An automatic sorting and sequential start-up control method for multiple sets of air ducts operating in disorder.

Abstract

This invention discloses an automatic sorting and sequential start-up control method for multiple sets of ventilation ducts operating in a disordered manner, relating to the field of control technology in fire protection and other special equipment industries. The method includes the following steps: collecting ventilation duct start-up sequence data; starting the first set of ventilation ducts based on the start-up sequence data; recording the actual start-up time of the ventilation ducts; monitoring the operating status of the ventilation ducts by setting sensors; and generating ventilation duct status data. This method compares and analyzes the collected ventilation duct operating status data with the historical operating status of ventilation ducts under various environments to determine the environment in which the ventilation ducts start, and adjusts the upper limit of the ventilation duct start-up time limit accordingly. This prevents ventilation ducts from failing to start if they require a long start-up time under headwinds or heavy loads, and analyzes whether the ventilation duct has completed start-up. Upon completion, the method automatically starts the next ventilation duct, achieving dynamic control of the ventilation duct start-up time interval and improving the efficiency of ventilation duct start-up.

Description

Technical Field

[0001] This invention relates to the field of control technology in the fire protection and other special equipment industries, specifically to an automatic sorting and sequential start-up control method for the disordered operation of multiple ventilation ducts. Background Technology

[0002] A ventilation duct is a structure or function that can be considered as a single unit. A fan is the smallest working motor unit in smoke exhaust / air supply or ventilation operation. Electrically driven smoke exhaust fire trucks, especially high-flow-rate smoke exhaust trucks, have special requirements for the flow rate and function of the supply and exhaust air. Multiple ventilation duct units are often installed, with multiple fans arranged in each unit. The operation of each ventilation duct unit is independently controllable. To reduce the impact on the power system caused by direct motor starting, a time-sharing and sequential starting method is often adopted to start them one by one. Ventilation ducts and fans may have different names and definitions in different applications, but they all belong to the aforementioned individually controllable assemblies and the smallest controllable supply and exhaust air units within the assemblies.

[0003] Currently, the operation control of multi-duct smoke exhaust vehicles in China is mostly based on unit operation, with the control panel having multiple corresponding operation buttons. However, when operating independently, it is difficult to ensure that the previous duct unit has finished starting before starting the next group of duct units. This means that the starting cannot be completely controlled according to time-sharing and sequence. As a result, the operator needs to observe and confirm for a long time before operating the next group of duct units. Operating multiple groups of duct units requires a lot of time and effort. When operating multiple groups of fans, if it is necessary to stop the previously working duct, it may affect the normal start-up of the fan that is currently starting. This conventional control method is very unfriendly to operators and greatly reduces operating efficiency.

[0004] In summary, the shortcomings of commonly used control schemes are as follows:

[0005] 1. When starting the ventilation unit, it cannot be operated in one go. It is necessary to observe whether the previous unit has finished starting before operating it, which reduces the operating efficiency.

[0006] 2. The system is highly dependent on the operator's operation time intervals and operation sequence;

[0007] 3. Cannot be operated randomly or arbitrarily. Summary of the Invention

[0008] The purpose of this invention is to provide an automatic sorting and sequential start-up control method for multiple sets of air ducts operating in a disordered manner, so as to solve at least one of the above-mentioned deficiencies in the prior art.

[0009] To achieve the above objectives, the present invention provides the following technical solution: an automatic sorting and sequential start-up control method for multiple sets of air ducts operating in a disordered manner, comprising the following steps:

[0010] S1. Collect the ventilation duct start-up sequence data, and start the first group of ventilation ducts according to the ventilation duct start-up sequence data;

[0011] S2. Record the actual start-up time of the ventilation duct and monitor the operating status of the ventilation duct by setting up sensors to generate ventilation duct status data;

[0012] S3. Based on the duct status data and the duct status reference feature vector of the start-up environment, analyze and process the environment when the duct is started to generate duct start-up environment analysis data. The duct start-up environment analysis data includes the duct initial load factor data and the corresponding set environment label.

[0013] S4. Based on the analysis data of the ventilation duct startup environment and the setting of the environment and startup time limit mapping rules, dynamically adjust the upper and lower limits of the ventilation duct startup time to generate the ventilation duct startup time range.

[0014] S5. Based on the duct status data and the feature vector indicating whether the duct has been started and completed, perform similarity analysis to determine whether the duct has been started and completed.

[0015] S6. If yes, then perform a start operation on the next air duct in the air duct start sequence data and return to step S2;

[0016] S7. If not, determine whether the actual start-up time of the ventilation duct is within the specified start-up time range.

[0017] S8. If yes, continue to wait and repeatedly verify the actual start-up time of the ventilation duct; if no, determine that the ventilation duct has failed to start abnormally, and analyze and process the abnormal type of the ventilation duct start-up failure based on the ventilation duct status data and the set ventilation duct abnormal type feature vector to generate ventilation duct start-up failure analysis data. Then, perform the start-up operation on the next ventilation duct in the ventilation duct start-up sequence data and return to step S2.

[0018] Furthermore, S1 includes the following steps:

[0019] S1.1 Generate ventilation duct start sequence data based on the triggering sequence of the operator's trigger switch or the selected preset ventilation duct start sequence;

[0020] S1.2 Execute the start operation on the first group of the air duct start sequence data.

[0021] Furthermore, S2 includes the following steps:

[0022] S2.1. When the ventilation duct is started, the start-up time of the ventilation duct is timed to obtain the actual start-up time of the ventilation duct;

[0023] S2.2. By setting up sensors, the three-phase current, motor speed, outlet air pressure, motor temperature and vibration information of the duct are collected in real time to generate duct status data. Among them, the three-phase current can be monitored by current transformer or Hall sensor, the motor speed can be monitored by speed sensor or Hall sensor, the outlet air pressure can be collected by pressure sensor, the motor temperature can be monitored by temperature sensor, and the vibration information can be monitored by accelerometer (such as triaxial accelerometer).

[0024] Furthermore, S3 includes the following steps:

[0025] S3.1. Based on the duct status data, calculate and analyze the average current rise slope and rotation speed rise slope of the duct before the set time (e.g., 2-3 seconds after startup) after the duct starts, and generate the initial current rise slope data and the initial rotation speed rise slope data of the duct respectively.

[0026] S3.2. Collect the average current rise slope and speed rise slope of the duct under normal conditions after starting in a normal environment, and generate the standard current rise slope data and the standard speed rise slope data of the duct in the early stage respectively.

[0027] S3.3. Based on the initial current rise slope data, initial rotation speed rise slope data, initial current rise standard slope data, and initial rotation speed rise standard slope data of the duct, the load factor of the duct is calculated and processed to generate the initial load factor data of the duct.

[0028] S3.4 Collect the load factor range of the duct corresponding to each set environmental label;

[0029] S3.5 Determine the environmental label corresponding to the wind duct load factor interval where the initial load factor data of the wind duct is located, and output the initial load factor data of the wind duct and the corresponding environmental label to generate wind duct start-up environmental analysis data.

[0030] Furthermore, S4 includes the following steps:

[0031] S4.1 Input the initial load factor data of the ventilation duct into the setting maximum start-up time mapping function, so that the output increases as the initial load factor data of the ventilation duct increases within the set range after the initial load factor data of the ventilation duct exceeds the set standard load factor, and outputs the upper limit of the ventilation duct start-up time. For example, the following formula can be used as a reference:

[0032]

[0033] in, The initial value of the upper limit of the ventilation duct start-up time is given by k, where M is the control coefficient for the upper limit of the ventilation duct start-up time, used to control the value range of the upper limit of the ventilation duct start-up time. M is greater than or equal to 1, and k is... The control coefficients of the function are used to control its steepness; LF represents the initial load factor data of the duct.

[0034] Furthermore, k can decrease as LF increases, and can be used to calculate the initial growth rate of LF, for example... , ;

[0035] S4.2. Based on the lower limit of the set duct start-up time and the upper limit of the output duct start-up time, the duct start-up time range is obtained.

[0036] Furthermore, S5 includes the following steps:

[0037] S5.1 Based on the current duct status data and set feature standard data within the sliding window, calculate various characteristics of the duct during operation and generate a duct status feature vector. The following features can be selected for calculation as needed, such as steady-state current, current fluctuation rate, three-phase current imbalance, the proportion of duct speed reaching the set rated speed, duct speed fluctuation rate, the proportion of duct wind pressure reaching the set rated wind pressure, and vibration effective value, etc.

[0038] S5.2. Based on the historical duct state feature vector, mark the duct state feature vector when the start-up is completed, and obtain the standard vector of duct state features;

[0039] S5.3. Normalize each term of the duct state feature vector and the duct state feature standard vector to generate the duct state feature normalized vector and the duct state feature standard normalized vector, respectively.

[0040] S5.4. Based on the cosine similarity algorithm, the normalized vector of the duct state features and the standard normalized vector of the duct state features are processed to calculate the similarity and generate the duct state start-up completion similarity data.

[0041] S5.5 Determine whether the similarity data of the duct status start-up completion is greater than the set duct status start-up completion similarity threshold. If yes, the duct start-up is complete; otherwise, the duct start-up is not complete.

[0042] Furthermore, S5.4 can be performed through the following steps:

[0043] S5.4a. Collect and set the weights of each item in the duct state feature vector, and generate duct state feature weight data;

[0044] S5.4b: Based on the weighted data of the duct state features, perform weighted cosine similarity calculation on the normalized vector of the duct state features and the standard normalized vector of the duct state features to generate similarity data of duct state start-up and completion.

[0045] Furthermore, the generation of duct start-up failure analysis data in S8 includes the following steps:

[0046] Set the normalized vector of the duct state features when the duct starts up abnormally as the normalized vector of the duct abnormal state features.

[0047] Based on the PCA (Principal Component Analysis) algorithm, the normalized vectors of all abnormal state features of the ventilation ducts in the historical ventilation duct start-up anomalies are reduced in dimensionality and then HDBSCAN cluster analysis is performed to generate multiple data clusters of ventilation duct start-up anomaly types.

[0048] Each duct start-up anomaly type data cluster is associated with a set duct start-up anomaly type label; when associating, the confidence level of the duct start-up anomaly type data cluster and the corresponding duct start-up anomaly type label can also be calculated, so as to provide a prompt when the duct anomaly state feature normalization vector corresponds to the duct start-up anomaly type data cluster with low confidence.

[0049] The average value of each term in the normalized vector of all abnormal state features of each ventilation duct in each data cluster of abnormal start-up type is calculated to obtain the average vector of the data cluster.

[0050] The cosine similarity (or weighted cosine similarity) between the normalized vector of the duct abnormal state features and the average vector of each data cluster is calculated. The maximum cosine similarity is extracted and checked against a set cosine similarity threshold. If the threshold is met, the duct start-up abnormality type label corresponding to the data cluster with the maximum cosine similarity is output; otherwise, "Unknown abnormality occurred" is output. Furthermore, the duct abnormal state feature vectors at the "Unknown abnormality occurred" stage can be collected. After a certain number are collected, new duct start-up abnormality type labels are added as needed, updating the corresponding duct start-up abnormality type data clusters, data cluster average vectors, and duct start-up abnormality type labels.

[0051] Furthermore, the method also includes the following steps:

[0052] Operators can remove the duct from the duct start-up sequence data at any time by triggering the duct's trigger switch again. If the duct has already been started, a shutdown operation can be performed on it.

[0053] Beneficial effects:

[0054] Compared with the prior art, the present invention provides an automatic sorting and sequential start-up control method for the disordered operation of multiple sets of ventilation ducts. By collecting the operating status data of the ventilation ducts and comparing and analyzing the operating status of the ventilation ducts in various environments in the past, the environment in which the ventilation ducts are started is determined, and the upper limit of the start-up time limit of the ventilation ducts is adjusted accordingly to prevent the ventilation ducts from being considered as starting failures if they take too long to start under headwinds or heavy loads.

[0055] Compared with the prior art, the present invention provides an automatic sorting and sequential start-up control method for multiple sets of air ducts operating in disorder. By comparing the air duct operation status data with the operation status data when the start-up is completed, it analyzes whether the air duct has been started. When the start-up is completed, it automatically starts the next air duct, realizing dynamic control of the air duct start-up time interval and improving the start-up efficiency of the air duct.

[0056] Compared with the prior art, the present invention provides an automatic sorting and sequential start-up control method for multiple groups of air ducts operating in disorder. By outputting air duct start-up failure analysis data when the air duct fails to start, the method analyzes the cause of the start-up failure, so as to facilitate the operator to decide on the subsequent strategy for the air duct that failed to start based on the cause of the start-up failure. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0058] Figure 1 This is a flowchart illustrating the method steps provided in an embodiment of the present invention;

[0059] Figure 2 Detailed step diagrams of S3 and S4 provided in the embodiments of the present invention;

[0060] Figure 3 This is a schematic diagram showing the numbering of the air duct and fan provided in an embodiment of the present invention;

[0061] Figure 4 A block diagram of the air duct sorting queue provided in an embodiment of the present invention;

[0062] Figure 5 A diagram illustrating the process of falling behind, provided for an embodiment of the present invention;

[0063] Figure 6 This is a diagram illustrating the phenomenon of falling behind when m>1, provided in an embodiment of the present invention.

[0064] Figure 7 This is a diagram illustrating the phenomenon of falling behind when m=1, provided in an embodiment of the present invention.

[0065] Figure 8 This is a schematic diagram of indentation when m>1, provided in an embodiment of the present invention.

[0066] Figure 9 This is a schematic diagram of indentation when m=1 provided in an embodiment of the present invention;

[0067] Figure 10 A diagram illustrating the wind turbine drive function provided in an embodiment of the present invention. Detailed Implementation

[0068] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0069] Exemplary embodiments will be described more fully below with reference to the accompanying drawings; however, these exemplary embodiments may be embodied in different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will enable those skilled in the art to fully understand the scope of this disclosure.

[0070] Where there is no conflict, the various embodiments of this disclosure and the features thereof in the embodiments may be combined with each other.

[0071] As used herein, the term “and / or” includes any and all combinations of one or more related enumerated entries.

[0072] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. As used herein, the singular forms “a” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that when the terms “comprising” and / or “made of” are used in this specification, the presence of a feature, integral, step, operation, element, and / or component is specified, but the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof is not excluded.

[0073] The embodiments described herein can be illustrated with reference to plan views and / or cross-sectional views, taking into account the ideal schematic diagrams of this disclosure. Therefore, the exemplary illustrations can be modified according to manufacturing techniques and / or tolerances. Thus, the embodiments are not limited to those shown in the drawings, but include modifications to configurations formed based on manufacturing processes. Therefore, the areas illustrated in the drawings are schematic in nature, and the shapes of the areas shown illustrate specific shapes of areas of an element, but are not intended to be limiting.

[0074] Unless otherwise specified, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art.

[0075] Please see Figures 1-2An automatic sorting and sequential start-up control method for multiple sets of air ducts operating in an unordered manner includes the following steps:

[0076] S1. Collect the ventilation duct start-up sequence data, and start the first group of ventilation ducts according to the ventilation duct start-up sequence data, including the following steps:

[0077] S1.1 Generate ventilation duct start sequence data based on the triggering sequence of the operator's trigger switch or the selected preset ventilation duct start sequence;

[0078] S1.2 Execute the start operation on the first group of air ducts in the air duct start sequence data.

[0079] S2. Record the actual start-up time of the ventilation duct, and monitor the operating status of the ventilation duct by setting sensors to generate ventilation duct status data, including the following steps:

[0080] S2.1. When the ventilation duct is started, the start-up time of the ventilation duct is timed to obtain the actual start-up time of the ventilation duct;

[0081] S2.2. By setting up sensors, the three-phase current, motor speed, outlet air pressure, motor temperature and vibration information of the duct are collected in real time to generate duct status data. Among them, the three-phase current can be monitored by current transformer or Hall sensor, the motor speed can be monitored by speed sensor or Hall sensor, the outlet air pressure can be collected by pressure sensor, the motor temperature can be monitored by temperature sensor, and the vibration information can be monitored by accelerometer (such as triaxial accelerometer).

[0082] S3. Based on the duct status data and the set startup environment duct status reference feature vector, analyze and process the environment during duct startup to generate duct startup environment analysis data. The duct startup environment analysis data includes duct initial load factor data and corresponding set environment labels (such as headwind / heavy load, normal, tailwind / light load, etc.), including the following steps:

[0083] S3.1. Based on the duct status data, calculate and analyze the average current rise slope and speed rise slope of the duct before the set time (e.g., 2-3 seconds after startup) after the duct starts, and generate the initial current rise slope data and the initial speed rise slope data of the duct respectively.

[0084] S3.2. Collect the average current rise slope and speed rise slope of the duct under normal conditions after starting in a normal environment, and generate the standard current rise slope data and the standard speed rise slope data of the duct in the early stage respectively.

[0085] S3.3. Based on the initial current rise slope data, initial rotation speed rise slope data, initial current rise standard slope data, and initial rotation speed rise standard slope data of the ventilation duct, the load factor of the ventilation duct is calculated and processed to generate the initial load factor data of the ventilation duct.

[0086] S3.4 Collect the load factor range of the duct corresponding to each set environmental label;

[0087] S3.5 Determine the environmental label corresponding to the wind tunnel load factor interval where the initial load factor data of the wind tunnel is located, and output the initial load factor data of the wind tunnel and the corresponding environmental label to generate the wind tunnel start-up environmental analysis data.

[0088] S4. Based on the analysis data of the ventilation duct startup environment and the setting of the environment-startup time limit mapping rules, dynamically adjust the upper and lower limits of the ventilation duct startup time to generate the ventilation duct startup time range, including the following steps:

[0089] S4.1 Input the initial load factor data of the ventilation duct into the setting maximum start-up time mapping function, so that the output increases as the initial load factor data of the ventilation duct increases within the set range after the initial load factor data of the ventilation duct exceeds the set standard load factor. The output is the upper limit of the ventilation duct start-up time. For example, you can refer to the following formula:

[0090]

[0091] in, The initial value of the upper limit of the ventilation duct start-up time is given by k, where M is the control coefficient for the upper limit of the ventilation duct start-up time, used to control the value range of the upper limit of the ventilation duct start-up time. M is greater than or equal to 1, and k is... The control coefficients of the function are used to control its steepness; LF represents the initial load factor data of the duct.

[0092] Furthermore, k can decrease as LF increases, and can be used to calculate the initial growth rate of LF, for example... , ;

[0093] S4.2. Based on the lower limit of the set duct start-up time and the upper limit of the output duct start-up time, the duct start-up time range is obtained.

[0094] S5. Based on the duct status data and the feature vector indicating whether the duct has completed startup, perform similarity analysis to determine whether the duct has finished starting up. This includes the following steps:

[0095] S5.1 Based on the current duct status data and set feature standard data within the sliding window, calculate various characteristics of the duct during operation and generate a duct status feature vector. The following features can be selected for calculation as needed, such as steady-state current, current fluctuation rate, three-phase current imbalance, the proportion of duct speed reaching the set rated speed, duct speed fluctuation rate, the proportion of duct wind pressure reaching the set rated wind pressure, and vibration effective value, etc.

[0096] S5.2. Based on the historical duct state feature vector, mark the duct state feature vector when the start-up is completed, and obtain the standard vector of duct state features;

[0097] S5.3 Normalize each term of the duct state feature vector and the duct state feature standard vector to generate the duct state feature normalized vector and the duct state feature standard normalized vector, respectively.

[0098] S5.4. Based on the cosine similarity algorithm, the normalized vector of the duct state features and the standard normalized vector of the duct state features are processed to calculate the similarity and generate the duct state start-up and completion similarity data.

[0099] In one embodiment, S5.4 can be replaced by the following steps:

[0100] S5.4a. Collect and set the weights of each item in the duct state feature vector, and generate duct state feature weight data;

[0101] S5.4b: Based on the weighted data of the duct state features, the normalized vector of the duct state features and the standard normalized vector of the duct state features are processed by weighted cosine similarity calculation to generate similarity data of duct state start-up and completion.

[0102] S5.5 Determine whether the similarity data of the duct status startup completion is greater than the set duct status startup completion similarity threshold. If yes, the duct startup is complete; otherwise, the duct startup is incomplete.

[0103] If the duct has not been started, the remaining start time of the duct can be roughly estimated by subtracting the current actual start time of the duct from the upper limit of the start time, or the start-up progress of the fan can be reflected by calculating the ratio of the current actual start time of the duct to the upper limit of the start time.

[0104] S6. If yes, then perform the start operation on the next air duct in the air duct start sequence data and return to step S2;

[0105] S7. If not, determine whether the actual start-up time of the ventilation duct is within the start-up time range of the ventilation duct;

[0106] S8. If yes, continue to wait and repeatedly verify the actual start-up time of the ventilation duct; if no, determine that the ventilation duct has failed to start abnormally, and analyze and process the abnormal type of the ventilation duct start-up failure based on the ventilation duct status data and the set ventilation duct abnormal type feature vector, generate ventilation duct start-up failure analysis data, and then perform the start-up operation on the next ventilation duct in the ventilation duct start-up sequence data, and return to step S2.

[0107] The process of generating analysis data for air duct startup failure includes the following steps:

[0108] (1) Set the normalized vector of the duct state features when the duct starts abnormally as the normalized vector of the duct abnormal state features;

[0109] (2) Based on the PCA (principal component analysis) algorithm, the normalized vectors of all duct abnormal state features in the historical duct start-up anomalies are reduced in dimension and then HDBSCAN cluster analysis is performed to generate multiple duct start-up anomaly type data clusters.

[0110] (3) Associate each duct start-up anomaly type data cluster with a set duct start-up anomaly type label; when associating, the confidence level of the duct start-up anomaly type data cluster and the corresponding duct start-up anomaly type label can also be calculated so as to provide a prompt when the duct anomaly state feature normalization vector corresponds to the duct start-up anomaly type data cluster with low confidence.

[0111] (4) Calculate the average value of each term of the normalized vector of all abnormal state features of each duct in the data cluster of each duct start-up abnormality type to obtain the average vector of the data cluster;

[0112] (5) Calculate the cosine similarity (or weighted cosine similarity) between the normalized vector of the duct abnormal state features and the average vector of each data cluster. Extract the maximum cosine similarity and determine whether it is greater than the set cosine similarity threshold. If yes, output the duct start-up abnormal type label corresponding to the data cluster with the maximum cosine similarity. If no, output "Unknown abnormality occurred". Further, the duct abnormal state feature vectors when "unknown abnormality occurred" can be collected. After reaching a certain number, add duct start-up abnormal type labels as needed, and update the corresponding duct start-up abnormal type data clusters, data cluster average vectors, and duct start-up abnormal type labels.

[0113] In one embodiment, reference Figure 3No[i] is the unique unit number of the ventilation duct, i∈N1; it is a code assigned to the ventilation duct according to a certain order or rule for easy identification. The code is not limited to numbering but can also be named. Regardless of the identification method used, it is considered the same method of identifying ventilation duct units; M[x] is the fan number within the ventilation duct unit, x∈N1, which is a code assigned to the fan within the ventilation duct according to a certain order or rule. The fan number is unique within the same ventilation duct unit. (Reference) Figure 4 K is the trigger sequence number of the ventilation duct, k∈N1, which is the sequence count of the ventilation ducts when they are started. When the kth ventilation duct is triggered to work, the number of the triggered ventilation duct No[i] is assigned to the variable Vk, that is, Vk= No[i], and Vk is the number of the kth triggered ventilation duct No[i]; at the same time, Vk is put into the Px[k] queue.

[0114] In one embodiment, the operator can remove the duct from the duct start-up sequence data at any time by triggering the duct's trigger switch again. If the duct has already started, a shutdown operation is performed on it. The deleted duct is removed from the duct start-up sequence data. For example, if the m-th triggered duct changes its working state to a stopped state, 1 ≦ m ≦ k, m ∈ N1, initially, the controller cannot know the specific value of m and the corresponding device number Px[m], and must identify the value of m. The method for identifying the value of m adopts a strategy of traversing the Px[k] array. Since which device is stopped can be monitored, i.e., the identification number No[i] of the stopped device is determined, when Px[n] = No[i], the corresponding value of n is the position value m of the dropped duct in the start-up queue, i.e., m = n. That is, the duct number where Px[n] or Px[m] is placed is the number No[i] corresponding to the dropped duct. After identifying the value of m, Px[m] is set to 0, and the original working duct number is cleared. After clearing the original equipment number, such as Figures 5-9 As shown; in the Px[k] working sequence, the invalid duct number "0" is meaningless. To avoid the accumulation of "0" in the working sequence, it needs to be removed. It is also necessary to fill the "0" values ​​of the earlier, later-entered ducts with the later-entered values ​​to remove the traces of "0". The method is to traverse Px[k]. When Px[c] = 0 and Px[c+1] ≠ 0, c ∈ N1, Px[c] = Px[c+1]. At the same time, let Px[c+1] = 0 and Q[c] = 1. Finally, the duct start-up sequence is formed, which can guide the next step of the fan start-up work. If it is necessary to count the number of working duct units N, then execute... .

[0115] In one embodiment, the fan inside the ventilation duct is started, and a startup function can be constructed as follows: Figure 10The inputs mainly include the duct number, the number of fans to be controlled, start / stop signals, and start time intervals. The outputs mainly include fan drive signals, fan start-up completion markers, start-up progress, and duct number. The main function is that when the condition is met (step S5 determines yes), a start signal is input, and fan drive signals are output sequentially at the set time intervals to control the fan action. When all fans in a duct unit have started, the fan start-up completion marker is set. The fan start-up progress can be viewed to see the position of the currently started fans.

[0116] In one embodiment, the duct numbers to be started can be sequentially retrieved from the duct start-up sequence, and the fan drive function can be called to drive the fan to run. The strategy is as follows: take the value of Px[1], iterate through No[i], when Px[1] = No[i1], the duct with the number No[i1] is the first duct unit to be started, and the fan drive function can be called to drive the fan in No[i1] to run; similarly, take the value of Px[2], iterate through No[i], when Px[2] = No[i2], and the start-up completion flag of No[i1] is set, the duct with the number No[i2] is the second duct unit to be started, and the fan drive function can be called to drive the fan in No[i2] to run; and so on, take the value of Px[k], iterate through No[i], when Px[k] = When No[ii] is started and the start-up completion flag of No[ii-1] is set, the duct with the number No[ii] is the i-th started duct unit. The fan drive function is called to drive the fan in No[ii] to run. After the start-up completion flag of the fan in the k-th duct No[ii] is set, the duct start-up ends. When a certain duct No[x] is working, due to the indentation function and the above operation, there is no No[x] value in Px[k], that is, Px[k] ≠ No[x], and the fan in duct No[x] stops immediately.

[0117] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. An automatic sorting and sequential start-up control method for multiple sets of air ducts operating in a disordered manner, characterized in that, Includes the following steps: S1. Collect the ventilation duct start-up sequence data, and start the first group of ventilation ducts according to the ventilation duct start-up sequence data; S2. Record the actual start-up time of the ventilation duct and monitor the operating status of the ventilation duct by setting up sensors to generate ventilation duct status data; S3. Based on the duct status data and the duct status reference feature vector of the start-up environment, analyze and process the environment when the duct is started to generate duct start-up environment analysis data. The duct start-up environment analysis data includes the duct initial load factor data and the corresponding set environment label. S4. Based on the analysis data of the ventilation duct startup environment and the setting of the environment and startup time limit mapping rules, dynamically adjust the upper and lower limits of the ventilation duct startup time to generate the ventilation duct startup time range. S5. Based on the duct status data and the feature vector indicating whether the duct has been started and completed, perform similarity analysis to determine whether the duct has been started and completed. S6. If yes, then perform a start operation on the next air duct in the air duct start sequence data and return to step S2; S7. If not, determine whether the actual start-up time of the ventilation duct is within the specified start-up time range. S8. If not, then determine that the ventilation duct has failed to start abnormally, and analyze and process the abnormal type of the ventilation duct start failure based on the ventilation duct status data and the set ventilation duct abnormal type feature vector to generate ventilation duct start failure analysis data. Then, perform the start operation on the next ventilation duct in the ventilation duct start sequence data and return to step S2. S2 includes the following steps: S2.

1. When the ventilation duct is started, the start-up time of the ventilation duct is timed to obtain the actual start-up time of the ventilation duct; S2.

2. By setting up sensors, the three-phase current, motor speed, outlet air pressure, motor temperature and vibration information of the air duct are collected in real time to generate air duct status data; S3 includes the following steps: S3.

1. Based on the duct status data, calculate and analyze the average current rise slope and rotation speed rise slope for the set time before the duct starts, and generate the initial current rise slope data and the initial rotation speed rise slope data of the duct respectively. S3.

2. Collect the average current rise slope and speed rise slope of the duct under normal conditions after starting in a normal environment, and generate the standard current rise slope data and the standard speed rise slope data of the duct in the early stage respectively. S3.

3. Based on the initial current rise slope data, initial rotation speed rise slope data, initial current rise standard slope data, and initial rotation speed rise standard slope data of the duct, the load factor of the duct is calculated and processed to generate the initial load factor data of the duct. S3.4 Collect the load factor range of the duct corresponding to each set environmental label; S3.5 Determine the environmental label corresponding to the wind duct load factor interval where the initial load factor data of the wind duct is located, and output the initial load factor data of the wind duct and the corresponding environmental label to generate wind duct start-up environmental analysis data; S4 includes the following steps: S4.1 Input the initial load factor data of the ventilation duct into the setting maximum start-up time mapping function, so that the output is within the set range and increases as the initial load factor data of the ventilation duct increases after the initial load factor data of the ventilation duct exceeds the set standard load factor, and outputs the upper limit of the start-up time of the ventilation duct. S4.

2. Based on the lower limit of the set duct start-up time and the upper limit of the output duct start-up time, the duct start-up time range is obtained.

2. The automatic sorting and sequential start-up control method for multiple sets of air ducts operating in disorder as described in claim 1, characterized in that, S1 includes the following steps: S1.1 Generate ventilation duct start sequence data based on the triggering sequence of the operator's trigger switch or the selected preset ventilation duct start sequence; S1.2 Execute the start operation on the first group of the air duct start sequence data.

3. The automatic sorting and sequential start-up control method for multiple sets of air ducts operating in disorder as described in claim 1, characterized in that, S5 includes the following steps: S5.1 Based on the current duct status data and set feature standard data within the sliding window, calculate various features of the duct during operation and generate a duct status feature vector. S5.

2. Based on the historical duct state feature vector, mark the duct state feature vector when the start-up is completed, and obtain the standard vector of duct state features; S5.

3. Normalize each term of the duct state feature vector and the duct state feature standard vector to generate the duct state feature normalized vector and the duct state feature standard normalized vector, respectively. S5.

4. Based on the cosine similarity algorithm, the normalized vector of the duct state features and the standard normalized vector of the duct state features are processed to calculate the similarity and generate the duct state start-up completion similarity data. S5.5 Determine whether the similarity data of the duct status start-up completion is greater than the set duct status start-up completion similarity threshold. If yes, the duct start-up is complete; otherwise, the duct start-up is not complete.

4. The automatic sorting and sequential start-up control method for unordered operation of multiple sets of ventilation ducts according to claim 3, characterized in that, S5.4 is replaced by the following steps: S5.4a. Collect and set the weights of each item in the duct state feature vector, and generate duct state feature weight data; S5.4b: Based on the weighted data of the duct state features, perform weighted cosine similarity calculation on the normalized vector of the duct state features and the standard normalized vector of the duct state features to generate similarity data of duct state start-up and completion.

5. The automatic sorting and sequential start-up control method for multiple sets of air ducts operating in disorder as described in claim 3, characterized in that, The steps involved in generating the duct start-up failure analysis data in S8 are as follows: Set the normalized vector of the duct state features when the duct starts up abnormally as the normalized vector of the duct abnormal state features. Based on the PCA algorithm, the normalized vectors of all abnormal state features of the ventilation duct in the historical ventilation duct start-up anomalies are reduced in dimensionality and then HDBSCAN clustering analysis is performed to generate multiple data clusters of ventilation duct start-up anomaly types. Associate each data cluster containing the abnormal start-up type of the ventilation duct with a label that sets the abnormal start-up type of the ventilation duct. The average value of each term in the normalized vector of all abnormal state features of each ventilation duct in each data cluster of abnormal start-up type is calculated to obtain the average vector of the data cluster. The cosine similarity between the normalized vector of the abnormal state features of the ventilation duct and the average vector of each data cluster is calculated. The maximum cosine similarity is extracted and it is determined whether it is greater than the set cosine similarity threshold. If it is, the ventilation duct start-up abnormality type label corresponding to the data cluster with the maximum cosine similarity is output; otherwise, "Unknown abnormality has occurred" is output.

6. The automatic sorting and sequential start-up control method for multiple sets of air ducts operating in disorder as described in claim 1, characterized in that, The method further includes the following steps: Operators can remove the duct from the duct startup sequence data at any time by triggering the duct's trigger switch again. If the duct has already been started, a shutdown operation will be performed on the duct.

Images