the value of the flow information at the previous moment written into the current flow information, further capturing the dependency of the flow time series on the flow at the current moment. The calculation expression is as follows: ； in, is the coefficient value of the flow information at the previous moment being written into the current flow information, Write value weight coefficient for traffic information, is the flow information of the hth flow transmission link at time t, The offset value for writing the flow information; The flow detection model calculation expression is as follows: ； in, is the weight coefficient of the i-th traffic perception layer, is the weight coefficient of the first layer traffic perception layer, is the weight coefficient of the kth traffic perception layer, Input traffic information to the traffic perception layer at the first moment, Enter the flow information for the t-1th moment, is the bias value of the first layer traffic perception layer, is the bias value of the i-th traffic perception layer, is the bias value of the kth traffic perception layer, e is the exponential constant, n is the current traffic perception layer calculation, k is the total number of traffic perception layers, The model operator indicates that the traffic detection model is running to calculate the traffic information.

5. A proportional flow control system according to claim 4, characterized in that: The performance evaluation unit evaluates the prediction capability of the flow detection model by calculating the accuracy of the flow information prediction value output by the flow detection model. The calculation expression is as follows: ； in, The accuracy of the traffic information prediction value output by the traffic detection model, is the real-time flow value, Output the traffic information prediction value for the traffic detection model, t is the time, T is the total time of traffic calculation; The accuracy of the traffic information prediction value output by the traffic detection model ranges from 0 to 1. The greater the accuracy, the more accurate the traffic information prediction value output by the traffic detection model is.

6. A proportional flow control system according to claim 5, characterized in that: The transient unit obtains the arrival rate of the traffic at the next moment by receiving the traffic information change rate of the traffic transmission link at the current moment and the traffic information prediction value output by the traffic detection model, and pre-calculates the change rate of the traffic transmitted by the traffic transmission link at the next moment. The higher the arrival rate, the faster the traffic transmission speed, and the lower the arrival rate, the slower the traffic transmission speed. If the arrival rate is 0, the traffic transmission speed is moderate. The steady-state unit is used to obtain average flow information of the flow transmission link.

7. A proportional flow control system according to claim 6, characterized in that: The determination unit receives the traffic information prediction value output by the traffic detection model, determines the traffic information of the traffic transmission link, and introduces a small amount of traffic into the traffic buffer queue if the traffic information of the traffic transmission link is less than the average traffic information and the traffic information tends to increase with traffic bursts; If the flow information of the traffic delivery link is less than the average flow information and the flow information decreases, it is a flow attenuation suppression state, and all flows are allowed to enter the flow buffer queue; If the flow information of the traffic transport link is greater than the average flow information and the amount of flow information increases, the control flow enters the flow buffer queue; The adjusting unit receives the judgment result of the flow transmission link flow information output by the judgment unit and the arrival rate output by the transient unit. If the flow load is greater than the maximum threshold, the flow load is released, that is, all flow data packets are discarded. If the flow load is not greater than the maximum threshold, the flow data packets are discarded in a dispersed manner. The flow data packet is used to distribute and adjust the flow transmission link through the transmission pipeline and the automatic control ball valve; The monitoring unit detects the flow buffer by monitoring the flow arrival rate and flow information of the flow buffer.

8. A proportional flow control device, which is implemented by a proportional flow control system as claimed in any one of claims 1 to 7, characterized in that: The device comprises: A proportional flow control device comprises a base portion (1), a liquid level tank (2) and a delivery pipeline (3); One end of the delivery pipeline (3) is connected to the liquid input device, and the other end is connected to the liquid level tank (2); The liquid level tank (2) is located at the top of the base part (1); The base portion (1) is used to support the liquid level tank (2).

9. A proportional flow control device according to claim 8, characterized in that: The base part (1) comprises a base bracket (11) and a base support leg (12); The delivery pipeline (3) comprises a delivery pipe (31), a first manual ball valve (32), a proportional control valve (33), a second manual ball valve (34), a pressure regulating valve (35), a third manual ball valve (36) and a signal line pipe (37); One end of the first manual ball valve (32) is connected to the delivery pipe (31), and the other end is connected to the proportional control valve (33); The other end of the proportional control valve (33) is connected to a second manual ball valve (34), and one side of the proportional control valve is connected to a signal line pipe (37), and the pressure regulating valve (35) and the proportional control valve (33) are connected via the signal line pipe (37); The other end of the proportional regulating valve (33) is connected to a second manual ball valve (34), and the other end of the second manual ball valve (34) is connected to a pressure regulating valve (35); The other end of the pressure regulating valve (35) is connected to a third manual ball valve (36); A signal line pipe (37) is provided on one side of the pressure regulating valve (35), and the pressure regulating valve (35) is connected to the proportional control valve (33) via the signal line pipe (37).

10. The proportional flow control device according to claim 9, characterized in that: The liquid level tank (2) comprises a tank body (22), and a plurality of spring legs (21) are arranged on the side wall of the tank body (22), and the plurality of spring legs (21) are spaced at equal distances and maintained symmetrically, for supporting and fixing the liquid level tank (2) on the base portion (1); A liquid level gauge (23) is arranged on the side wall of the tank body (22), wherein the liquid level gauge (23) is perpendicular to the horizontal ground and is used to measure the liquid level in the liquid level tank (2); A liquid discharge port (24) is provided at one end of the tank body (22) for discharging liquid inside the tank body (22).

11. A proportional flow control device according to claim 10, characterized in that: The top of the liquid level tank (2) is respectively provided with a fourth spare port (25), a reflux port (26), a pure water port (27), a nitrogen port (28), a first spare port (29), a second spare port (210), and a third spare port (211).

Images