An intelligent analysis method and system for the usage status of a forklift

Abstract

The present invention relates to the technical field of state monitoring, and specifically to an intelligent analysis method and system for the usage state of a forklift, which includes the following steps: Based on the forklift operation data, characteristic parameters including vehicle speed, fork load, and vehicle body attitude angle are extracted, and the fluctuation ranges of the characteristic parameters are divided into intervals. In the present invention, by dividing the fluctuation range intervals and performing normalization processing, the comparability and stability of the characteristic parameters are optimized. The weighted distance calculation and dynamic weight adjustment are adopted to improve the sample classification accuracy, effectively weaken the coupling interference between parameters in the operation data, combine the density distribution value and the density peak for agglomerative clustering, accurately identify the diversity of operation modes, strengthen the ability to distinguish complex operation behaviors, and realize the precise division of multi-stage states by extracting the weighted eigenvalue in the operation stage and calculating the volatility, enhancing the identification of abnormal operation modes and the systematic ability of operation behavior feature classification, and providing multi-dimensional decision support for equipment management.

Description

Technical Field

[0001] The present invention relates to the technical field of status monitoring, and particularly to an intelligent analysis method and system for the usage status of forklifts. Background Art

[0002] The technical field of status monitoring includes the monitoring and data analysis of the operating status of various devices or systems in order to obtain their operating parameters and behavior patterns. The core content of this technical field is to collect, analyze, and process the operating data of devices or systems to obtain information on device status, fault warning, operating efficiency, etc. Status monitoring technology usually covers aspects such as sensor data acquisition, signal transmission, status data parsing, information storage and processing, forming a systematic data monitoring and management method. By systematically constructing a monitoring platform integrating hardware and software, the real-time grasp and analysis of device status information are realized, providing technical support for the operation management of various devices.

[0003] Among them, the intelligent analysis method for the usage status of forklifts obtains the operating parameters (position, speed, load, etc.) of forklifts in real time through data acquisition devices, and uses data processing methods for status judgment, operating mode recognition, and anomaly discrimination to achieve multi-dimensional status analysis. This technology is based on the system data of modules such as forklift controllers and ECUs to obtain device status, operating efficiency, fault information, etc. Through sensor data acquisition, signal transmission, status parsing and storage, a monitoring platform integrating hardware and software is constructed to realize the real-time monitoring, analysis, module testing, updating, and adjustment of forklift status, providing data support for vehicle maintenance and controller diagnosis.

[0004] The prior art does not fully consider the influence of the fluctuation range interval when processing characteristic parameters, which easily leads to insufficient accuracy of parameter characteristic expression. The sample classification method lacks the ability to dynamically adjust weights, making it impossible to reasonably allocate the importance of different characteristics in operating parameters according to changes in scenarios, affecting the recognition effect in complex scenarios. In the analysis of operating modes, there is a lack of a dynamic clustering method based on density distribution, making it difficult to capture the characteristic distribution laws of diverse operation modes. The division of operating stages mainly relies on static thresholds, unable to effectively analyze the change trends between associated characteristics, restricting the early discovery and accurate determination of abnormal states, and reducing the accuracy and reliability of status monitoring and management. Summary of the Invention

[0005] The purpose of the present invention is to solve the deficiencies existing in the prior art and propose an intelligent analysis method and system for the usage status of forklifts.

[0006] To achieve the above purpose, the present invention adopts the following technical solution: An intelligent analysis method for the usage status of forklifts, including the following steps:

[0007] S1: Based on the forklift operation data, extract characteristic parameters including vehicle speed, fork load, and vehicle body attitude angle, divide the fluctuation range of the characteristic parameters into intervals, calculate the fluctuation means and variances of multiple intervals, normalize all the characteristic parameters, and generate initial characteristic weight values;

[0008] S2: Based on the initial characteristic weight values, calculate the weighted distances of each forklift operation sample point on the vehicle speed sensor and the position change rate, calculate the within-class sample weighted average distance and the between-class sample weighted average distance, normalize and adjust the characteristic weights for the difference between the two, and generate dynamic characteristic weight values;

[0009] S3: Based on the dynamic characteristic weight values, calculate the weighted density distribution values of the forklift operation sample points, determine the density peaks in combination with the vehicle speed and fork load characteristic values, perform merging clustering on the density peaks, re-divide the clustering samples, and calculate the weighted density values for the clustering centers to generate the operation mode clustering results;

[0010] S4: Based on the operation mode clustering results, divide the sample groups according to the operation stages of acceleration, turning, and steady operation in the forklift operation data, extract the weighted characteristic values of the vehicle body attitude angle and the position change rate in each stage, calculate the volatility of the weighted characteristic values and re-adjust the weights to generate the stage-based operation mode classification results;

[0011] S5: Based on the stage-based operation mode classification results, extract the fluctuation value ranges of the fork load and the vehicle body attitude angle in the differentiated operation stages, conduct classification statistics on the fluctuation value distributions of each type of sample, calculate the distribution means and the fluctuation intervals, and generate the forklift usage status classification index set.

[0012] The initial characteristic weight values specifically refer to the characteristic parameter fluctuation range, multi-interval fluctuation means, variances, and normalized characteristic parameters. The dynamic characteristic weight values include weighted distances, within-class and between-class sample average distances, and difference normalization. The operation mode clustering results specifically refer to density peaks, merging clustering, and clustering center weighted density. The stage-based operation mode classification results include stage division, volatility calculation, and weight adjustment. The forklift usage status classification index set specifically refers to the fork load fluctuation range, vehicle body attitude angle fluctuation range, fluctuation value distribution statistics, distribution means, and fluctuation intervals.

[0013] As a further solution of the present invention, the specific steps for obtaining the initial characteristic weight values are as follows:

[0014] S111: Extract the vehicle speed, fork load, and vehicle body attitude angle characteristic parameters in the forklift operation data, analyze the fluctuation range of the characteristic parameters item by item according to the discrete interval rule, divide the data set within each interval into separate regions, and generate the fluctuation mean and the fluctuation variance by performing item-by-item mean operation and variance calculation on the sample points within the regions;

[0015] S112: Normalize the mean value of fluctuations and the variance of fluctuations using the formula:

[0016] ;

[0017] Calculate and generate a normalized feature parameter matrix;

[0018] Wherein, represents the eigenvalue of the th row and th column of the normalized feature matrix, represents the eigenvalue of the th row and th column of the original feature matrix, represents the mean value of the th column feature, represents the variance of the th column feature, represents the smoothing coefficient used to avoid a zero denominator;

[0019] S113: Calculate the mean value between columns of the normalized feature parameter matrix, analyze the initial weight values column by column in sequence, use the matrix column mean vector calculation as the initial weight vector, and generate the initial feature weight values based on the superposition processing of the deviation between the column mean and the multi-eigenvalues of the matrix columns.

[0020] As a further solution of the present invention, the steps for obtaining the dynamic feature weight value are specifically as follows:

[0021] S211: According to the initial feature weight value, perform parameter weighting processing on the operation sample points of the vehicle speed sensor and the position change rate respectively, calculate the weight distribution of the vehicle speed and position change rate parameters for the sample points by calling the initial weight value, and at the same time calculate the cumulative value of the squared differences of each sample point on the weighted distance to obtain a weighted sample point distance set;

[0022] S212: Separate the weighted distance of the intra-class sample points and the weighted distance of the inter-class sample points from the weighted sample point distance set, calculate the weighted mean values of the intra-class and inter-class sample points respectively according to the sample point categories by calling the weighted distance information of the sample points within the set, gradually aggregate the sample category information, and obtain the weighted average distance of the intra-class sample points and the weighted average distance of the inter-class sample points to generate a sample weighted average distance parameter set;

[0023] S213: Based on the sample weighted average distance parameter set, calculate the difference between the weighted average distances of the inter-class and intra-class samples point by point using the formula:

[0024] ;

[0025] Calculate and generate the dynamic feature weight value;

[0026] Among them, represents the dynamic feature weight value, represents the within-class weighted average distance of the th sample, represents the between-class weighted average distance of the th sample, represents the initial feature weight value of the th sample point, is a dynamic adjustment coefficient for smoothing weight calculation, represents the total number of samples, is used to adjust the influence of the sample point weight on the difference normalization, represents the normalized sum of the weight differences.

[0027] As a further solution of the present invention, the steps for obtaining the clustering result of the operation mode are specifically as follows:

[0028] S311: Call the dynamic feature weight value, calculate the weighted density distribution value of the forklift operation sample points, perform weighted density calculation on each sample point through the vehicle speed and the fork load eigenvalue, accumulate the sample point density values point by point, and perform density classification processing in combination with the feature distribution parameters to generate a set of weighted density distribution values;

[0029] S312: Combine the set of weighted density distribution values, and by comparing the distribution values of the vehicle speed and the fork load eigenvalue with the overall density distribution point by point, use the formula:

[0030] ;

[0031] Determine the local density peak of each sample point to generate a set of density peaks;

[0032] Among them, represents the density peak, represents the th weighted density value of the sample point, represents the dynamic weight of the vehicle speed feature, represents the dynamic weight of the fork load feature, is an adjustment parameter for adjusting the smoothness of the weight on the density calculation, represents the dynamic smoothing effect of the fork load weight on the density value, The operation selects the local density maximum value as the density peak;

[0033] S313: Perform merging clustering on the set of density peaks. By calculating the weighted density difference between the density peak of each sample point and other sample points, judge the weighted distance distribution between sample points, and reclassify the clustering result by gradually aggregating the sample point attribution information to generate a clustering sample division;

[0034] S314: Invoke the clustering sample division, calculate the weighted density value of each cluster center, sequentially perform weighted summation on the vehicle speed and forklift load eigenvalue of the sample points belonging to the cluster center, and dynamically adjust the center eigenvalue in combination with the mean distribution of the in-class sample density value to generate the operation mode clustering result.

[0035] As a further solution of the present invention, the specific steps for obtaining the phased operation mode classification result are as follows:

[0036] S411: Invoke the operation mode clustering result, divide the sample group item by item according to the acceleration, turning and steady running stages in the forklift operation data, and sequentially perform operation feature grouping processing on the sample group by extracting the position parameter and attitude angle eigenvalue of the sample points in the running stage to generate the phased sample group division;

[0037] S412: Extract the vehicle body attitude angle and position change rate eigenvalue of each stage in the phased sample group division, sequentially perform weighted processing on the sample point eigenvalue by combining the dynamic feature weight value, and perform weighted cumulative processing on the feature data in the running stage to generate the phased weighted eigenvalue set;

[0038] S413: Calculate the volatility of the phased weighted eigenvalue set, and by readjusting the weight, use the formula:

[0039] ;

[0040] Calculate and generate the readjusted weight parameter;

[0041] wherein, represents the readjusted weight parameter, represents the volatility of the weighted eigenvalue set, represents the mean of the weighted eigenvalue set, represents the minimum volatility threshold, which is used to adjust the flexibility and sensitivity of the weight, and increases the response ability of the formula to small fluctuations;

[0042] S414: Invoke the readjusted weight parameter, re-evaluate and classify the weighted eigenvalue of each stage in combination with the phased sample group division information, and sequentially classify the readjusted classification features in the running stage by calculating the comprehensive feature parameter of the running stage to generate the phased operation mode classification result.

[0043] As a further solution of the present invention, the specific steps for obtaining the forklift use state classification index set are as follows:

[0044] S511: Invoke the classification result of the phased operation mode, extract the fluctuation value ranges of the forklift load and the vehicle body attitude angle in the differential operation stage. By sequentially screening the dynamic change values of the forklift load and the vehicle body attitude angle parameters for the sample points within the operation stage, gradually classify and summarize the fluctuation value ranges of the sample points to generate a set of sample fluctuation values for the operation stage;

[0045] S512: Extract the fluctuation value distributions of each type of sample in the set of sample fluctuation values for the operation stage. By sequentially calculating the interval characteristics of the fluctuation values of the samples and combining the distribution ranges of each type of sample for classification statistics, gradually classify and summarize the characteristic parameters of the sample point fluctuation ranges to generate a classification statistical result of the sample fluctuation value distributions;

[0046] S513: Invoke the classification statistical result of the sample fluctuation value distributions, calculate the fluctuation mean value and the fluctuation interval of each type of sample distribution. By performing weighted operations on the fluctuation value characteristics point by point and using the formula:

[0047] ;

[0048] Calculate and generate the sample distribution mean value and the fluctuation interval;

[0049] Wherein, represents the weighted fluctuation value of the sample distribution, represents the fluctuation value of the th sample, represents the mean value of the fluctuation values of all sample points, represents the dynamic weight of the th sample point, is the non - linear influence of the smoothing coefficient on adjusting the weight, represents the smoothing effect of the dynamic adjustment weight on the distribution characteristics,

[0050] S514: Invoke the sample distribution mean value and the fluctuation interval, and in combination with the classification statistical results within the operation stage, gradually classify and summarize the fluctuation value ranges and the mean value characteristic parameters of each type of sample, recalculate the classification indicators for the dynamic change parameter characteristics of each type of sample in the operation stage to generate a set of classification indicators for the forklift usage status.

[0051] An intelligent analysis system for forklift usage status, the intelligent analysis system for forklift usage status is used to execute the above - mentioned intelligent analysis method for forklift usage status, and the system includes:

[0052] The feature extraction module, based on the forklift operation data, extracts feature values from the vehicle speed, forklift load, and vehicle body attitude angle, calculates the fluctuation mean value and variance, and normalizes the feature values to obtain the initial feature weight values;

[0053] The dynamic weight adjustment module calculates the weighted distance of sample points based on the initial feature weight values, obtains the weighted average distances of the samples inside and outside the class, performs normalization adjustment through the difference, updates the weight of each feature, and generates dynamic feature weight values;

[0054] The clustering analysis module calculates the weighted density distribution of the forklift operation sample points based on the dynamic feature weight values, determines the density peaks, performs the merging clustering of the density peaks, calculates the weighted density of the clustering centers, and generates the operation mode clustering results;

[0055] The operation mode classification module divides the forklift operation data into acceleration, turning, and stable operation stages based on the operation mode clustering results, extracts the weighted features of the vehicle body attitude angle and the position change rate, calculates the volatility and adjusts the weights, and generates the stage operation mode classification results;

[0056] The state classification index generation module extracts the fluctuation ranges of the forklift load and the vehicle body attitude angle in the differential operation stages based on the stage operation mode classification results, statistically analyzes the distribution of the fluctuation values, calculates the distribution mean and interval, and generates the forklift usage state classification index set.

[0057] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0058] In the present invention, by extracting feature parameters such as vehicle speed, forklift load, and vehicle body attitude angle, dividing the fluctuation range intervals and performing normalization processing, the comparability and stability of the feature parameters are optimized. The use of weighted distance calculation and dynamic weight adjustment improves the sample classification accuracy and effectively weakens the coupling interference between parameters in the operation data. Combining the density distribution value and the merging clustering of the density peaks accurately identifies the diversity of operation modes and enhances the ability to distinguish complex operation behaviors. By extracting the weighted feature values in the operation stages and calculating the volatility, the precise division of multi-stage states is realized, the recognition of abnormal operation modes and the systematic ability of operation behavior feature classification are enhanced, and multi-dimensional decision-making support is provided for equipment management. Description of the Drawings

[0059] Figure 1 It is a schematic diagram of the working process of the present invention;

[0060] Figure 2 It is a flow chart of the steps for obtaining the initial feature weight values of the present invention;

[0061] Figure 3 It is a flow chart of the steps for obtaining the dynamic feature weight values of the present invention;

[0062] Figure 4 It is a flow chart of the steps for obtaining the operation mode clustering results of the present invention;

[0063] Figure 5Flow chart of the steps for obtaining the classification results of the phased operation mode of the present invention;

[0064] Figure 6 Flow chart of the steps for obtaining the classification index set of the usage status of the forklift of the present invention. Detailed implementation manners

[0065] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. 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.

[0066] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present invention. In addition, in the description of the present invention, "a plurality of" means two or more unless otherwise specifically defined.

[0067] Embodiment 1

[0068] Please refer to Figure 1 , the present invention provides a technical solution: an intelligent analysis method for the usage status of a forklift, including the following steps:

[0069] S1: Based on the forklift operation data, extract characteristic parameters including vehicle speed, fork load, and vehicle body attitude angle, divide the fluctuation range of the characteristic parameters into intervals, calculate the fluctuation mean and variance of multiple intervals, normalize all the characteristic parameters, and generate an initial characteristic weight value;

[0070] S2: Based on the initial characteristic weight value, calculate the weighted distance of each forklift operation sample point on the vehicle speed sensor and the position change rate, calculate the weighted average distance of the samples within the class and the weighted average distance of the samples between classes, normalize the difference between the two to adjust the characteristic weight, and generate a dynamic characteristic weight value;

[0071] S3: Based on the dynamic characteristic weight value, calculate the weighted density distribution value of the forklift operation sample point, determine the density peak in combination with the vehicle speed and fork load characteristic values, perform merging clustering on the density peak, re-divide the clustering samples, calculate the weighted density value of the clustering center, and generate the operation mode clustering result;

[0072] S4: Based on the operation mode clustering results, divide the sample groups according to the running stages of acceleration, turning, and stable running in the forklift operation data. Extract the weighted eigenvalue of the vehicle body attitude angle and the position change rate in each stage, calculate the volatility of the weighted eigenvalue, and readjust the weights to generate the classification results of the stage operation mode.

[0073] S5: Based on the classification results of the stage operation mode, extract the fluctuation value ranges of the forklift load and the vehicle body attitude angle in the differential operation stages, conduct a classification statistics on the fluctuation value distribution of each type of sample, calculate the distribution mean and the fluctuation interval, and generate the forklift usage status classification index set.

[0074] The initial feature weight values specifically include the feature parameter fluctuation range, the multi-interval fluctuation mean, the variance, and the normalized feature parameters. The dynamic feature weight values include the weighted distance, the average distance between samples inside and outside the class, and the difference normalization. The operation mode clustering results specifically refer to the density peak, the merged clustering, and the weighted density of the clustering center. The classification results of the stage operation mode include the stage division, the volatility calculation, and the weight adjustment. The forklift usage status classification index set specifically includes the forklift load fluctuation range, the vehicle body attitude angle fluctuation range, the fluctuation value distribution statistics, the distribution mean, and the fluctuation interval.

[0075] Please refer to Figure 2 , and the specific steps for obtaining the initial feature weight values are as follows:

[0076] S111: Extract the characteristic parameters of the vehicle speed, the forklift load, and the vehicle body attitude angle in the forklift operation data. Analyze the fluctuation range of the characteristic parameters item by item according to the discrete interval rule, divide the data set in each interval into separate regions, and generate the fluctuation mean and the fluctuation variance by performing the item-by-item mean operation and variance calculation on the sample points in the region.

[0077] Sample and extract the vehicle speed, the forklift load, and the vehicle body attitude angle parameters recorded during the forklift operation process respectively according to the time series. Divide the extreme values, the mean value, and the fluctuation range in the time series sampling data into intervals. For the data in each divided interval, calculate the fluctuation mean in each interval, and calculate the fluctuation variance in each interval. Combine the fluctuation mean and the fluctuation variance in each interval to gradually construct a fluctuation parameter set with intervals as the unit, and generate the fluctuation mean and the fluctuation variance.

[0078] More specifically, for characteristic parameters such as the vehicle speed, fork load, vehicle body attitude angle, and steering angle torque of a forklift, it is necessary to divide the fluctuation range of the characteristic parameters into intervals. When collecting data, first define the basic value range of each parameter. For example, the vehicle speed is generally between 0 - 25 km / h, the fork load can vary within 0 - 5000 kg, the value of the vehicle body attitude angle is usually between -10° and 10°, and the steering angle torque is between 0 - 200 Nm. Then, analyze the time series data of each parameter, calculate the maximum value, minimum value, and mean value within each time segment. For example, within a certain minute, the vehicle speed may vary between 12 km / h and 18 km / h, then its mean value is (12 + 18) / 2 = 15 km / h. The variance calculation requires summing the deviations of each data point from the mean value and dividing by the total number of data points. For example, if the vehicle speed data points within this time period are 12, 14, 16, 18 km / h respectively, then its variance is [((12 - 15)² + (14 - 15)² + (16 - 15)² + (18 - 15)²) / 4] = 4.5; for the fork load, if the data within a certain time period is 500 kg, 800 kg, 1200 kg, 1000 kg, then the mean value is (500 + 800 + 1200 + 1000) / 4 = 875 kg, and the variance calculation method is the same; for the vehicle body attitude angle, assuming the values within a certain time period are -3°, 0°, 2°, 5°, then the mean value is (-3 + 0 + 2 + 5) / 4 = 1°, and the variance is [((-3 - 1)² + (0 - 1)² + (2 - 1)² + (5 - 1)²) / 4] = 6; through the above calculations, the fluctuation of each characteristic parameter can be quantified to ensure the accuracy of subsequent normalization processing.

[0079] S112: Normalize the fluctuation mean and fluctuation variance, using the formula:

[0080] ;

[0081] Calculate and generate a normalized characteristic parameter matrix;

[0082] Among them, represents the eigenvalue of the th row and th column of the normalized characteristic matrix, represents the eigenvalue of the th row and th column of the original characteristic matrix, represents the mean value of the th column characteristic, represents the variance of the th column characteristic, represents the smoothing coefficient used to avoid a zero denominator;

[0083] Formula:

[0084] ;

[0085] The advantage of the formula is that by introducing a smoothing parameter it avoids calculation errors caused by a zero denominator, and at the same time realizes a normalization operation by combining the mean and variance of each feature, enabling calculations of different features on the same scale and enhancing the comparability between data.

[0086] Detailed explanation of the formula and the derivation process of formula calculation: For each element of the feature parameter matrix , when collecting, the sampling data per second in the forklift operation record is used as input data. By calculating the mean and variance for the same column of data (i.e., the same feature parameter column), where is the total number of data samples, the normalization value of each element is calculated using the mean and variance. For example, for , , , , substitute them into the formula:

[0087] ;

[0088] Calculate the normalization values of all elements in the matrix in sequence to generate a normalized feature parameter matrix.

[0089] This result shows that through the normalized feature parameter matrix, the weights between features can be further calculated and the influence degree of each feature on the final target can be analyzed.

[0090] S113: Perform column - mean calculation on the normalized feature parameter matrix, analyze the initial weight values column by column in sequence, use the matrix column - mean vector calculation as the initial weight vector, and generate the initial feature weight values based on the superposition processing of the deviation between the column mean and the multi - eigenvalue of the matrix columns.

[0091] For the normalized data values of each column in the matrix, perform a summation operation on the normalized values in sequence and take the mean. Use the formula to calculate the column - mean weight, where is the number of data points in the column. Combining the column - mean weight, calculate the weight adjustment amount through the deviation value between the column mean and each eigenvalue in the matrix. The adjustment amount is realized through the formula , where is the matrix normalization value, is the column - mean weight. Gradually generate the adjusted weight vector. After completing the weight adjustment, perform column - accumulation operation and normalization processing in combination with the adjusted weight vector to generate the initial feature weight values.

[0092] Please refer toFigure 3 , the steps for obtaining the dynamic feature weight value are specifically as follows:

[0093] S211: According to the initial feature weight value, perform parameter weighting processing on the operation sample points of the vehicle speed sensor and the position change rate respectively. By calling the initial weight value, calculate the weight distribution of the vehicle speed and position change rate parameters for the sample points, and at the same time calculate the cumulative value of the squared differences of each sample point on the weighted distance to obtain the weighted sample point distance set;

[0094] By extracting the real-time monitoring data of the vehicle speed sensor and the time series data corresponding to the position change rate, establish the weight mapping of each sample point, and call the initial weight value to calculate the cumulative value of the squared differences of each sample point on the weighted distance point by point. The extraction process of the vehicle speed data is completed through continuous monitoring of the sensor, and the vehicle speed change data is recorded at 5-second intervals to generate a sample point sequence, where each sample point data is the vehicle speed value at the current moment. The position change rate is calculated by recording the spatial coordinate changes of each sample point, and using the formula to calculate the position change rate, is the spatial displacement between the current sample point and the previous sample point, is the time interval between two recordings. After obtaining the vehicle speed and position change rate, calculate the weighted distance according to the initial weight value according to the weight distribution rule, and use the formula , is the initial weight value, is the vehicle speed or position change rate value, is the overall sample mean. The squared differences are accumulated point by point to form a complete sample distance set, and the weighted sample point distance set is obtained.

[0095] S212: From the weighted sample point distance set, separate the within-class sample point weighted distance and the between-class sample point weighted distance. By calling the weighted distance information of the sample points within the set, calculate the weight means of the within-class and between-class sample points respectively according to the categories to which the sample points belong, and gradually aggregate the sample category information to obtain the within-class sample weighted average distance and the between-class sample weighted average distance, and generate a sample weighted average distance parameter set;

[0096] By analyzing the category labels of the sample points, calculate the within-class weighted distance according to the weighted distance between the sample points within the same category. Call the category label field of the sample points to extract the sample point data of each category in a grouped form, and use the average value of the weighted distances between each pair of samples after grouping as the within-class weighted average distance. The formula is expressed as , where is the number of pairs of category sample points, is the sample point and is the weighted distance, represents the category For the sample set, the weighted distance between classes is calculated using the weighted distance between sample points of different classes. According to the classes of the sample points, perform cross-class averaging. By the above calculations, the weighted average distance within classes and the weighted average distance between classes of samples are obtained respectively, generating a parameter set of the weighted average distance of samples.

[0097] S213: Based on the parameter set of the weighted average distance of samples, by calculating the difference between the weighted average distance between classes and within classes of samples point by point, using the formula:

[0098] ;

[0099] calculate and generate the dynamic feature weight value;

[0100] where, represents the dynamic feature weight value, represents the weighted average distance within classes of the th sample, represents the weighted average distance between classes of the th sample, represents the initial feature weight value of the th sample point, is the dynamic adjustment coefficient for smoothing the weight calculation, represents the total number of samples, is used to adjust the influence of the sample point weight on the normalization of the difference, represents the normalized sum of the weight differences.

[0101] Formula:

[0102] ;

[0103] The advantage of the formula is that by introducing the square root and the dynamic adjustment coefficient 1, combined with the influence of the sample weight , the weighted difference between classes and within classes is smoothly adjusted, and at the same time, the sensitivity and calculation accuracy of the result are improved through point-by-point normalization calculation.

[0104] Detailed explanation of the formula and the derivation process of the formula calculation:

[0105] First, extract the weighted average distance within classes and the weighted average distance between classes from the parameter set of the weighted average distance of samples. These parameters are obtained through the above calculations. Let the initial total number of samples be , where the initial weight values are all distributed in the range of 0.1 to 1.0. Through the formula , assuming the dynamic adjustment coefficient , the following are the parameter values:

[0106] ;

[0107] ;

[0108] ;

[0109] Substitute point by point into the formula for calculation:

[0110] For the first sample point:

[0111] ;

[0112] Repeat the above process to calculate the contribution value of each sample point and accumulate item by item:

[0113] Finally:

[0114] ;

[0115] This result indicates that the dynamic feature weights of the sample points are adjusted to the normalized weighted result after overall smoothing, and the dynamic feature weight values are obtained through calculation , providing a standardized weight value benchmark for subsequent calculations and further analysis.

[0116] Please refer to Figure 4 , and the specific steps for obtaining the operation mode clustering result are as follows:

[0117] S311: Call the dynamic feature weight value, calculate the weighted density distribution value of the forklift operation sample points, perform weighted density calculation on each sample point through the vehicle speed and fork load characteristic values, accumulate the sample point density values point by point, and combine the feature distribution parameters for density classification processing to generate a set of weighted density distribution values;

[0118] Extract the vehicle speed and fork load parameter characteristics in the forklift operation data, calculate the weighted density distribution value point by point, and through calling the feature weight value, perform normalization processing on the vehicle speed parameter value and the fork load parameter value respectively according to the specific weight distribution rule. After normalization, substitute the feature value of each sample point into the weighted density calculation formula , where is the density value of the sample point , is the th feature value of the sample point , is the weight value of the corresponding feature. By accumulating the weighted contribution values of all features of the sample points, a set of weighted density values for each sample point is obtained. The accumulation of the density distribution values needs to combine the overall density range of the sample points. By normalizing each density value, the density values are distributed within a unified interval, and then classified and processed in combination with the density range of the feature distribution. By calculating the median, maximum, and minimum values of the density value distribution interval and setting the segmentation threshold, the classification processing of the density distribution set is completed, and a set of weighted density distribution values is generated.

[0119] S312: Combine the set of weighted density distribution values. By comparing the distribution values of the vehicle speed and the forklift load feature values with the overall density distribution point by point, using the formula:

[0120] ;

[0121] Determine the local density peak of each sample point to generate a set of density peaks;

[0122] Among them, represents the density peak, represents the th weighted density value of the sample point, represents the dynamic weight of the vehicle speed feature, represents the dynamic weight of the forklift load feature, is an adjustment parameter used to adjust the smoothness of the weight on density calculation, represents the dynamic smoothing effect of the forklift load weight on the density value, The operation selects the local density maximum value as the density peak;

[0123] Formula:

[0124] ;

[0125] The benefit of the formula is that by introducing the adjustment parameter in the square root denominator, the influence of the forklift load feature weight is smoothed, making the calculation of the density peak more accurate. At the same time, by screening the local density through the maximum value operation, the significance of the result is improved;

[0126] Detailed explanation of the formula and the derivation process of the formula calculation:

[0127] Suppose in a set of sample data, the calculated value of is respectively, assume the corresponding vehicle speed weights are respectively, the forklift load weights are respectively, and the adjustment coefficient

[0128] ;

[0129] ;

[0130] ;

[0131] ;

[0132] ;

[0133] Maximum value is ;

[0134] The result shows that after the density peak is calculated, the density peak of the sample point is the highest, indicating that its characteristic weight contributes more to the weighted density of vehicle speed and fork load, and can be used as an important node in the agglomerative clustering.

[0135] S313: Perform agglomerative clustering on the density peak set. By calculating the weighted density difference between the density peak of each sample point and other sample points, judge the distribution of the weighted distance between sample points, and reclassify the clustering result by gradually aggregating the attribution information of sample points to generate a clustering sample partition;

[0136] Extract the weighted density peak of the sample point. By comparing the difference in weighted density values and the difference in spatial distribution between sample points, use the calculation formula to obtain the weighted density difference between sample points, judge whether the difference is less than the threshold. The threshold is calculated based on 20% of the maximum density difference. For example, if the maximum density difference is set to 8, the threshold is 1.6. Aggregate the sample points with a difference less than the threshold. By further calculating the weighted average distance from the clustering center to the sample point, update the density value of the clustering center point, and loop the above steps again to adjust the attribution category of the sample point and form a clustering result to generate a clustering sample partition.

[0137] S314: Call the clustering sample partition, calculate the weighted density value of each clustering center, and dynamically adjust the center characteristics by sequentially weighted summing the vehicle speed and fork load characteristic values of the sample points belonging to the clustering center, and combining the mean distribution of the within-class sample density values to generate an operation mode clustering result.

[0138] Select the sample point set of each clustering center, extract the vehicle speed characteristics and fork load characteristics of the sample points belonging to each clustering center, calculate the weighted density value by weighted averaging the vehicle speed characteristic values and combining the density value and weight parameter of the fork load characteristic, and then combine the reciprocal of the square of the distance from the density value to the clustering center to calculate the correction value of the density center, perform weighted density correction on each clustering center and generate an operation mode clustering result.

[0139] Please refer toFigure 5 Specifically, the steps for obtaining the classification result of the phased operation mode are as follows:

[0140] S411: Invoke the clustering result of the operation mode. Divide the sample groups item by item according to the acceleration, turning, and steady running stages in the forklift operation data. By extracting the position parameters of the sample points and the attitude angle eigenvalue in the running stage, perform running feature grouping processing on the sample groups in sequence to generate the phased sample group division;

[0141] Gradually analyze the characteristic parameters of the acceleration stage. Invoke the vehicle speed change sequence recorded in the forklift operation data, and extract the positions where the change values in the vehicle speed data are greater than a certain set threshold as the judgment criteria for the acceleration stage. Classify the sample points higher than the vehicle speed change threshold into the acceleration stage, and extract the characteristic of the vehicle body attitude angle recorded in the corresponding sample points. When extracting the turning stage, by invoking the direction change parameter and the fork load change value recorded in the forklift operation data, combine the time series analysis of each sample point to analyze the time period when the direction change angle is greater than a certain angle threshold, and classify the corresponding sample points into the turning stage. At the same time, extract the vehicle body attitude angle and the position change rate eigenvalue of this stage. The steady running stage is obtained by excluding the sample points in the acceleration and turning stages, calculating the vehicle speed volatility and the fork load volatility of the remaining sample points, and demarcating the low volatility sample points according to the volatility mean and standard deviation as the steady stage sample group. Gradually aggregate the data in the above divided stages and establish the sample group of the running stage to generate the phased sample group division.

[0142] S412: Extract the vehicle body attitude angle and the position change rate eigenvalue of each stage in the phased sample group division. By combining the dynamic feature weight value, perform point-by-point weighted processing on the sample point eigenvalues, and perform weighted cumulative processing on the feature data of the running stage to generate the phased weighted eigenvalue set;

[0143] For the weighted processing of the acceleration stage, calculate and extract the absolute difference between the vehicle body attitude angle and the position change rate of the sample points, and sequentially assign the dynamic feature weight value to the two parameters. Adjust the weight ratio according to the distribution mean of the two features, and accumulate the weighted results of the two to obtain the weighted eigenvalue of the acceleration stage. For the turning stage, invoke the direction change data and the fork load characteristic data of the sample points, and calculate the weighted ratio of the dynamic feature weight value and the direction change value as the weighted feature component of the turning stage. Sequentially accumulate this feature component with the fork load characteristic. The feature weighting of the steady running stage uses the weighted product of the absolute fluctuation values of the vehicle speed and the position change rate of the sample points and the dynamic feature weight value, and combines the time series of the running stage to calculate the cumulative result. Gradually accumulate and classify the eigenvalue distributions of the three stages to generate the phased weighted eigenvalue set.

[0144] S413: Calculate the volatility of the phased weighted eigenvalue set, and by readjusting the weights, use the formula:

[0145] ;

[0146] Calculate and generate the re-adjusted weight parameters;

[0147] Wherein, represents the re-adjusted weight parameter, represents the volatility of the weighted eigenvalue set, represents the mean value of the weighted eigenvalue set, represents the minimum volatility threshold, which is used to adjust the flexibility and sensitivity of the weight, and increases the response ability of the formula to small fluctuations;

[0148] Formula:

[0149] ;

[0150] The benefit of the formula is that by introducing the minimum volatility threshold , the sensitivity of the calculation is increased. Especially when the change of eigenvalues is small, it can respond to tiny changes, enhancing the model's ability to identify outliers. Such a setting makes the weight adjustment more precise, applicable to unstable data sources, and improves the practical application value of the model. Formula details and formula calculation derivation process: Assume that the volatility obtained from the data is 0.03, the mean value is 0.01, and the minimum volatility threshold is set to 0.05, and these parameters are adjusted according to the actual situation. These values can be obtained through actual data monitoring. For example, the volatility can be calculated through the standard deviation within a certain time period, and the mean value is the average value of the data within the same time period.

[0151] ;

[0152] ;

[0153] ;

[0154] ;

[0155] ;

[0156] The result shows that the adjusted weight parameter It is approximately 1.76, which means that according to the current volatility and mean, by adjusting the minimum volatility threshold, we can obtain a relatively high weight parameter, which can make a more sensitive response to small fluctuations when dealing with actual data, increasing the adaptability of the model to small changes in the data. Further derivation shows that this weight can be directly used in subsequent calculations or model adjustments to optimize the model performance.

[0157] S414: Call the re-adjusted weight parameter, re-evaluate and classify the weighted eigenvalue of each stage in combination with the phased sample group division information, calculate the comprehensive feature parameter of the operation stage through classification calculation, and classify the re-adjusted classification features of the operation stage in turn to generate the phased operation mode classification result.

[0158] By extracting the sample point feature distribution and corresponding dynamic weights of the acceleration, turning, and steady running stages one by one, calculate the weighted mean of the weighted feature distribution in the acceleration stage, judge the classification attribution of each feature point according to the mean interval, extract the direction change and forklift load fluctuation eigenvalue in the turning stage, and classify and aggregate the samples in combination with the weighted distribution in turn. For the steady running stage, calculate the weight normalization distribution of the vehicle speed fluctuation value of the sample points, and gradually re-classify and aggregate the sample points in the three stages to generate the phased operation mode classification result.

[0159] Please refer to Figure 6 , and the steps for obtaining the forklift usage status classification index set are specifically as follows:

[0160] S511: Call the phased operation mode classification result, extract the fluctuation value range of the forklift load and the vehicle body attitude angle in the differential operation stage, and gradually classify and summarize the dynamic change values of the forklift load and the vehicle body attitude angle parameters of the sample points in the operation stage to generate the operation stage sample fluctuation value set;

[0161] First, perform point-by-point analysis on the forklift load value and attitude angle parameter recorded for each sample point in the operation stage. Calculate the fluctuation range through the time series fluctuation change of each group of sample points. Use the sliding window method to extract the local maximum and minimum values based on the time series characteristics and calculate the difference to obtain the fluctuation range value. After completing the extraction of the fluctuation range value, re-classify the samples in combination with the differential operation stage of the sample group classification, and store the extracted fluctuation range corresponding to the classified operation stage sample points to ensure that each group of fluctuation value ranges corresponds to the operation stage characteristics; after all operation stage sample groups are classified and stored, gradually accumulate the fluctuation ranges of all sample groups, perform upper and lower quartile statistics on the classified fluctuation range values and calculate the overall fluctuation interval, and re-evaluate the fluctuation value difference characteristics of each operation stage group through the calculated fluctuation interval to generate the operation stage sample fluctuation value set.

[0162] S512: Extract the fluctuation value distribution of each type of sample in the sample fluctuation value set during the operation stage. Through calculating the interval characteristics of the sample fluctuation values in sequence and combining with the distribution range of each type of sample for classification statistics, gradually classify and summarize the characteristic parameters of the sample point fluctuation range to generate the classification statistics result of the sample fluctuation value distribution.

[0163] First, conduct distribution statistics on the set of fluctuation values in each type of sample during the operation stage. Summarize the overall distribution of each type of sample by calculating basic statistics such as the mean, median, and standard deviation in the sample distribution. Then, combine the upper and lower quartiles of the fluctuation values for classification statistics of the distribution, and gradually refine the classification intervals based on the skewness characteristics of the distribution. For example, samples with fluctuation values higher than two standard deviations of the overall distribution mean are defined as the high-fluctuation value interval; samples with fluctuation values lower than one standard deviation of the overall mean are defined as the low-fluctuation value interval, and the remaining samples are divided into the medium-fluctuation value interval. After completing the classification of the distribution, count the proportion of the number of samples in each interval for the sample fluctuation distribution of each type of operation stage respectively, and analyze its contribution to the overall distribution category by category. Finally, record the classified statistical results in combination with the upper and lower boundaries of the fluctuation interval, and gradually accumulate the statistical characteristics of each type to generate the classification statistics result of the sample fluctuation value distribution.

[0164] S513: Invoke the classification statistics result of the sample fluctuation value distribution, calculate the fluctuation mean and fluctuation interval of each type of sample distribution, through weighted operation on each point of the fluctuation value characteristics and using the formula:

[0165] ;

[0166] Calculate and generate the sample distribution mean and fluctuation interval;

[0167] Among them, represents the weighted fluctuation value of the sample distribution, represents the th fluctuation value of the sample, represents the mean of the fluctuation values of all sample points, represents the th dynamic weight of the sample point, is the non-linear influence of the smoothing coefficient to adjust the weight, represents the smoothing effect of the dynamic adjustment weight on the distribution characteristics, represents the total number of samples;

[0168] Formula:

[0169] ;

[0170] The benefit of the formula is that by introducing the dynamic weight and the non-linear smoothing adjustment coefficient 2. It can effectively balance the influence of sample weights on the distribution of fluctuation values, further enhance the correlation between sample weights and fluctuation characteristics, and thus improve the detail level of the fluctuation value distribution.

[0171] Detailed Explanation of the Formula and Derivation Process of Formula Calculation:

[0172] First, define each parameter: Represents the weighted fluctuation value of the sample distribution. Represents the fluctuation value of the th sample. Represents the mean of the fluctuation values of all sample points. Represents the dynamic weight of the th sample point. 2 is the smoothing coefficient to adjust the non - linear influence of the weight.

[0173] Description of the Parameter Acquisition Method:

[0174] : Obtained by monitoring the fluctuation value range of sample points. The difference between the local maximum and minimum values of each group of sample points is calculated using a sliding window, and the difference is taken as the fluctuation value.

[0175] : Obtained by calculating the mean of the fluctuation values of sample points. The calculation formula is , where is the total number of samples.

[0176] : The dynamic weight is extracted through the operation - stage characteristics of sample points. The weight value is dynamically adjusted according to the importance of the operation stage, and can be determined by calculating the proportion of the comprehensive influence degree of the forklift load and the vehicle body attitude angle in the operation stage.

[0177] 2: The smoothing coefficient is obtained through experimental calibration, and the adjustment value is dynamically set according to the influence range of the sample fluctuation value on the overall feature distribution.

[0178] Parameter Assignment and Formula Derivation:

[0179] Suppose there are 5 sample points, with fluctuation values , , , , , weights , , , , , smoothing coefficient .

[0180] 1. Calculate the mean value 。

[0181] 2. Calculate the weighted fluctuation value point by point:

[0182] For : 。

[0183] For : 。

[0184] For : 。

[0185] For : 。

[0186] For : 。

[0187] 3. Calculate the denominator part:

[0188] 。

[0189] 。

[0190] 。

[0191] 。

[0192] 。

[0193] 4. Substitute into the formula for calculation:

[0194] 。

[0195] The result shows that the weighted fluctuation value of the sample distribution is 0.383, which is related to the calculated mean value, weight, and fluctuation range, indicating that the fluctuation characteristics in the operation stage have completed the weighted distribution process.

[0196] S514: Call the mean value and fluctuation interval of the sample distribution, gradually classify and summarize the fluctuation value range and mean characteristic parameters of each type of sample in combination with the classification statistical results in the operation stage, recalculate the classification indicators for the dynamic change parameter characteristics of each type of sample in the operation stage, and generate the classification index set for the forklift usage status.

[0197] First, the fluctuation mean and interval characteristics are respectively corresponding to each type of operation stage sample, and the dynamic change characteristics of each type of operation stage are divided through the mean characteristics of the distribution and the upper and lower interval boundaries; after gradually classifying the fluctuation ranges of each type of sample, the distribution characteristics of the operation characteristic parameters of each type of sample are calculated by combining the change values of the forklift fork load and the vehicle body attitude angle; after completing the classification of all characteristic parameters, the dynamic change characteristics of all operation stage samples are gradually summarized to generate the forklift usage status classification index set.

[0198] An intelligent analysis system for forklift usage status, which is used to execute the above intelligent analysis method for forklift usage status. The system includes:

[0199] The feature extraction module extracts eigenvalue from vehicle speed, forklift fork load, and vehicle body attitude angle based on forklift operation data, calculates the fluctuation mean and variance, normalizes the eigenvalue, and obtains the initial feature weight value;

[0200] The dynamic weight adjustment module calculates the weighted distance of sample points based on the initial feature weight value, obtains the weighted average distance of samples inside and outside the class, performs normalization adjustment through the difference, updates the weight of each feature, and generates the dynamic feature weight value;

[0201] The clustering analysis module calculates the weighted density distribution of forklift operation sample points based on the dynamic feature weight value, determines the density peak, performs the merging clustering of the density peak, calculates the weighted density of the clustering center, and generates the operation mode clustering result;

[0202] The operation mode classification module divides the forklift operation data into acceleration, turning, and smooth operation stages based on the operation mode clustering result, extracts the weighted features of the vehicle body attitude angle and the position change rate, calculates the volatility and adjusts the weight, and generates the stage operation mode classification result;

[0203] The status classification index generation module extracts the fluctuation ranges of the forklift fork load and the vehicle body attitude angle in different operation stages based on the stage operation mode classification result, statistically analyzes the distribution of the fluctuation values, calculates the distribution mean and interval, and generates the forklift usage status classification index set.

[0204] Embodiment 2

[0205] This embodiment is basically the same as Embodiment 1, except that: compared with Embodiment 1, in this embodiment, step S1 further includes:

[0206] Based on the operation data of the forklift controller and the ECU computer, extract the version information including vehicle model, identification number, and year.

[0207] Specifically, based on the operating data of the forklift controller and the ECU computer, vehicle model, identification number, and year / version information are extracted. It is necessary to obtain the data stored in the ECU from the forklift control system. First, connect to the forklift control system through the communication interface of the ECU, and use the CAN bus or serial communication protocols such as UART, SPI, etc. to read the firmware version, hardware serial number, and software version information in the storage unit. Then, extract the vehicle identification number, which is usually stored in the EEPROM or Flash storage unit. The information needs to be obtained by reading the data at specific addresses and verified in combination with the system registry content provided by the forklift controller, thereby obtaining the complete vehicle identification number. For the vehicle model information, it can be confirmed by parsing the device identification field in the ECU firmware. This field may contain information such as the manufacturer code, device type, load capacity, etc. For example, if the device identification field returned by the ECU is "FLK-3500", it means the forklift model is FLK-3500 and its rated load capacity is 3500 kg. To obtain the production year of the forklift, it is necessary to parse the ECU system log or production data record, which contains the device factory time. If the log record shows that the first operation time of the device is "2021-06-15", then the vehicle may be produced in 2021. Combining with other hardware information can further verify its accuracy. Finally, after integrating these data, a basic information table of the vehicle is formed, including the model, identification number, and production year.

[0208] The above are only the preferred embodiments of the present invention and do not limit the present invention in other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A method for intelligent analysis of forklift usage status, characterized in that: The following steps are involved: S1: Based on the forklift operation data, feature parameters including vehicle speed, fork load, and vehicle body posture angle are extracted, the fluctuation range of the feature parameters is divided into intervals, the fluctuation mean and variance of multiple intervals are calculated, all feature parameters are normalized, and the initial feature weight value is generated; S2: Based on the initial feature weight value, calculate the weighted distance of each forklift operation sample point on the vehicle speed sensor and the position change rate, calculate the weighted average distance of samples within the class and the weighted average distance of samples between classes, normalize the difference between the two and adjust the feature weight to generate a dynamic feature weight value; The steps for obtaining the dynamic feature weight value are specifically as follows: S211: performing parameter weighting processing on the running sample points of the vehicle speed sensor and the position change rate according to the initial feature weight value, calculating the weight distribution of the vehicle speed and position change rate parameters on the sample points by calling the initial weight value, and calculating the accumulated square difference value of each sample point on the weighted distance, to obtain a weighted sample point distance set; S212: separating the weighted distance of sample points within a class and the weighted distance of sample points between classes from the weighted sample point distance set, and calculating the weighted mean values ​​of sample points within a class and between classes respectively according to the classes to which the sample points belong by calling the weighted distance information of the sample points within the set, and gradually aggregating the sample class information to obtain the weighted average distance of samples within a class and the weighted average distance of samples between classes, and generating a sample weighted average distance parameter set; S213: Based on the sample weighted average distance parameter set, the difference between the weighted average distances of samples between classes and within classes is calculated point by point, using the formula: ； Calculate and generate dynamic feature weight values; in, represents the dynamic feature weight value, Indicates The weighted average distance within the class of samples, Indicates The weighted average distance between classes of samples, Indicates The initial feature weight value of the sample point, Dynamic adjustment coefficients are used for smooth weight calculations. represents the total number of samples, Used to adjust the influence of sample point weight on difference normalization. represents the normalized sum of weight differences; S3: Based on the dynamic feature weight value, the weighted density distribution value of the forklift operation sample point is calculated, the density peak value is determined by combining the vehicle speed and the fork load feature value, the density peak value is merged and clustered, the cluster samples are re-divided, the weighted density value of the cluster center is calculated, and the operation mode clustering result is generated; S4: Based on the operation mode clustering result, the sample groups are divided according to the operation stages of acceleration, turning, and stable operation in the forklift operation data, and the weighted feature values ​​of the vehicle body posture angle and position change rate in each stage are extracted. The fluctuation rate of the weighted feature values ​​is calculated and the weights are readjusted to generate the staged operation mode classification results; S5: Based on the classification results of the staged operation modes, the fluctuation value ranges of the fork load and the vehicle body posture angle in the differentiated operation stages are extracted, the fluctuation value distribution of each type of sample is classified and counted, the distribution mean and the fluctuation range are calculated, and a forklift usage status classification indicator set is generated.

2. The intelligent analysis method for forklift usage status according to claim 1 is characterized in that: The steps for obtaining the initial feature weight value are specifically as follows: S111: extracting characteristic parameters of vehicle speed, fork load and vehicle body posture angle from the forklift operation data, analyzing the fluctuation range of the characteristic parameters item by item according to the discrete interval rule, dividing the data set in each interval into separate areas, and generating fluctuation mean and fluctuation variance by performing item by item mean operation and variance calculation on sample points in the area; S112: Normalize the fluctuation mean and fluctuation variance using the formula: ； Calculate and generate a normalized feature parameter matrix; in, Represents the normalized feature matrix Line The eigenvalues ​​of the columns, Represents the original feature matrix Line The eigenvalues ​​of the columns, Representative The mean of the column features, Representative The variance of the column features, represents the smoothing coefficient used to avoid the denominator being zero; S113: Calculate the inter-column mean of the normalized feature parameter matrix, analyze its initial weight value column by column, use the matrix column mean vector calculation as the initial weight vector, and generate the initial feature weight value based on the column mean and the deviation superposition processing between the matrix column multiple eigenvalues.

3. The intelligent analysis method for forklift usage status according to claim 1 is characterized in that: The steps for obtaining the operation mode clustering result are specifically as follows: S311: calling the dynamic feature weight value, calculating the weighted density distribution value of the forklift operation sample point, performing weighted density calculation on each sample point through the vehicle speed and fork load feature value, accumulating the sample point density value point by point, and performing density classification processing in combination with the feature distribution parameter to generate a weighted density distribution value set; S312: Combine the weighted density distribution value set, compare the distribution values ​​of the vehicle speed and fork load characteristic values ​​with the overall density distribution point by point, and use the formula: ； Determine the local density peak of each sample point and generate a density peak set; in, represents the density peak, Indicates The weighted density value of sample points, represents the dynamic weight of vehicle speed characteristics, The dynamic weight representing the fork load characteristics, To adjust the parameters used to adjust the smoothness of the weight to density calculation, Indicates the dynamic smoothing effect of the fork load weight on the density value, The operation selects the local density maximum as the density peak; S313: performing merging clustering on the density peak set, determining the weighted distance distribution between the sample points by calculating the weighted density difference between the density peak of each sample point and other sample points, and reclassifying the clustering results by gradually aggregating the sample point attribution information to generate cluster sample division; S314: Call the cluster sample division, calculate the weighted density value of each cluster center, perform weighted summation of the vehicle speed and fork load characteristic values ​​of the sample points belonging to the cluster center in sequence, and dynamically adjust the center characteristics in combination with the mean distribution of the sample density values ​​within the class to generate the operation mode clustering result.

4. The method for intelligent analysis of forklift usage status according to claim 3, characterized in that: The steps for obtaining the classification results of the staged operation modes are specifically as follows: S411: calling the operation mode clustering result, dividing the sample groups item by item according to the acceleration, turning and stable operation stages in the forklift operation data, extracting the sample point position parameters and attitude angle feature values ​​of the operation stage, performing operation feature grouping processing on the sample groups in turn, and generating stage-by-stage sample group division; S412: extracting the characteristic values ​​of the vehicle body posture angle and position change rate in each stage of the phased sample group division, weighting the characteristic values ​​of the sample points point by point by combining the dynamic characteristic weight values, and performing weighted accumulation processing on the characteristic data of the running stage to generate a phased weighted characteristic value set; S413: Calculate the volatility of the phased weighted eigenvalue set, and readjust the weights using the formula: ； Calculate and generate the re-adjusted weight parameters; in, represents the re-adjusted weight parameter, represents the volatility of the weighted eigenvalue set, represents the mean of the weighted eigenvalue set, Represents the minimum fluctuation threshold, which is used to adjust the flexibility and sensitivity of the weights and increase the formula's ability to respond to small fluctuations; S414: Call the readjusted weight parameters, re-evaluate and classify the weighted feature values ​​of each stage in combination with the phased sample group division information, calculate the comprehensive feature parameters of the operation stage by classification, classify the readjusted classification features of the operation stage in turn, and generate the phased operation mode classification results.

5. The method for intelligent analysis of forklift usage status according to claim 4, characterized in that: The steps for obtaining the forklift usage status classification index set are specifically as follows: S511: calling the classification result of the staged operation mode, extracting the fluctuation value range of the fork load and the body posture angle in the differentiated operation stage, and gradually summarizing and classifying the fluctuation value range of the sample points by sequentially screening the dynamic change values ​​of the fork load and the body posture angle parameters for the sample points in the operation stage, and generating a sample fluctuation value set for the operation stage; S512: extracting the fluctuation value distribution of each type of samples in the sample fluctuation value set of the operation stage, calculating the fluctuation value interval characteristics of the samples in turn and combining the distribution range of each type of samples for classification statistics, and gradually classifying and summarizing the characteristic parameters of the fluctuation range of the sample points to generate the classification statistics results of the sample fluctuation value distribution; S513: Call the sample fluctuation value distribution classification statistics result, calculate the fluctuation mean and fluctuation range of each type of sample distribution, and perform weighted operation on the fluctuation value characteristics point by point and use the formula: ； Calculate and generate sample distribution mean and fluctuation range; in, represents the weighted fluctuation value of the sample distribution, Indicates The fluctuation value of the samples, Represents the mean of the fluctuation values ​​of all sample points, Indicates The dynamic weight of the sample points, The nonlinear effect of adjusting the weights for the smoothing coefficients, represents the smoothing effect of dynamically adjusting weights on distribution characteristics, Indicates the total number of samples; S514: calling the sample distribution mean and fluctuation range, and gradually classifying and summarizing the fluctuation value range and mean characteristic parameters of each type of sample in combination with the classification statistics results in the operation stage, recalculating the classification index for the dynamic change parameter characteristics of each type of operation stage sample, and generating a forklift usage status classification index set.

6. A forklift usage status intelligent analysis system, characterized in that: According to the method for intelligent analysis of forklift usage status according to any one of claims 1 to 5, the system comprises: The feature extraction module extracts feature values ​​from vehicle speed, fork load, and vehicle body posture angle based on forklift operation data, calculates the fluctuation mean and variance, normalizes the feature values, and obtains the initial feature weight value; The dynamic weight adjustment module calculates the weighted distance of the sample points based on the initial feature weight value, obtains the weighted average distance of samples inside and outside the class, performs normalization adjustment through the difference, updates the weight of each feature, and generates a dynamic feature weight value; The cluster analysis module calculates the weighted density distribution of the forklift operation sample points based on the dynamic feature weight values, determines the density peaks, performs merging clustering of the density peaks, calculates the weighted density of the cluster centers, and generates the operation mode clustering results; The operation mode classification module divides the forklift operation data into acceleration, turning, and stable operation stages based on the operation mode clustering results, extracts weighted features of the vehicle body posture angle and position change rate, calculates the fluctuation rate and adjusts the weight, and generates a staged operation mode classification result; The state classification index generation module extracts the fluctuation range of the fork load and the body posture angle in the differentiated operation stage based on the staged operation mode classification result, performs statistics on the fluctuation value distribution, calculates the distribution mean and interval, and generates a forklift use state classification index set.

Images