A hoist control method with adaptive parameter adjustment

Abstract

The present invention relates to the technical field of hoist control, and discloses a hoist control method with adaptive parameter adjustment, the method comprising: collecting hoist operation status data; constructing a loss function of model parameters in a hoist intelligent control model based on a training data set, and optimizing and solving the loss function; using the hoist intelligent control model to receive the hoist operation status data, and optimizing the hoist control strategy, obtaining multi-stage inverter parameter adjustment results, and covering the inverter parameter settings. The present invention extracts state time series information features in combination with the hoist operation status data at adjacent operating moments, performs parameter mapping and parameter adjustment probability calculation on the current inverter parameter settings, wherein the greater the frequency switching amplitude, the greater the parameter adjustment probability, reducing the loss of the inverter in the hoist, retaining the inverter parameter adjustment results with a parameter adjustment probability higher than a preset threshold, and realizing adaptive parameter adjustment of the inverter in the hoist.

Description

Technical Field

[0001] The invention relates to the technical field of hoist control, and in particular to a hoist control method with adaptive parameter adjustment. Background Art

[0002] Hoists are important equipment widely used in industrial production such as mines, ports, metallurgy, and electricity. They are responsible for lifting materials or ores from low places to high places, or transporting them to designated locations. The operating performance and safety of the hoist directly affect production efficiency and economic benefits. Therefore, it is of great engineering application value to study the control system of the hoist and realize efficient and safe automatic control. With the development of industrial automation technology, traditional hoist control systems have gradually exposed some shortcomings. Especially under the influence of environmental changes, load fluctuations and other factors, traditional control methods such as PID control and linear control are often difficult to adapt to the uncertainty and changes of system parameters, resulting in complex use problems.

[0003] At present, most of the control methods of elevators are divided into the following two categories:

[0004] 1. Choose to use a servo motor to drive the cargo platform of the elevator: This method uses the position mode of the servo motor to set the speed and movement target position to achieve high-speed positioning. Although it is very easy to achieve the function, the servo motor is relatively expensive, especially when the load is large and a high-power servo system must be used, the cost is even higher.

[0005] 2. Choose the inverter drive + barcode position feedback control method: This method uses an inverter to drive an ordinary three-phase motor, plus a barcode to feedback the actual lifting position, and uses the feedback position to change the motor speed until it stops at the corresponding work position; although this method uses an inverter instead of a servo motor, which greatly reduces material costs, the barcode feedback system increases costs, and the control method is much more complicated than the servo system, which is also a less ideal control method. Summary of the invention

[0006] In view of this, the present invention provides a hoist control method with adaptive parameter adjustment, which monitors the working status of the hoist in real time and automatically adjusts the control strategy and parameters, so that the hoist always operates in an optimized manner under various working conditions, thereby improving work efficiency, and uses a new control method of inverter drive + proximity switch sensing to control the hoist.

[0007] To achieve the above object, the present invention provides a method for controlling a hoist with adaptive parameter adjustment, comprising the following steps:

[0008] S1: Collecting the hoist operation status data, wherein the hoist operation status data includes the number and position of the hoist start and stop points, the inverter configuration parameters, the inverter parameter settings and the load weight;

[0009] S2: constructing a model structure of an intelligent control model of an elevator based on the elevator operation status data, and forming a training data set, wherein the intelligent control model of the elevator takes the elevator operation status data as input and takes the adjustment result of the multi-stage inverter parameters as output;

[0010] S3: Based on the training data set, a loss function of the model parameters in the elevator intelligent control model is constructed, and the loss function is optimized to obtain the optimal model parameters. Based on the optimal model parameters and the model parameters, an intelligent control model of the elevator is established, wherein the optimization solution adopts an improved artificial immune method;

[0011] S4: Utilize the elevator intelligent control model to receive the elevator operation status data, optimize the elevator control strategy, obtain the multi-stage inverter parameter adjustment results, and overwrite the inverter parameter settings.

[0012] As a further improvement method of the present invention:

[0013] Optionally, collecting the elevator operation status data in step S1 includes:

[0014] The elevator is built by frequency converter drive and proximity switch sensing. The elevator operation process is divided into multiple working sections according to the number of elevator start and stop points. The area between two adjacent elevator start and stop points is a working section. The acceleration and deceleration time and frequency converter parameters of each working section are initially set. The frequency converter parameters are the frequency converter frequency. When the proximity switch at the elevator start and stop point senses the transported material, the frequency converter automatically switches to low frequency mode and performs deceleration operation. After passing the elevator start and stop point, the frequency converter runs according to the set frequency converter parameters. When the frequency of the frequency converter is abnormal, the elevator is shut down.

[0015] During the operation of the elevator, the elevator operation status data at different operation times are collected, where the elevator operation status data at the tth operation time is: :

[0016]

[0017] in:

[0018] T stands for transpose;

[0019] Indicates the number and location of the start and stop points of the elevator , Indicates the number of start and stop points of the elevator;

[0020] express The location of the start and stop points of the elevator. Indicates The location of the start and stop points of the elevator. ;

[0021] Indicates the inverter configuration parameters of the hoist. They are the minimum frequency and maximum frequency of the inverter. Indicates the deceleration time in the initial setting of the inverter parameters. Indicates the acceleration time in the initially set inverter parameters; in the embodiment of the present invention, the inverter needs to increase the frequency during the acceleration time and needs to reduce the frequency during the deceleration time;

[0022] Indicates the inverter parameter settings at the tth operating moment, It represents the inverter parameter sequence of the jth working section at the tth operating moment, wherein the jth working section is the working section of the elevator between the j+1th elevator start-stop point and the jth elevator start-stop point. , Indicates the total number of working sections of the elevator;

[0023] Divide each work section into S work sub-sections. represents the frequency converter of the S working subsections in the j-th working section at the t-th operating moment, represents the frequency converter frequency of the sth working subsection of the jth working section at the tth operating moment,

[0024] It indicates the weight of the transport equipment that the elevator needs to transport at the tth operating moment.

[0025] Optionally, in step S2, the model structure of the elevator intelligent control model is constructed based on the elevator operation status data, including:

[0026] A model structure of an intelligent control model of an elevator is constructed based on the operation status data of the elevator, wherein the intelligent control model of the elevator includes an input layer, a timing information extraction layer, a parameter adjustment probability calculation layer and a parameter mapping layer. The input layer is used to receive the operation status data of the elevator at multiple adjacent operation times. The timing information extraction layer adopts the structure of a long short-term memory network to extract and process the state timing information of the operation status data of the elevator at multiple adjacent operation times to obtain the state timing information characteristics. The parameter adjustment probability calculation layer is used to obtain the operation status data of the elevator at the current t-th operation time and calculate the parameter adjustment probability of each working section. The parameter mapping layer maps the inverter parameter settings at the current t-th operation time in combination with the state timing information characteristics to obtain the inverter parameter adjustment results of multiple working sections, retains the inverter parameter adjustment results whose parameter adjustment probability is higher than the preset threshold, and covers the inverter parameter settings at the current t-th operation time with the retained results as the multi-segment inverter parameter adjustment results;

[0027] Collect training data for optimizing model parameters in the elevator intelligent control model to form a training data set data:

[0028]

[0029] in:

[0030] represents the training data subset with num number of work sections, Indicates the preset maximum number of working sections;

[0031] Represents a subset of training data The D group of training data in Represents a subset of training data The dth group of training data in The dth group of state time series information features representing the operation state data of the hoist at multiple adjacent operation times under the number of working sections num, Indicates the characteristics of state timing information The corresponding inverter parameter settings to be optimized by the strategy, Indicates the characteristics of state timing information The corresponding optimal multi-stage inverter parameter adjustment results.

[0032] Optionally, in step S3, constructing a loss function of model parameters in the elevator intelligent control model based on the training data set includes:

[0033] The loss function of the model parameters in the elevator intelligent control model is constructed based on the training data set, where the model parameters to be optimized in the elevator intelligent control model are the mapping matrix parameters in the parameter mapping layer, and the loss function of the mapping matrix parameters corresponding to the number of elevator working sections num is:

[0034]

[0035]

[0036] in:

[0037] Represents the loss function of the mapping matrix parameters corresponding to the number of hoist working sections num, Indicates the mapping matrix parameters corresponding to the number num of hoist working sections to be optimized. ;

[0038] Indicates the use of mapping matrix parameters Status timing information characteristics And inverter parameter settings Mapping is performed, and the mapped parameter adjustment results are obtained;

[0039] represents the L1 norm;

[0040] Represents the Hadamard product, which multiplies corresponding elements of matrices or vectors of the same dimension one by one.

[0041] Optionally, optimizing and solving the loss function to obtain optimal model parameters includes:

[0042] The loss function The optimization solution process is:

[0043] S31: Randomly generate a set of mapping matrix parameters , the generated mapping matrix parameters Perform iterative mapping to obtain U groups of mapping matrix parameters, where the iterative mapping formula of the u-th group of mapping matrix parameters is:

[0044]

[0045] in:

[0046] represents the u-th group of mapping matrix parameters obtained by iterative mapping, ;

[0047] Represents the mapping matrix parameters The upper limit of

[0048] represents the iteration coefficient; in the embodiment of the present invention, Set to 3.5;

[0049] S32: Set the current number of iterations of each group of mapping matrix parameters to h, the maximum number of iterations to H, and the initial value of h to 0, then The h-th iteration result is ;

[0050] S33: Calculate the mutation probability of the mapping matrix parameters, then the mapping matrix parameters The probability of mutation is :

[0051]

[0052] in:

[0053] represents an exponential function with a natural constant as base;

[0054] represents the coefficient of variation;

[0055] S34: performing mutation or immune evaluation processing on the mapping matrix parameters based on the mutation probability to generate the h+1th iteration result of the mapping matrix parameters;

[0056] S35: Let h = h + 1 until the maximum number of iterations is reached, and use the U group mapping matrix parameters obtained in the current iteration as the loss function in turn The input value is used to obtain the loss function value of each group of mapping matrix parameters, and the mapping matrix parameter with the smallest loss function value is selected as the optimal model parameter corresponding to the number of hoist working sections num .

[0057] Optionally, in step S34, the mapping matrix parameters are mutated or immune evaluated based on the mutation probability to generate the h+1th iteration result of the mapping matrix parameters, including:

[0058] The mapping matrix parameters The probability of mutation treatment is , the probability of immune assessment treatment is , mapping matrix parameters The formula for mutation processing is:

[0059]

[0060] in:

[0061] The mean is 0 and the variance is The normal distribution of represents the gamma function;

[0062] represents a normal distribution with a mean of 0 and a variance of 1;

[0063] Represents a random number between -1 and 1

[0064] To obey The random amount of To obey The random amount of

[0065] And manually screen the mutation processing results:

[0066]

[0067] Mapping matrix parameters The formula for immune assessment processing is:

[0068]

[0069] in:

[0070] Represents a random number between 0 and 1;

[0071] It represents the mapping matrix parameter with the smallest loss function value after being substituted into the loss function in the U group mapping matrix obtained in the h-th iteration.

[0072] Optionally, in step S4, the elevator intelligent control model is used to receive the elevator operation status data, and the elevator control strategy is optimized to obtain the multi-stage inverter parameter adjustment result, including:

[0073] The elevator intelligent control model is used to receive the elevator operation status data and optimize the elevator control strategy. The optimization process of the elevator control strategy at the t-th operating moment is:

[0074] S41: Input layer elevator operation status data of multiple adjacent operation times ;

[0075] S42: The time series information extraction layer performs state time series information extraction processing on the elevator operation state data of multiple adjacent operation moments to obtain the state time series information feature of the t-th operation moment :

[0076]

[0077] in:

[0078] Represents the Hadamard product, which multiplies corresponding elements of matrices or vectors of the same dimension one by one; The dimension and Consistency;

[0079] express The attention weight, express In the embodiment of the present invention, if the attention weight is too large, it means that the number and position of the start and stop points of the elevator have changed, and the acceleration and deceleration time and the inverter parameters of each working section need to be reset;

[0080] S43: The parameter adjustment probability calculation layer obtains the elevator operation status data at the current t-th operation time , and calculate the parameter adjustment probability of each working section, where the parameter adjustment probability of the jth working section is :

[0081]

[0082] in:

[0083] Indicates the preset control parameters. Set to 0.5;

[0084] S44: Parameter mapping layer combined with state timing information features The inverter parameter settings for the current t-th running moment Mapping is performed, where the mapping formula is:

[0085]

[0086] in:

[0087] Indicates the number of working sections of the elevator The corresponding optimal model parameters; The dimension of the result of the calculation is Consistency;

[0088] represents the inverter parameter adjustment result of the j-th working section obtained by mapping;

[0089] Included The inverter parameter adjustment results of each working section;

[0090] S45: If If it is higher than the preset threshold, In Replace with ,in ;

[0091] S46: Repeat step S45, The replacement result is , as the multi-stage inverter parameter adjustment result at the t-th operating moment;

[0092] Adjust the inverter parameter settings of the hoist at the current t-th operating moment to .

[0093] In order to solve the above problem, the present invention provides an electronic device, the electronic device comprising:

[0094] A memory storing at least one instruction;

[0095] Communication interface, enabling electronic equipment to communicate; and

[0096] The processor executes the instructions stored in the memory to implement the above-mentioned elevator control method with adaptive parameter adjustment.

[0097] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned hoist control method with adaptive parameter adjustment.

[0098] Compared with the prior art, the present invention proposes a hoist control method with adaptive parameter adjustment, which has the following advantages:

[0099] Firstly, this scheme proposes an adaptive parameter adjustment method for the hoist, which monitors the current inverter parameters, working section and load weight of the hoist in real time, constructs the hoist operation status data representing the current operation status of the hoist, extracts the state time series information features in combination with the elevator operation status data at the adjacent operation time, performs parameter mapping and parameter adjustment probability calculation on the current inverter parameter setting, wherein the larger the frequency switching amplitude, the greater the parameter adjustment probability, reduces the loss of the inverter in the hoist, and increases the service life of the inverter, retains the inverter parameter adjustment results with a parameter adjustment probability higher than the preset threshold, and covers the inverter parameter setting at the current operation time with the retained results as the multi-stage inverter parameter adjustment result, realizes the adaptive parameter adjustment of the inverter in the hoist, and then controls the hoisting speed of the hoist; in the process of solving the model parameters, a chaotic mapping method is used to generate multiple groups of parameters, and the parameters are mutated and immune evaluated respectively in combination with the immune principle in biology, retains the better parameters, and mutates the worse parameters, improves the solution speed of the model parameters, and uses multiple random variables for iteration to avoid falling into the local optimal solution.

[0100] At the same time, this solution uses a new control method of inverter drive + proximity switch sensing to control the elevator, replaces the servo with an ordinary motor, replaces the barcode feedback device with a cheap proximity switch, and replaces the more complex control method of barcode feedback closed-loop control speed with the simple multi-speed output function of the inverter, which significantly reduces the cost, about 1-2 times cheaper than the servo system or barcode feedback system. The method is simple, easy to program, and quick to debug. No special modules are required for pulse positioning output or barcode feedback input, and only simple switch quantity processing is required. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] Figure 1 A schematic flow chart of a method for controlling a hoist with adaptive parameter adjustment provided in accordance with an embodiment of the present invention.

[0102] Figure 2 A frequency converter wiring diagram of a frequency converter in a hoist provided in one embodiment of the present invention.

[0103] Figure 3 This is a timing diagram of the multi-speed and external terminal actions of the inverter provided by an embodiment of the present invention.

[0104] Figure 4 A programming diagram for automatic multi-speed processing of a frequency converter provided by an embodiment of the present invention.

[0105] Figure 5 A touch screen interface for setting inverter parameters provided by an embodiment of the present invention.

[0106] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0107] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0108] The embodiment of the present application provides a method for controlling a hoist with adaptive parameter adjustment. The execution subject of the method for controlling a hoist with adaptive parameter adjustment includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for controlling a hoist with adaptive parameter adjustment can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0109] Example 1

[0110] A method for controlling a hoist with adaptive parameter adjustment comprises the following steps:

[0111] S1: Collecting the elevator operation status data, wherein the elevator operation status data includes the number and position of the elevator start and stop points, the inverter configuration parameters, the inverter parameter settings and the load weight.

[0112] The step S1 of collecting the elevator operation status data includes:

[0113] The elevator is built by frequency converter drive and proximity switch sensing. The elevator operation process is divided into multiple working sections according to the number of elevator start and stop points. The area between two adjacent elevator start and stop points is a working section. The acceleration and deceleration time and frequency converter parameters of each working section are initially set. The frequency converter parameters are the frequency converter frequency. When the proximity switch at the elevator start and stop point senses the transported material, the frequency converter automatically switches to low frequency mode and performs deceleration operation. After passing the elevator start and stop point, the frequency converter runs according to the set frequency converter parameters.

[0114] During the operation of the elevator, the elevator operation status data at different operation times are collected, where the elevator operation status data at the tth operation time is: :

[0115]

[0116] in:

[0117] T stands for transpose;

[0118] Indicates the number and location of the start and stop points of the elevator , Indicates the number of start and stop points of the elevator;

[0119] express The location of the start and stop points of the elevator. Indicates The location of the start and stop points of the elevator. ;

[0120] Indicates the inverter configuration parameters of the hoist. They are the minimum frequency and maximum frequency of the inverter. Indicates the deceleration time in the initial setting of the inverter parameters. Indicates the acceleration time in the initially set inverter parameters; in the embodiment of the present invention, the inverter needs to increase the frequency during the acceleration time and needs to reduce the frequency during the deceleration time;

[0121] Indicates the inverter parameter settings at the tth operating moment, It represents the inverter parameter sequence of the jth working section at the tth operating moment, wherein the jth working section is the working section of the elevator between the j+1th elevator start-stop point and the jth elevator start-stop point. , Indicates the total number of working sections of the elevator;

[0122] Divide each work section into S work sub-sections. represents the frequency converter of the S working subsections in the j-th working section at the t-th operating moment, represents the frequency converter frequency of the sth working subsection of the jth working section at the tth operating moment,

[0123] It indicates the weight of the transport equipment that the elevator needs to transport at the tth operating moment.

[0124] S2: Construct the model structure of the elevator intelligent control model based on the elevator operation status data and form a training data set.

[0125] The model structure of the elevator intelligent control model is constructed based on the elevator operation status data in step S2, including:

[0126] A model structure of an intelligent control model of an elevator is constructed based on the operation status data of the elevator, wherein the intelligent control model of the elevator includes an input layer, a timing information extraction layer, a parameter adjustment probability calculation layer and a parameter mapping layer. The input layer is used to receive the operation status data of the elevator at multiple adjacent operation times. The timing information extraction layer adopts the structure of a long short-term memory network to extract and process the state timing information of the operation status data of the elevator at multiple adjacent operation times to obtain the state timing information characteristics. The parameter adjustment probability calculation layer is used to obtain the operation status data of the elevator at the current t-th operation time and calculate the parameter adjustment probability of each working section. The parameter mapping layer maps the inverter parameter settings at the current t-th operation time in combination with the state timing information characteristics to obtain the inverter parameter adjustment results of multiple working sections, retains the inverter parameter adjustment results whose parameter adjustment probability is higher than the preset threshold, and covers the inverter parameter settings at the current t-th operation time with the retained results as the multi-segment inverter parameter adjustment results;

[0127] Collect training data for optimizing model parameters in the elevator intelligent control model to form a training data set data:

[0128]

[0129] in:

[0130] represents the training data subset with num number of work sections, Indicates the preset maximum number of working sections;

[0131] Represents a subset of training data The D group of training data in Represents a subset of training data The dth group of training data in The dth group of state time series information features representing the operation state data of the hoist at multiple adjacent operation times under the number of working sections num, Indicates the characteristics of state timing information The corresponding inverter parameter settings to be optimized by the strategy, Indicates the characteristics of state timing information The corresponding optimal multi-stage inverter parameter adjustment results.

[0132] S3: Construct a loss function of the model parameters in the elevator intelligent control model based on the training data set, optimize and solve the loss function to obtain the optimal model parameters, and establish the elevator intelligent control model based on the optimal model parameters and model parameters.

[0133] In the step S3, the loss function of the model parameters in the elevator intelligent control model is constructed based on the training data set, including:

[0134] The loss function of the model parameters in the elevator intelligent control model is constructed based on the training data set, where the model parameters to be optimized in the elevator intelligent control model are the mapping matrix parameters in the parameter mapping layer, and the loss function of the mapping matrix parameters corresponding to the number of elevator working sections num is:

[0135]

[0136]

[0137] in:

[0138] Represents the loss function of the mapping matrix parameters corresponding to the number of hoist working sections num, Indicates the mapping matrix parameters corresponding to the number num of hoist working sections to be optimized. ;

[0139] Indicates the use of mapping matrix parameters Status timing information characteristics And inverter parameter settings Mapping is performed, and the mapped parameter adjustment results are obtained;

[0140] represents the L1 norm;

[0141] Represents the Hadamard product, which multiplies corresponding elements of matrices or vectors of the same dimension one by one.

[0142] The optimization and solution of the loss function to obtain the optimal model parameters includes:

[0143] The loss function The optimization solution process is:

[0144] S31: Randomly generate a set of mapping matrix parameters , the generated mapping matrix parameters Perform iterative mapping to obtain U groups of mapping matrix parameters, where the iterative mapping formula of the u-th group of mapping matrix parameters is:

[0145]

[0146] in:

[0147] represents the u-th group of mapping matrix parameters obtained by iterative mapping, ;

[0148] Represents the mapping matrix parameters The upper limit of

[0149] represents the iteration coefficient; in the embodiment of the present invention, Set to 3.5;

[0150] S32: Set the current number of iterations of each group of mapping matrix parameters to h, the maximum number of iterations to H, and the initial value of h to 0, then The h-th iteration result is ;

[0151] S33: Calculate the mutation probability of the mapping matrix parameters, then the mapping matrix parameters The probability of mutation is :

[0152]

[0153] in:

[0154] represents an exponential function with a natural constant as base;

[0155] represents the coefficient of variation;

[0156] S34: performing mutation or immune evaluation processing on the mapping matrix parameters based on the mutation probability to generate the h+1th iteration result of the mapping matrix parameters;

[0157] S35: Let h = h + 1 until the maximum number of iterations is reached, and use the U group mapping matrix parameters obtained in the current iteration as the loss function in turn The input value is used to obtain the loss function value of each group of mapping matrix parameters, and the mapping matrix parameter with the smallest loss function value is selected as the optimal model parameter corresponding to the number of hoist working sections num .

[0158] In the step S34, the mapping matrix parameters are mutated or immune evaluated based on the mutation probability to generate the h+1th iteration result of the mapping matrix parameters, including:

[0159] The mapping matrix parameters The probability of mutation treatment is , the probability of immune assessment treatment is , mapping matrix parameters The formula for mutation processing is:

[0160]

[0161] in:

[0162] The mean is 0 and the variance is The normal distribution of represents the gamma function;

[0163] represents a normal distribution with a mean of 0 and a variance of 1;

[0164] Represents a random number between -1 and 1

[0165] To obey The random amount of To obey The random amount of

[0166] And manually screen the mutation processing results:

[0167]

[0168] Mapping matrix parameters The formula for immune assessment processing is:

[0169]

[0170] in:

[0171] Represents a random number between 0 and 1;

[0172] It represents the mapping matrix parameter with the smallest loss function value after being substituted into the loss function in the U group mapping matrix obtained in the h-th iteration.

[0173] S4: Utilize the elevator intelligent control model to receive the elevator operation status data, optimize the elevator control strategy, obtain the multi-stage inverter parameter adjustment results, and overwrite the inverter parameter settings.

[0174] In the step S4, the elevator intelligent control model is used to receive the elevator operation status data, and the elevator control strategy is optimized to obtain the multi-stage inverter parameter adjustment results, including:

[0175] The elevator intelligent control model is used to receive the elevator operation status data and optimize the elevator control strategy. The optimization process of the elevator control strategy at the t-th operating moment is:

[0176] S41: Input layer elevator operation status data of multiple adjacent operation times ;

[0177] S42: The time series information extraction layer performs state time series information extraction processing on the elevator operation state data of multiple adjacent operation moments to obtain the state time series information feature of the t-th operation moment :

[0178]

[0179] in:

[0180] Represents the Hadamard product, which multiplies corresponding elements of matrices or vectors of the same dimension one by one; The dimension and Consistency;

[0181] express The attention weight, express In the embodiment of the present invention, if the attention weight is too large, it means that the number and position of the start and stop points of the elevator have changed, and the acceleration and deceleration time and the inverter parameters of each working section need to be reset;

[0182] S43: The parameter adjustment probability calculation layer obtains the elevator operation status data at the current t-th operation time , and calculate the parameter adjustment probability of each working section, where the parameter adjustment probability of the jth working section is :

[0183]

[0184] in:

[0185] Indicates the preset control parameters. Set to 0.5;

[0186] S44: Parameter mapping layer combined with state timing information features The inverter parameter settings for the current t-th running moment Mapping is performed, where the mapping formula is:

[0187]

[0188] in:

[0189] Indicates the number of working sections of the elevator The corresponding optimal model parameters; The dimension of the result of the calculation is Consistency;

[0190] represents the inverter parameter adjustment result of the j-th working section obtained by mapping;

[0191] Included The inverter parameter adjustment results of each working section;

[0192] S45: If If it is higher than the preset threshold, In Replace with ,in ;

[0193] S46: Repeat step S45, The replacement result is , as the multi-stage inverter parameter adjustment result at the t-th operating moment;

[0194] Adjust the inverter parameter settings of the hoist at the current t-th operating moment to .

[0195] Example 2

[0196] like Figure 2 As shown, it is a wiring diagram of the inverter provided by an embodiment of the present invention, taking the two start and stop points of the Delta VFD022C43A-21 inverter as an example, which can realize the frequency adjustment of the inverter in the hoist control method with adaptive parameter adjustment in Example 1.

[0197] The control process of the hoist includes: determining the number of inverter multi-speed output points according to the number of start and stop points of the hoist; correctly setting the inverter parameters; programming the inverter operation and multi-speed part, such as Figure 3 1 is a timing diagram of multi-speed and external terminal actions provided by an embodiment of the present invention with reference to the Delta C2000 inverter user manual.

[0198] like Figure 4 As shown, it is a programming diagram of the inverter multi-speed automatic processing provided by an embodiment of the present invention. The inverter multi-speed part is divided into two parts: the conditional judgment assignment of the automatic rising section and the automatic falling section and the judgment of the output conditions of the multi-speed output point.

[0199] The multi-speed switching can only be performed in the complete automatic rising section or automatic falling section, otherwise the inverter will directly switch to the low-frequency mode. The complete automatic rising section means that the lifting starts from the lower limit position of the hoist and stops until the upper limit position is sensed. The whole process is uninterrupted. If it is paused during the rising period, it will no longer belong to the complete automatic rising section if it continues to rise, and it can only rise at a low speed. The complete automatic falling section means that the lifting starts from the upper limit position of the hoist and stops until the lower limit position is sensed. The whole process is uninterrupted. If it is paused during the falling period, it will no longer belong to the complete automatic falling section if it continues to fall, and it can only fall at a low speed.

[0200] To avoid the situation where the system does not know where to stop, continues to move quickly according to the original fast motion time, and then switches to the second multi-speed, which will cause the target limit proximity switch to be sensed at high speed and cannot stop at the correct position.

[0201] Example 3

[0202] Taking the hoisting condition of the Delta VFD022C43A-21 inverter with two start and stop points as an example, an initial setting of the inverter parameters is provided. The contents of the inverter parameters initially set are as follows: 1. Frequency source and operation command source setting: The inverter defaults to AUTO mode when it leaves the factory. Do not change to HAND mode and keep the default mode; frequency command source 00-20 is set to 0 (input by digital operator), and operation command source 00-21 is set to 1 (operated by external terminal); 2. Acceleration and deceleration time setting, the first acceleration time 01-12 is set to 5 (the default is 10, i.e. 0.1 seconds), the first deceleration time 01-13 is set to 2 (the default is 10, i.e. 0.1 seconds), and the acceleration and deceleration and S curve time unit 01-45 is set to 0 (the default is 0, i.e. 0.01 seconds); 3. Multi-speed input point setting: Multi-function input command 1 (MI1) 02-01 is set to 1 (default 1, i.e. multi-speed command 1 / multi-point position command 1 , which is the high-speed operation stage of the inverter), multi-function input command 2 (MI2) 02-02 is set to 2, (default 2, that is, multi-speed command 1 / multi-point position command 1, which is the high-speed operation stage of the inverter); 4. Multi-speed frequency setting: the first speed 04-00 is preset to 50.00 Hz (the default is 0.0 Hz, which is the high-speed stage), corresponding to the speed when MI1 is connected, and the second speed 04-01 is preset to 10.00 Hz (the default is 0.0 Hz, which is the low-speed stage), corresponding to the speed when MI2 is connected; 5. Inverter fault output setting: multi-function output 1 (RLY1, i.e. RB1-RC1) 02-13 is set to 11 (when the inverter detects an abnormal condition (except Lv shutdown), the contact will "act"); 6. Braking resistor setting: no parameter setting is required, just connect the line and it can be used.

[0203] Some of the inverter parameters initially set can be transferred to Figure 5 The inverter parameter setting touch screen interface shown is for easy setting; after the initial setting of the parameters, the inverter will operate normally, and an adaptive parameter adjustment hoist control method in Example 1 is used to monitor the current inverter parameters, working section and load weight of the hoist in real time, as well as adaptive adjustment of the inverter parameters, so as to improve the stability of the hoist operation.

[0204] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0205] It should be noted that the serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. And the terms "including", "comprising" or any other variants thereof in this article are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "including a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0206] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0207] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for controlling a hoist with adaptive parameter adjustment, characterized in that: The method comprises: S1: Collecting the hoist operation status data, wherein the hoist operation status data includes the number and position of the hoist start and stop points, the inverter configuration parameters, the inverter parameter settings and the load weight; S2: construct the model structure of the elevator intelligent control model based on the elevator operation status data and form a training data set; The intelligent control model of the hoist includes an input layer, a timing information extraction layer, a parameter adjustment probability calculation layer and a parameter mapping layer. The input layer is used to receive the hoist operation status data of multiple adjacent operation moments. The timing information extraction layer adopts the structure of a long short-term memory network to extract the state timing information of the hoist operation status data of multiple adjacent operation moments to obtain the state timing information characteristics. The parameter adjustment probability calculation layer is used to obtain the hoist operation status data of the current t-th operation moment and calculate the parameter adjustment probability of each working section. The parameter mapping layer maps the inverter parameter setting of the current t-th operation moment in combination with the state timing information characteristics to obtain the inverter parameter adjustment results of multiple working sections, retains the inverter parameter adjustment results with a parameter adjustment probability higher than a preset threshold, and covers the inverter parameter setting of the current t-th operation moment with the reserved results as the multi-segment inverter parameter adjustment results; Collect training data for optimizing model parameters in the elevator intelligent control model to form a training data set data: in: represents the training data subset with num number of work sections, Indicates the preset maximum number of working sections; Represents a subset of training data The D group of training data in Represents a subset of training data The dth group of training data in The dth group of state time series information features representing the operation state data of the hoist at multiple adjacent operation times under the number of working sections num, Indicates the characteristics of state timing information The corresponding inverter parameter settings to be optimized by the strategy, Indicates the characteristics of state timing information The corresponding optimal multi-stage inverter parameter adjustment results; S3: construct a loss function of the model parameters in the elevator intelligent control model based on the training data set, optimize and solve the loss function to obtain the optimal model parameters, and establish the elevator intelligent control model based on the optimal model parameters and the model parameters; The loss function of the model parameters in the elevator intelligent control model is constructed based on the training data set, where the model parameters to be optimized in the elevator intelligent control model are the mapping matrix parameters in the parameter mapping layer, and the loss function of the mapping matrix parameters corresponding to the number of elevator working sections num is: in: Represents the loss function of the mapping matrix parameters corresponding to the number of hoist working sections num, Indicates the mapping matrix parameters corresponding to the number num of hoist working sections to be optimized. ; Indicates the use of mapping matrix parameters Status timing information characteristics And inverter parameter settings Mapping is performed, and the mapped parameter adjustment results are obtained; represents the L1 norm; Represents the Hadamard product, which multiplies corresponding elements of matrices or vectors of the same dimension one by one; S4: Utilize the elevator intelligent control model to receive the elevator operation status data, optimize the elevator control strategy, obtain the multi-stage inverter parameter adjustment results, and overwrite the inverter parameter settings.

2. The method for controlling a hoist with adaptive parameter adjustment according to claim 1, characterized in that: The S1 step includes: During the operation of the elevator, the elevator operation status data at different operation times are collected, where the elevator operation status data at the tth operation time is: : in: T stands for transpose; Indicates the number and location of the start and stop points of the elevator , Indicates the number of start and stop points of the elevator; express The location of the start and stop points of the elevator. Indicates The location of the start and stop points of the elevator. ; Indicates the inverter configuration parameters of the hoist. They are the minimum frequency and maximum frequency of the inverter. Indicates the deceleration time in the initial setting of the inverter parameters. Indicates the acceleration time in the initial setting of the inverter parameters; Indicates the inverter parameter settings at the tth operating moment, It represents the inverter parameter sequence of the jth working section at the tth operating moment, wherein the jth working section is the working section of the elevator between the j+1th elevator start-stop point and the jth elevator start-stop point. , Indicates the total number of working sections of the elevator; Divide each work section into S work sub-sections. represents the frequency converter of the S working subsections in the j-th working section at the t-th operating moment, represents the frequency converter frequency of the sth working subsection of the jth working section at the tth operating moment, ; It indicates the weight of the transport equipment that the elevator needs to transport at the tth operating moment.

3. The method for controlling a hoist with adaptive parameter adjustment according to claim 1, characterized in that: The optimization and solution of the loss function to obtain the optimal model parameters includes: The loss function The optimization solution process is: S31: Randomly generate a set of mapping matrix parameters , the generated mapping matrix parameters Perform iterative mapping to obtain U groups of mapping matrix parameters, where the iterative mapping formula of the u-th group of mapping matrix parameters is: in: represents the u-th group of mapping matrix parameters obtained by iterative mapping, ; Represents the mapping matrix parameters The upper limit of represents the iteration coefficient; S32: Set the current number of iterations of each group of mapping matrix parameters to h, the maximum number of iterations to H, and the initial value of h to 0, then The h-th iteration result is ; S33: Calculate the mutation probability of the mapping matrix parameters, then the mapping matrix parameters The probability of mutation is : in: represents an exponential function with a natural constant as base; represents the coefficient of variation; S34: performing mutation or immune evaluation processing on the mapping matrix parameters based on the mutation probability to generate the h+1th iteration result of the mapping matrix parameters; S35: Let h = h + 1 until the maximum number of iterations is reached, and use the U group mapping matrix parameters obtained in the current iteration as the loss function in turn The input value is used to obtain the loss function value of each group of mapping matrix parameters, and the mapping matrix parameter with the smallest loss function value is selected as the optimal model parameter corresponding to the number of hoist working sections num .

4. The method for controlling a hoist with adaptive parameter adjustment according to claim 3, characterized in that: The S4 step includes: The elevator intelligent control model is used to receive the elevator operation status data and optimize the elevator control strategy. The optimization process of the elevator control strategy at the t-th operating moment is: S41: Input layer elevator operation status data of multiple adjacent operation times ; S42: The time series information extraction layer performs state time series information extraction processing on the elevator operation state data of multiple adjacent operation moments to obtain the state time series information feature of the t-th operation moment : in: Represents the Hadamard product, which multiplies corresponding elements of matrices or vectors of the same dimension one by one; express The attention weight, express The attention weight of S43: The parameter adjustment probability calculation layer obtains the elevator operation status data at the current t-th operation time , and calculate the parameter adjustment probability of each working section, where the parameter adjustment probability of the jth working section is : in: Indicates the preset control parameters. Set to 0.5; S44: Parameter mapping layer combined with state timing information features The inverter parameter settings for the current t-th running moment Mapping is performed, where the mapping formula is: in: Indicates the number of working sections of the elevator The corresponding optimal model parameters; represents the inverter parameter adjustment result of the j-th working section obtained by mapping; Included The inverter parameter adjustment results of each working section; S45: If If it is higher than the preset threshold, In Replace with ,in ; S46: Repeat step S45, The replacement result is , as the multi-stage inverter parameter adjustment result at the t-th operating moment; Adjust the inverter parameter settings of the hoist at the current t-th operating moment to .

5. The method for controlling a hoist with adaptive parameter adjustment according to claim 3, characterized in that: In the step S34, the mapping matrix parameters are mutated or immune evaluated based on the mutation probability to generate the h+1th iteration result of the mapping matrix parameters, including: The mapping matrix parameters The probability of mutation treatment is , the probability of immune assessment treatment is , mapping matrix parameters The formula for mutation processing is: in: The mean is 0 and the variance is The normal distribution of represents the gamma function; represents a normal distribution with a mean of 0 and a variance of 1; Represents a random number between -1 and 1 To obey The random amount, To obey The random amount of And manually screen the mutation processing results: Mapping matrix parameters The formula for immune assessment processing is: in: Represents a random number between 0 and 1; It represents the mapping matrix parameter with the smallest loss function value after being substituted into the loss function in the U group mapping matrix obtained in the h-th iteration.

Images