Concrete pumping equipment imaging method, device and medium
By constructing a single and group portrait model of concrete pumping equipment, using label sets and machine learning algorithms, massive data processing problems are solved, intuitive, accurate description and comprehensive comparison and analysis of the equipment group are realized, and data in the existing technology is solved Unsystematic problems.
Patent Information
- Application Number
- CN202210146321.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-02-17
- Publication Date
- 2024-12-10
- Estimated Expiration
- 2042-02-17
Smart Images
Figure CN114510472B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of engineering machinery, and in particular to a concrete pumping equipment imaging method, device and medium. Background Art
[0002] At present, with the increasing scale of concrete pumping equipment, the data collected by its various sensors are also gradually forming massive data. However, there is no systematic and effective processing solution for these massive data in the existing technology. They are often simply stored first, and then the data is selected for analysis when there is a business need. This temporary and targeted data analysis solution wastes massive data and cannot fully describe the characteristics of concrete pumping equipment, especially the characteristics of the concrete pumping equipment group cannot be intuitively and accurately described, and thus cannot support a comprehensive and objective comparative analysis of the equipment group. Summary of the invention
[0003] The object of the present invention is to provide a concrete pumping equipment imaging method, device and medium, which are used to at least partially solve the above technical problems.
[0004] In order to achieve the above-mentioned purpose, the present invention provides a concrete pumping equipment portrait method, comprising: constructing a single portrait model of the concrete pumping equipment represented by its label set; for an equipment group having multiple concrete pumping equipment, using an objective weighting method to determine the weight value of each concrete pumping equipment in the equipment group, and determining the label feature value of the equipment group based on the label feature value of each concrete pumping equipment and the weight value, and constructing a corresponding group portrait model based on the determined label feature value of the equipment group; and using a visualization method to process the single portrait model and / or the group portrait model to generate corresponding single equipment portraits and / or equipment group portraits.
[0005] Preferably, the construction of a monomer portrait model of concrete pumping equipment shown by its label set includes: acquiring and cleaning basic data of the concrete pumping equipment; classifying the cleaned basic data into labels, including classification into basic information labels, construction performance labels and performance performance labels; determining a parent label for showing equipment performance, and selecting several labels from the construction performance label and / or the performance performance label as child labels of the parent label, and setting weight values for the child labels; and performing statistical analysis on the basic data corresponding to each label to obtain corresponding label feature values, performing model training on the label feature values based on a machine learning algorithm, and obtaining the monomer portrait model shown by the label set.
[0006] Preferably, the construction performance label includes any one or more of a construction location label, a boom posture label, a construction intensity label, a construction frequency label and a construction extent label.
[0007] Preferably, the performance label includes any one or more of the following labels: a comprehensive energy consumption level label, a pumping system energy consumption level label, an idle system energy consumption level label and a driving system energy consumption level label set based on energy consumption data; a stroke in place rate level label, a pumping efficiency level label and a suction efficiency level label set based on efficiency data; an engine power level label set based on engine data; a pressure shock level label set based on pressure data; a hydraulic system oil temperature level label set based on oil temperature data; a mechanical system fault level label, a hydraulic system fault level label, an electrical system fault level label, a chassis and engine fault level label and other fault level labels set based on fault warning data; and a boom stability level label and a pipe blockage rate level label set based on boom data.
[0008] Preferably, the determined parent tag and the corresponding child tag include any one or more of the following:
[0009] The parent tag is determined to be the equipment construction saturation tag, and the corresponding child tags include the construction intensity tag, the construction frequency tag, and the construction breadth tag;
[0010] The parent label is determined to be the equipment overall efficiency grade label, and the corresponding child labels include the stroke in place rate grade label, the pumping efficiency grade label and the suction efficiency grade label;
[0011] The parent tag is determined to be the overall energy consumption level tag of the equipment, and the corresponding child tags include the comprehensive energy consumption level tag, the pumping system energy consumption level tag, the idle system energy consumption level tag, and the driving system energy consumption level tag;
[0012] The parent tag is determined to be the equipment boom reliability level tag, and the corresponding child tags include the boom stability level tag, the pipe blocking rate level tag, and the mechanical system failure level tag;
[0013] The parent tag is determined to be a power reliability level tag, and the corresponding child tags include the engine power level tag and the chassis and engine failure level tags;
[0014] The parent label is determined to be a hydraulic system reliability level label, and the corresponding child labels include a hydraulic system oil temperature level label and a hydraulic system fault level label.
[0015] Preferably, the determined parent tag and the corresponding child tag also include:
[0016] The parent tag is determined to be the overall performance tag of the equipment, and the corresponding child tags include any one or more of the following tags: the overall efficiency level label, the overall energy consumption level label, the boom reliability level label, the power reliability level label and the hydraulic system reliability level label.
[0017] Preferably, the statistical analysis of the basic data corresponding to each label to obtain the corresponding label characteristic value includes: obtaining the mean, median, first quartile or third quartile of the operating condition data in the basic data corresponding to each label; and performing weighted summation of all the obtained means, medians, first quartiles or third quartiles to obtain the label characteristic value.
[0018] Preferably, the model training of the label feature values based on the machine learning algorithm includes: performing outlier processing on the label feature values; clustering the label feature values after removing the outliers using the K-means clustering algorithm to obtain multiple classes and the center point of each class; classifying the label feature values according to the size of the label feature values corresponding to the center points; comparing the label feature values after removing the outliers with each of the center points, and selecting the level label corresponding to the nearest center point as the level label of the corresponding label feature value; and forming the monomer portrait model based on the level label, the corresponding weight set corresponding to the level label, and the basic information label.
[0019] Preferably, the objective weighting method includes an entropy weight method, and using the entropy weight method to determine the weight value of each concrete pumping equipment in the equipment group includes: calculating the information entropy of the label feature value of each concrete pumping equipment; and determining the weight value of each concrete pumping equipment in the equipment group based on the calculated information entropy.
[0020] Preferably, the constructing of the group portrait model of the device group is configured to be performed under custom dimensions, and the custom dimensions include a time dimension, an attribute dimension and a behavior dimension.
[0021] The present invention also provides a concrete pumping equipment profiling device, comprising: a memory storing a program that can be run on a processor; and the processor configured to implement any of the above-mentioned concrete pumping equipment profiling methods when executing the program.
[0022] The present invention also provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute any of the above-mentioned concrete pumping equipment profiling methods.
[0023] Through the above technical scheme, the present invention constructs individual portraits and group portraits for concrete pumping equipment, which not only makes full use of the massive data generated by the concrete pumping equipment, but also can intuitively and accurately describe the concrete pumping equipment group, thereby supporting comprehensive and objective comparative analysis of multiple equipment groups.
[0024] Other features and advantages of the present invention will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present invention but do not constitute a limitation of the present invention. In the accompanying drawings:
[0026] Figure 1 It is a flow chart of a concrete pumping equipment profiling method according to an embodiment of the present invention.
[0027] Figure 2 is a schematic diagram of a process for constructing a single-body portrait model of a device in a preferred embodiment of the present invention; and
[0028] Figure 3 It is a flow chart of model training of label feature values based on machine learning algorithm in a preferred embodiment of the present invention. DETAILED DESCRIPTION
[0029] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.
[0030] Figure 1 FIG. 1 is a flow chart of a concrete pumping equipment profiling method according to an embodiment of the present invention. Figure 1 As shown, the concrete pumping equipment portrait method includes the following steps S100-S300.
[0031] Step S100, constructing a single-body portrait model of the concrete pumping equipment represented by its tag set.
[0032] It should be noted that for the convenience of description, the concrete pumping equipment will be referred to as equipment below.
[0033] For step S100, Figure 2 FIG. 1 is a flow chart of constructing a single-body portrait model of a device in a preferred embodiment. Figure 2 As shown, the following steps may be included:
[0034] Step S110, acquiring and cleaning basic data of the concrete pumping equipment.
[0035] The basic data includes the attribute data and working condition data of the equipment. The attribute data includes the model, delivery time, location, customer name, regional concrete characteristics, etc. of the equipment, and the working condition data includes data collected by all equipment sensors and transmitted to the big data platform by the controller through the GPS terminal, such as working time, pumping volume, efficiency, energy consumption, oil temperature, pipe blocking rate, faults, speed fluctuation value, etc. In addition, all the basic data can be stored in the big data platform, and the embodiment of the present invention can directly obtain these basic data from the big data platform.
[0036] Among them, cleaning the basic data may refer to preprocessing the data by means of filling, interpolation, etc. For example, when the equipment is pumping, the time difference between two consecutive oil temperature data is within n minutes (where n is greater than zero), the oil temperature data is linearly interpolated according to time; when the equipment pumping volume decreases abnormally and recovers to the original value before the decrease or exceeds the original value before the decrease within n minutes, the first volume record that recovers to the original value before the decrease or exceeds the original value before the decrease within n minutes is used to cover the first record that starts to decrease; or when the equipment is pumping, the speed fluctuation value has vacancies due to the frequency of data upload, and the vacant value is interpolated by filling from top to bottom.
[0037] Step S120 , classifying the cleaned basic data into labels, including classification into basic information labels, construction performance labels and performance performance labels.
[0038] Preferably, for step S120, the cleaned basic data may be first classified into the following three types of labels:
[0039] 1) Basic information labels, such as: short-meter segment labels, medium-meter segment labels, long-meter segment labels, etc. defined according to the equipment model; domestic chassis labels and imported chassis labels defined according to the equipment chassis model; and vehicle age labels defined according to the equipment delivery time.
[0040] 2) Construction performance labels, such as one or more of a construction location label, a boom posture label, a construction intensity label, a construction frequency label, and a construction extent label.
[0041] Among them, the construction performance labels include the common construction position label, which is defined according to the equipment construction position and is used to support the analysis of the equipment construction environment characteristics; the boom posture label is defined according to the equipment boom data and is used to support the analysis of the equipment construction scene characteristics; the construction intensity label is defined according to the pumping volume and working hours, and is used to reflect the equipment construction intensity and support the life prediction analysis of the whole machine and spare parts; the construction frequency label is defined according to the construction days and working hours, and is used to reflect the equipment utilization rate; the construction breadth label is defined according to the construction area and the number of construction coverage areas, and is used to reflect the breadth of the customer's business scope.
[0042] 3) Performance tags, such as any one or more of the following tags a)-g):
[0043] a) Comprehensive energy consumption level label, pumping system energy consumption level label, idling system energy consumption level label and driving system energy consumption level label set based on energy consumption data. It should be noted that these labels are used to reflect the composition of equipment energy consumption and provide targeted guidance for equipment to reduce energy consumption.
[0044] b) The stroke rate level label, pumping efficiency level label, and suction efficiency level label set based on efficiency data. It should be noted that these labels are used to reflect the pumping performance of the equipment and provide targeted guidance for the efficiency optimization of the equipment.
[0045] c) Engine power level label set based on engine data, which is used to reflect the performance of the equipment engine and provide support for predictive maintenance of the engine.
[0046] d) Pressure shock level label set based on pressure data, which is used to reflect the pumping stability of the equipment and provide overpressure warning and operation guidance suggestions for the equipment.
[0047] e) The hydraulic system oil temperature level label set based on the oil temperature data is used to reflect the hydraulic oil temperature status of the equipment and provide over-temperature warning for the equipment.
[0048] f) Mechanical system fault level labels, hydraulic system fault level labels, electrical system fault level labels, chassis and engine fault level labels and other fault level labels set based on fault warning data. It should be noted that these labels are used to reflect the health status of each system of the equipment and provide support for predictive maintenance of the equipment.
[0049] g) Boom stability level label and pipe blocking rate level label set based on boom data. It should be noted that these two labels are used to reflect the performance of the equipment boom, support the material condition characteristic analysis of the equipment pumping concrete, and provide pipe blocking warning and operation guidance suggestions for the equipment.
[0050] Step S130, determine the parent tag used to show the performance of the equipment, select several tags from the construction performance tag and / or the performance performance tag as the child tags of the parent tag, and set weight values for the child tags. Among them, the basic information tag is irrelevant to the performance of the equipment, and there is no need to consider its tag weight, so it does not participate in the determination of the parent tag and the child tag. Accordingly, in a preferred embodiment, the determined parent tag and the corresponding child tag include any one or more of the following:
[0051] a) The parent tag is determined to be the equipment construction saturation tag, and the corresponding child tags include the construction intensity tag, the construction frequency tag, and the construction breadth tag;
[0052] b) The parent label is determined to be the equipment overall efficiency grade label, and the corresponding child labels include the stroke in place rate grade label, the pumping efficiency grade label and the suction efficiency grade label;
[0053] c) The parent tag is determined to be the overall energy consumption level tag of the equipment, and the corresponding child tags include the comprehensive energy consumption level tag, the pumping system energy consumption level tag, the idle system energy consumption level tag and the driving system energy consumption level tag;
[0054] d) The parent tag is determined to be the equipment boom reliability level tag, and the corresponding child tags include the boom stability level tag, the pipe blocking rate level tag, and the mechanical system failure level tag;
[0055] e) The parent tag is determined to be a power reliability level tag, and the corresponding child tags include the engine power level tag and the chassis and engine failure level tags;
[0056] f) The parent tag is determined to be a hydraulic system reliability level tag, and the corresponding child tags include a hydraulic system oil temperature level tag and a hydraulic system fault level tag.
[0057] In a more preferred embodiment, the parent tag can also be determined as the overall performance tag of the equipment, and the corresponding child tag includes any one or more of the following tags determined as parent tags in a)-f) above: the overall efficiency level tag, the overall energy consumption level tag, the boom reliability level tag, the power reliability level tag, and the hydraulic system reliability level tag. That is, for the determined parent tags, they can also be classified and configured with a parent tag, and the parent tag becomes a child tag of its parent tag.
[0058] Furthermore, after configuring the parent tag and the corresponding child tag, the subjective weighting method can be used to set weights for the child tags of the parent tag, that is, to weight the child tags of each parent tag based on the experience of relevant technical experts.
[0059] Step S140, statistically analyzing the basic data corresponding to each label to obtain the corresponding label feature value, performing model training on the label feature value based on a machine learning algorithm, and obtaining the monomer portrait model represented by the label set.
[0060] Among them, the label characteristic value refers to the label variable used to show the equipment feature classification, and the statistical analysis of the basic data corresponding to each label to obtain the corresponding label characteristic value includes but is not limited to: obtaining the mean, median, first quartile or third quartile of the operating condition data in the basic data corresponding to each label; and performing weighted summation of all the obtained means, medians, first quartiles or third quartiles to obtain the label characteristic value.
[0061] Further, Figure 3 FIG. 1 is a flow chart of model training of label feature values based on a machine learning algorithm in a preferred embodiment of the present invention. Figure 3 As shown, the following steps may be included:
[0062] Step S131, performing outlier processing on the tag feature value.
[0063] In the example taking the pipe blockage rate data as an example, the interquartile range method is used to process the outliers of the pipe blockage rate characteristic value, including: the third quartile of the pipe blockage rate characteristic value is counted as Q3, the first quartile is counted as Q1, the difference between Q3 and Q1 is counted as IQR, and the data greater than (Q3+1.5*IQR) or less than (Q1-1.5*IQR) is removed.
[0064] Step S132: clustering the label feature values after removing outliers using the K-means clustering algorithm to obtain multiple classes and the center point of each class.
[0065] Based on the example of pipe blockage rate data, the pipe blockage rate feature value after removing outliers is clustered with K=3 to obtain three clusters G i and its center point m i .
[0066] Step S133, classifying the label feature value according to the size of the label feature value corresponding to the center point.
[0067] That is, the clusters are marked according to the size of the label characteristic value corresponding to the center point to be classified into multiple data levels. Continuing with the example of the pipe blockage rate data, the clusters are marked according to the size of the pipe blockage rate characteristic value corresponding to the center point of the three classes, and three classifications are obtained: K1 = "mild pipe blockage", K2 = "moderate pipe blockage", and K3 = "severe pipe blockage". In addition, corresponding to the above-mentioned outlier step, the pipe blockage rate characteristic value after cleaning is subjected to an outlier test. If it is greater than (Q3 + 1.5 * IQR) or less than (Q1-1.5 * IQR), K4 = "pipe blockage data abnormality" is also used as the level label of the pipe blockage rate of the equipment.
[0068] Step S134, comparing the label feature value after removing the outliers with each of the center points, and selecting the level label corresponding to the nearest center point as the level label of the corresponding label feature value.
[0069] Step S135, forming the monomer portrait model based on the level label, the corresponding weight set corresponding to the level label and the basic information label.
[0070] Accordingly, the single image model training based on the K-means clustering algorithm is implemented through steps S131 to S135. However, it should be noted that the machine learning algorithm available in the embodiment of the present invention is not limited to the K-means clustering algorithm, and one or more algorithms such as the SVM (Support Vector Machine) algorithm and the decision tree algorithm may also be used.
[0071] Step S200, for an equipment group having multiple concrete pumping equipment, an objective weighting method is used to determine the weight value of each concrete pumping equipment in the equipment group, and the label feature value of the equipment group is determined based on the label feature value of each concrete pumping equipment and the weight value, and a corresponding group portrait model is constructed based on the determined label feature value of the equipment group.
[0072] Among them, the objective weighting method includes but is not limited to the entropy weight method, the coefficient of variation method, the standard deviation method, and the correlation coefficient method. Taking the entropy weight method as an example, using the entropy weight method to determine the weight value of each concrete pumping equipment in the equipment group may include the following steps: calculating the information entropy of the label feature value of each concrete pumping equipment; and determining the weight value of each concrete pumping equipment in the equipment group based on the calculated information entropy. In this regard, in the following, the specific calculation method will be introduced through examples, which will not be repeated here.
[0073] In a more preferred embodiment, the construction of the group portrait model of the device group is configured to be performed under a custom dimension, and the custom dimension includes a time dimension, an attribute dimension, and a behavior dimension. Among them, the time dimension includes a natural time dimension and a pumping volume node dimension. The natural time dimension refers to the monthly / quarterly / half-year / year / full life cycle, and the pumping volume node dimension refers to a time node for each N pumps (N is a positive integer) of the device since delivery. The attribute dimension includes the device model dimension, the regional dimension, and the customer dimension. The behavior dimension refers to starting from a certain event or a certain indicator, including the existence of a certain fault warning, the indicator being within a certain range, and the indicator change level being within a certain range.
[0074] The specific implementation of step S200 is described below by taking an example of determining a group portrait based on the entropy weight method. In this example, the group portrait can be determined by following the steps S1-S7:
[0075] Step S1: Based on the statistical time period of m days and the number of devices under the custom dimension of n, construct the initial evaluation matrix A = (a ij ) m×n .
[0076] Among them, a ij It represents the label feature value of the j-th device on the i-th day, and the label feature value is the weighted sum of the mean, median, first quartile, third quartile, etc. of the operating condition data corresponding to the label.
[0077] Step S2: Standardize the initial evaluation matrix A to obtain a matrix B = (b ij ) m×n .
[0078] Step S3: Calculate the information entropy e of the j-th device label feature value j :
[0079]
[0080] in,
[0081] Step S4: Calculate the weight value w of the tag feature value of each device j :
[0082]
[0083] Among them, d j =1-e j .
[0084] Step S5: According to the weight value calculated in step S4, obtain the label feature value L of the device group:
[0085]
[0086] Among them, X j is the label feature value of the j-th device.
[0087] Step S6: for all tag feature values, repeat steps S1-S5 to obtain the device group tag feature value / single device tag feature value and weight as shown in Table 1 below.
[0088] Table 1 Equipment group portrait labels
[0089]
[0090] in, Get the device group portrait label feature value set L = {L1, L2, L3, ... L q}.
[0091] Step S7: Based on the device group portrait label feature value set, use the label feature value X obtained in step S5 j The single-body portrait model shown in the figure is obtained by labeling feature values L = {L1, L2, L3, ... L q}The group portrait model shown.
[0092] Through this example, it can be seen that in step S200, the entropy weight method is used to profile the concrete pumping equipment group under the custom dimension, that is, the weight value of the individual in the equipment group is obtained based on the information entropy contained in the individual equipment in the equipment group. This method is conducive to building a more objective and accurate group portrait, facilitating the comprehensive comparative analysis of multiple equipment groups, and providing decision makers with systematic and clear data support.
[0093] Step S300: Process the single-device portrait model and / or the group portrait model using a visualization method to generate corresponding single-device portraits and / or device group portraits.
[0094] For example, visualization methods include commonly used chart forms, that is, a single device portrait or a group of devices can be displayed in the form of a chart.
[0095] In summary, through the above steps S100 to S300, the concrete pumping equipment profiling method according to the embodiment of the present invention mainly has the following three effects:
[0096] First, based on the basic data of concrete pumping equipment, the embodiment of the present invention cleans the working condition data generated by massive equipment when it is working, defines labels and sets weights for some sub-labels, and performs weighted summation on the mean, median, first quartile, third quartile, etc. of the working condition data corresponding to the labels to obtain the label feature value, and uses a machine learning algorithm to perform model training on the label feature value, and constructs a single equipment portrait for concrete pumping equipment, which provides a way for relevant technicians and equipment users to fully understand the equipment, so that they can timely grasp the working condition and health status of the equipment by continuously monitoring the changes in the equipment portrait, and at the same time help customers to reasonably schedule and predictively maintain the equipment, reduce equipment downtime, and improve equipment utilization.
[0097] Second, the embodiments of the present invention use, for example, the entropy weight method to obtain the weight value of the individual in the device group under the custom dimension based on the information entropy contained in the individual device, so as to construct a device group portrait under the custom dimension, so that business personnel and management personnel can systematically and accurately understand the overall picture of a certain type of device group, and provide intuitive and clear data support when comparing and analyzing certain types of device groups, which is conducive to assisting decision makers to make more objective and relatively optimal decisions in the personalized marketing of complete equipment and related spare parts.
[0098] Third, the embodiments of the present invention can effectively utilize the massive data generated by concrete pumping equipment to achieve multi-dimensional analysis and dynamic management of equipment and comprehensive and objective comparative analysis of groups, which not only avoids the waste of massive data, but also better meets the needs of customers and technicians.
[0099] Another embodiment of the present invention further provides a concrete pumping equipment profiling device, the device comprising: a memory storing a program that can be run on a processor; and the processor configured to implement any of the above-mentioned concrete pumping equipment profiling methods when executing the program.
[0100] Another embodiment of the present invention further proposes a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute any of the above concrete pumping equipment image processing methods. The machine-readable storage medium includes, but is not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (Flash Memory) or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical storage, magnetic cassette, magnetic disk storage or other magnetic storage devices, and other media that can store program codes.
[0101] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium. The processor may be a general-purpose processor, a special-purpose processor, a conventional processor, a digital signal processor (DSP), a plurality of microprocessors, one or more microprocessors associated with a DSP core, a controller, a microcontroller, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) circuit, any other type of integrated circuit (IC), a state machine, etc.
[0102] The preferred embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. However, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, a variety of simple modifications can be made to the technical solution of the present invention, and these simple modifications all belong to the protection scope of the present invention.
[0103] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.
[0104] In addition, various embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the present invention, they should also be regarded as the contents disclosed by the present invention.
Claims
1. A method for profiling concrete pumping equipment, characterized in that: The concrete pumping equipment portrait method comprises: Construct a single-body portrait model of concrete pumping equipment shown by its tag set, including: Acquiring and cleaning basic data of the concrete pumping equipment; Classifying the cleaned basic data into labels, including classification into basic information labels, construction performance labels and performance performance labels; Determine a parent tag for showing equipment performance, select several tags from the construction performance tag and / or the performance performance tag as child tags of the parent tag, and set weight values for the child tags; and Performing statistical analysis on the basic data corresponding to each label to obtain the corresponding label feature value, performing model training on the label feature value based on a machine learning algorithm, and obtaining the monomer portrait model represented by the label set; For a device group having multiple concrete pumping devices, an objective weighting method is used to determine the weight value of each concrete pumping device in the device group, and a label feature value of the device group is determined based on the label feature value of each concrete pumping device and the weight value, and a corresponding group portrait model is constructed based on the determined label feature value of the device group; and The single-device portrait model and / or the group portrait model are processed by a visualization method to generate corresponding single-device portraits and / or device group portraits.
2. The concrete pumping equipment image processing method according to claim 1, characterized in that: The construction performance label includes any one or more of a construction position label, a boom posture label, a construction intensity label, a construction frequency label, and a construction extent label; The performance label includes any one or more of the following labels: a comprehensive energy consumption level label, a pumping system energy consumption level label, an idle system energy consumption level label and a driving system energy consumption level label set based on energy consumption data; a stroke in place rate level label, a pumping efficiency level label and a suction efficiency level label set based on efficiency data; an engine power level label set based on engine data; Pressure shock level label set based on pressure data; hydraulic system oil temperature level label set based on oil temperature data; Mechanical system fault level labels, hydraulic system fault level labels, electrical system fault level labels, chassis and engine fault level labels and other fault level labels set based on fault warning data; And the boom stability grade label and pipe blocking rate grade label set based on the boom data; The determined parent tag and the corresponding child tag include any one or more of the following: The parent tag is determined to be the equipment construction saturation tag, and the corresponding child tags include the construction intensity tag, the construction frequency tag, and the construction breadth tag; The parent label is determined to be the equipment overall efficiency grade label, and the corresponding child labels include the stroke in place rate grade label, the pumping efficiency grade label and the suction efficiency grade label; The parent tag is determined to be the overall energy consumption level tag of the equipment, and the corresponding child tags include the comprehensive energy consumption level tag, the pumping system energy consumption level tag, the idle system energy consumption level tag, and the driving system energy consumption level tag; The parent tag is determined to be the equipment boom reliability level tag, and the corresponding child tags include the boom stability level tag, the pipe blocking rate level tag, and the mechanical system failure level tag; The parent tag is determined to be a power reliability level tag, and the corresponding child tags include the engine power level tag and the chassis and engine failure level tags; The parent label is determined to be a hydraulic system reliability level label, and the corresponding child labels include a hydraulic system oil temperature level label and a hydraulic system fault level label.
3. The concrete pumping equipment image processing method according to claim 2, characterized in that: The determined parent tags and corresponding child tags also include: The parent tag is determined as the overall performance label of the equipment, and the corresponding child tags include any one or more of the following tags determined as parent tags in claim 2: the overall efficiency level label, the overall energy consumption level label, the boom reliability level label, the power reliability level label and the hydraulic system reliability level label.
4. The concrete pumping equipment image processing method according to claim 1, characterized in that: The statistical analysis of the basic data corresponding to each tag to obtain the corresponding tag feature value includes: Obtain the mean, median, first quartile or third quartile of the operating condition data in the basic data corresponding to each label; and A weighted sum is performed on all the obtained means, medians, first quartiles or third quartiles to obtain the label feature value.
5. The concrete pumping equipment image processing method according to claim 1, characterized in that: The performing model training on the label feature value based on the machine learning algorithm includes: Performing outlier processing on the label feature value; The K-means clustering algorithm is used to cluster the label feature values after removing outliers to obtain multiple classes and the center point of each class; Classify the label feature value according to the size of the label feature value corresponding to the center point; Compare the label feature value after removing the outliers with each of the center points, and select the level label corresponding to the center point closest to the center point as the level label of the corresponding label feature value; and The monomer portrait model is formed based on the level label, the corresponding weight set corresponding to the level label and the basic information label.
6. The concrete pumping equipment image processing method according to claim 1, characterized in that: The objective weighting method includes an entropy weight method, and the entropy weight method is used to determine the weight value of each concrete pumping equipment in the equipment group, including: Calculate the information entropy of the label feature values of each concrete pumping equipment; and The weight value of each concrete pumping equipment in the equipment group is determined based on the calculated information entropy.
7. The concrete pumping equipment image processing method according to claim 1, characterized in that: The construction of the group portrait model of the device group is configured to be performed under custom dimensions, and the custom dimensions include time dimension, attribute dimension and behavior dimension.
8. A concrete pumping equipment imaging device, characterized in that: The concrete pumping equipment imaging device comprises: a memory storing a program that can be executed on the processor; and The processor is configured to implement the concrete pumping equipment imaging method according to any one of claims 1 to 7 when executing the program.
9. A machine-readable storage medium, characterized in that: The machine-readable storage medium stores instructions for enabling a machine to execute the concrete pumping equipment imaging method according to any one of claims 1 to 7.