Judgment method and system for rapidly evaluating water content grade of solid waste

Abstract

The invention relates to the technical field of solid waste image recognition, in particular to a judgment method and system for rapidly evaluating the water content grade of solid waste. Acquiring multi-angle image data and humidity data of the target solid waste, and preprocessing the multi-angle image data and the humidity data to obtain preprocessed data; presetting a solid waste water content identification model, and inputting the preprocessed data into the solid waste water content identification model; performing grade judgment on the water content of the solid waste according to an output result of the solid waste water content identification model. Four CCD cameras are adopted to collect images from different angles, meanwhile, a hygrometer on a sliding rail is used for collecting humidity data in real time, intelligent analysis is conducted on the collected data through a deep neural network model, and the solid waste is divided into four grades A, B, C and D according to the water content. The water content of the solid waste is rapidly and accurately evaluated, the evaluation efficiency is remarkably improved, and the labor cost is reduced.

Description

Technical Field

[0001] The invention relates to the technical field of solid waste image recognition, and in particular to a method and system for quickly evaluating the moisture content level of solid waste. Background Art

[0002] With the continuous development of society, the amount of solid waste generated has increased year by year. At present, the application of solid waste recycled in the market is generally based on direct combustion as fuel. The water content of solid waste has a great influence on the calorific value of combustion. Therefore, for large quantities of solid waste, its water content directly affects its value. At present, solid waste recycling manufacturers mainly judge the water content of large quantities of solid waste manually, mainly in the form of visual observation, with a handheld hygrometer to measure the humidity information around the waste to be tested, because the appearance of batch solid waste of different materials and different humidity will be different. Workers grade the water content of batch solid waste based on experience, among which Class A has the lowest water content and the best combustion performance during the production process. The relative water content of Class BCD increases in turn, and Class D solid waste is judged to be unusable for combustion. This form of determination depends on the workers' experience and technical level, and has subjective factors in scoring, so it is impossible to be absolutely objective. The present invention relies on image recognition and machine learning algorithms, and through multi-sensor coupling, it can efficiently and accurately determine the moisture content level of batch solid waste, thereby improving the determination efficiency and reducing the labor cost of identification, providing a new technical solution for the current intelligent recycling of solid waste. Summary of the invention

[0003] In view of the problems existing in the prior art, the present invention is proposed.

[0004] Therefore, the problem to be solved by the present invention is how to solve the problem that the water content of large quantities of solid waste directly affects its value. At present, solid waste recycling manufacturers mainly judge the water content of large quantities of solid waste manually, mainly in the form of visual observation, with a handheld hygrometer to measure the humidity information around the waste to be tested, because the appearance of batch solid wastes of different materials and different humidity will be different. Workers grade the water content of batch solid waste based on their experience, among which Class A has the least water content and the best combustion performance during the production process, and the relative water content of Classes BCD increases in turn, and Class D solid waste is judged to be unusable for combustion. This form of judgment depends on the experience and technical level of the workers, and there are subjective factors in the scoring, and it is impossible to be absolutely objective.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In a first aspect, an embodiment of the present invention provides a method for quickly evaluating the moisture content level of solid waste, which includes acquiring multi-angle image data and humidity data of target solid waste, and preprocessing the multi-angle image data and humidity data to obtain preprocessed data;

[0007] Presetting a solid waste moisture content identification model, and inputting the preprocessed data into the solid waste moisture content identification model;

[0008] The moisture content of the solid waste is graded according to the output result of the solid waste moisture content identification model.

[0009] As a preferred solution of the method for quickly evaluating the moisture content level of solid wastes described in the present invention, wherein: the multi-angle image data includes image data acquired by CCD cameras at four different angles; the humidity data includes humidity information collected by four hygrometers moving on a preset path;

[0010] The preprocessing includes data acquisition preprocessing and data enhancement preprocessing;

[0011] The data acquisition preprocessing includes image data processing and humidity data processing;

[0012] The data enhancement preprocessing is used to expand the data set after the data acquisition preprocessing.

[0013] As a preferred solution of the method for quickly evaluating the moisture content level of solid wastes described in the present invention, the image data processing includes:

[0014] The images acquired at four different angles are judged separately to obtain the moisture content grade judgment results at each angle;

[0015] The judgment results from the four angles are integrated and the lowest level is selected as the image judgment level of this batch of solid waste.

[0016] As a preferred solution of the method for rapidly evaluating the moisture content level of solid wastes described in the present invention, the humidity data processing includes:

[0017] Three sampling points are selected at equal distances on each suspension beam, and the hygrometer moves back and forth on the slide rail to collect humidity data at each sampling point;

[0018] The collected humidity data is weighted averaged, where the weights from the center to the outside are 0.4, 0.3, 0.2, and 0.1;

[0019] Calculate the average of the weighted average humidity values ​​on the four cantilever beams to obtain the final humidity value.

[0020] As a preferred solution of the method for rapidly evaluating the moisture content level of solid wastes described in the present invention, the solid waste moisture content identification model includes:

[0021] The input layer is used to receive the preprocessed image data and humidity data;

[0022] At least one feature extraction layer, used to extract feature information of input data;

[0023] The output layer is used to output the determination results of the moisture content level of solid waste.

[0024] As a preferred solution of the method for quickly evaluating the moisture content level of solid wastes described in the present invention, the solid waste moisture content identification model is a deep neural network model, including:

[0025] Multiple convolutional layers to extract image features;

[0026] Multiple pooling layers to reduce feature dimensionality;

[0027] Fully connected layer, used for feature fusion;

[0028] Classification layer, used for level determination.

[0029] As a preferred solution of the method for quickly evaluating the moisture content level of solid wastes described in the present invention, the moisture content level of solid wastes is determined by:

[0030] Solid waste is divided into four grades: A, B, C, and D according to the moisture content;

[0031] Among them, the moisture content of Class A is less than 5%, the moisture content of Class B is 5% to 20%, the moisture content of Class C is 20% to 40%, and the moisture content of Class D is greater than 40%.

[0032] In a second aspect, an embodiment of the present invention provides a system for quickly evaluating the moisture content level of solid waste, which includes a data acquisition module for acquiring multi-angle image data and humidity data of target solid waste, and preprocessing the multi-angle image data and humidity data to obtain preprocessed data;

[0033] A model operation module is used to preset a solid waste moisture content identification model and input the preprocessed data into the solid waste moisture content identification model;

[0034] The result output module determines the level of the water content of the solid waste according to the output result of the solid waste water content identification model.

[0035] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the method for quickly assessing the moisture content level of solid waste as described in the first aspect of the present invention are implemented.

[0036] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the method for rapidly assessing the moisture content level of solid waste as described in the first aspect of the present invention are implemented.

[0037] The beneficial effects of the present invention are as follows: the present invention realizes all-round and three-dimensional monitoring of solid waste by arranging four suspension beams on the gantry, installing a CCD camera at the end of each suspension beam to obtain multi-angle images, and installing a hygrometer on the suspension beam slide rail for dynamic collection. This multi-source data collection method not only avoids the subjectivity and one-sidedness of traditional manual observation, but also can capture the characteristic information of solid waste at different angles in real time, greatly improving the accuracy and comprehensiveness of data collection.

[0038] By setting three equidistant sampling points on each cantilever beam and using a weight distribution method that decreases from the center to the outside (0.4, 0.3, 0.2, 0.1), the humidity distribution of solid waste is accurately quantified. This weighted calculation method based on spatial position not only takes into account the actual characteristics of solid waste accumulation, but also effectively eliminates the interference caused by environmental factors, making humidity assessment more accurate and reliable.

[0039] The multi-layer deep neural network structure, including multiple convolutional layers for feature extraction, pooling layers for dimensionality reduction, fully connected layers for feature fusion, and classification layers for grade determination, forms an end-to-end intelligent recognition system. The model can automatically learn the deep features of solid waste, avoid the uncertainty of traditional manual experience judgment, and significantly improve the accuracy and reliability of the judgment results.

[0040] By classifying solid waste into four levels according to moisture content (Class A <5%, Class B 5% to 20%, Class C 20% to 40%, Class D > 40%), and combining image judgment and comprehensive analysis of humidity data, a scientific evaluation system has been established. This refined classification method is directly related to the combustion efficiency of solid waste, providing a clear guiding basis for subsequent treatment and effectively improving resource utilization efficiency.

[0041] The present invention organically combines multi-sensor data collection, intelligent algorithm processing, and automated determination technologies to build a complete rapid assessment system for the moisture content of solid waste. Compared with traditional manual determination methods, this system not only shortens the assessment time from several hours to minutes, but also increases the accuracy rate to more than 95%, while significantly reducing labor costs and realizing the intelligentization and standardization of solid waste treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 A flowchart for the method of quickly assessing the moisture content level of solid waste;

[0044] Figure 2 A diagram of a computer device for quickly assessing the moisture content of solid waste;

[0045] Figure 3 A schematic diagram of Class A (low water content, high calorific value) solid waste for rapid assessment of the moisture content level of solid waste;

[0046] Figure 4 A schematic diagram of Class B (low water content, high calorific value) solid waste for rapid assessment of the moisture content level of solid waste;

[0047] Figure 5 A schematic diagram of Class C (low water content, high calorific value) solid waste for rapid assessment of the moisture content level of solid waste;

[0048] Figure 6 A schematic diagram of Class D (low water content, high calorific value) solid waste for rapid assessment of the moisture content level of solid waste;

[0049] Figure 7 The present invention is a schematic diagram of a device for rapidly evaluating the moisture content level of solid waste. DETAILED DESCRIPTION

[0050] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0051] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0052] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0053] Example 1

[0054] Reference Figure 1 to Figure 7 , which is the first embodiment of the present invention, provides a method for quickly evaluating the moisture content level of solid waste, including:

[0055] S100: acquiring multi-angle image data and humidity data of target solid waste, and preprocessing the multi-angle image data and humidity data to obtain preprocessed data;

[0056] S101: The multi-angle image data includes image data acquired by CCD cameras at four different angles; the humidity data includes humidity information collected by four hygrometers moving on a preset path;

[0057] Preprocessing includes data collection preprocessing and data enhancement preprocessing;

[0058] The data acquisition preprocessing includes image data processing and humidity data processing;

[0059] Data augmentation preprocessing is used to expand the data set after data collection preprocessing.

[0060] Preferably, four CCD cameras at different angles are used to obtain the batch of solid waste images required for rating, a total of four images;

[0061] Four CCD cameras are fixed at the ends of four suspension beams, facing downward toward the center of the solid waste pile. The four suspension beams are fixed to a gantry by a rivet.

[0062] The collected 4 image information are transmitted to the terminal recognition part.

[0063] S102: Image data processing includes:

[0064] The images acquired at four different angles are judged separately to obtain the moisture content grade judgment results at each angle;

[0065] The judgment results from the four angles are integrated and the lowest level is selected as the image judgment level of this batch of solid waste.

[0066] Preferably, the solid waste is imaged by CCD cameras at four angles, and the images taken at each angle are judged by an algorithm program to obtain the solid waste grade taken at each angle, and the judgment of the four images is combined, and the final output result is the lowest grade of the four images. For example, if the judgment grade of three images is A and the judgment grade of one image is B, the final output of the judgment grade of the batch of solid waste is B;

[0067] The image classification and recognition code adopts a modular design, separating different functional modules, including data set training (train), model testing (test), and input prediction (predict), making the code structure clear and easy to understand and maintain;

[0068] The algorithm uses the cross entropy loss function and Adam optimizer, inputs the model's output and corresponding labels into the loss function, calculates the loss between the model's predicted value and the true value label, and adaptively adjusts the learning rate according to the historical gradient of each parameter to speed up convergence and improve model performance, thereby minimizing the loss function.

[0069] S103: Humidity data processing includes:

[0070] Three sampling points are selected at equal distances on each suspension beam, and the hygrometer moves back and forth on the slide rail to collect humidity data at each sampling point;

[0071] The collected humidity data is weighted averaged, where the weights from the center to the outside are 0.4, 0.3, 0.2, and 0.1;

[0072] Calculate the average of the weighted average humidity values ​​on the four cantilever beams to obtain the final humidity value.

[0073] Preferably, four hygrometers are used to collect humidity information of the batch of solid waste to be graded. Although humidity information is affected by environmental factors such as weather, the influence of humidity changes caused by weather is a factor that needs to be considered in the evaluation of the grade of the batch of solid waste.

[0074] Four hygrometers move back and forth on the suspension beam slide rail to record humidity information at multiple points.

[0075] Three sampling points are selected at equal distances on each cantilever beam wall. A motor is installed on each crossbeam to record the humidity information at each sampling point when the hygrometer slides back and forth.

[0076] The collected humidity information is calculated using a weighted average to obtain the final humidity value;

[0077] The weights of the observation points on each beam from the center to the outside are 0.4, 0.3, 0.2, and 0.1;

[0078] The humidity information obtained by weighted average on each of the four beams is averaged to calculate the final humidity value;

[0079] The final calculated humidity value is transmitted to the terminal.

[0080] S200: Preset a solid waste moisture content identification model, and input the preprocessed data into the solid waste moisture content identification model;

[0081] S201: Solid waste moisture content identification model includes:

[0082] The input layer is used to receive the preprocessed image data and humidity data;

[0083] At least one feature extraction layer, used to extract feature information of input data;

[0084] The output layer is used to output the determination results of the moisture content level of solid waste.

[0085] S202: The solid waste moisture content identification model is a deep neural network model, including:

[0086] Multiple convolutional layers to extract image features;

[0087] Multiple pooling layers to reduce feature dimensionality;

[0088] Fully connected layer, used for feature fusion;

[0089] Classification layer, used for level determination.

[0090] S300: Determine the level of moisture content of solid waste according to the output result of the solid waste moisture content identification model.

[0091] S301: Determination of moisture content level of solid waste includes:

[0092] Solid waste is divided into four grades: A, B, C, and D according to the moisture content;

[0093] Among them, the moisture content of Class A is less than 5%, the moisture content of Class B is 5% to 20%, the moisture content of Class C is 20% to 40%, and the moisture content of Class D is greater than 40%.

[0094] Preferably, solid waste is divided into four categories according to its moisture content, namely, A, B, C, and D, wherein category A has the lowest moisture content and the highest calorific value when burned, and category B, C, and D have increasing relative moisture contents, among which batch solid waste of category D has the highest moisture content.

[0095] Preferably, the machine learning uses the RESNET18 neural network model to learn the moisture content of large quantities of solid waste of different grades that have been manually rated before, and find the most appropriate training parameters.

[0096] Based on the information collected by multiple sensors, the RESNET18 neural network model algorithm that has been trained specifically for this device is used to evaluate the moisture level of large quantities of solid waste that need to be rated;

[0097] The evaluation results are transferred to the output part and displayed to the user.

[0098] The humidity values ​​collected by multiple sensors and the photo data taken by 4 CCDs are used to evaluate the final moisture content level based on the RESNET18 neural network model algorithm.

[0099] Resnet18 residual neural network model, the algorithm is implemented as:

[0100] It consists of multiple residual blocks, each of which contains two 3x3 convolutional layers with BatchNormalization and ReLU activation functions between each layer. In each residual block, the input passes through the first convolutional layer, then through BatchNormalization and ReLU activation function, then through the second convolutional layer, and finally the input is added to the residual (i.e., skip connection) and then through the ReLU activation function.

[0101] Device selection:

[0102] ①Frame structure part:

[0103] A gantry is fixed downward at the midpoint of the gantry, with four suspension beams extending outward from the center, and slide rail devices are reserved on the suspension beams.

[0104] ②Multi-sensor information collection part:

[0105] CCD camera: The system is equipped with four CCD cameras, which are fixed at the ends of four cantilever beams. Each camera faces downward and is aimed at the center of the solid waste pile, so that image information of the solid waste can be obtained from different angles.

[0106] Hygrometer: Four high-precision hygrometers are installed on the slide rails of the suspension beam and move back and forth to collect humidity information at various points.

[0107] ③Terminal algorithm identification and rating part:

[0108] Collect data information of samples of different grades, specifically: A-samples with a moisture content of less than 5%, B-samples with a moisture content of 5% to 20%, C-samples with a moisture content of 20% to 40%, and D-samples with a moisture content of more than 40%. The incineration calorific value of the boiler room can reflect the moisture content level of the batch of solid waste from the recycling station, and this is used as a basis for assessing the level of the batch of solid waste. First, collect the collected batch of solid waste image information, obtain its true level, and label it, dividing 80% into training set training set and 20% into test set test set. Based on this, Resnet18 neural network is used for image feature learning, and the number of convolution layers epoch and learning rate Learning rate are dynamically adjusted according to the accuracy of the training results. After the training is completed, the model is divided into 5 parts for cross-validation to prevent the model from overfitting, and finally output a recognition model with the best training results. After receiving the solid waste image information, the model can predict its level, and at the same time, combined with the humidity information collected by the hygrometer, comprehensively determine the moisture content level of the solid waste.

[0109] Output section:

[0110] The evaluation results are presented through a user interface, including

[0111] Includes information collected by 4 moisture meters, image prediction results and comprehensive assessment of the final moisture content level prediction of the batch of solid waste

[0112] 2. Operation process:

[0113] Data collection:

[0114] The center point is preset, and the information collected by the hygrometer can reflect the humidity of the batch of solid waste to a certain extent. At the same time, 4 CCD cameras shooting from 4 angles can ensure the comprehensiveness of image information and reduce the occurrence of accidental phenomena. The hygrometer slides at a constant speed on the preset path, and collects this information every fixed number of seconds, set to 1m / s, and collects information once per second. This data can be adjusted according to the size of the device. The measured humidity information is weighted averaged and output to ensure that the measured humidity information is the humidity information at the center point of the solid waste.

[0115] Result Output: The final moisture content level and its visual representation are displayed on the control panel, and the operator can make further processing decisions based on this information.

[0116] 3. Application scenarios

[0117] The device and system can be deployed in waste treatment plants to automatically assess the moisture content of incoming solid waste, thereby optimizing combustion efficiency, increasing calorific value, and reducing equipment maintenance costs.

[0118] Data Validation:

[0119] The operation cycle is based on one month. The collected data is compared with the humidity information corresponding to its actual combustion. The image information with incorrect evaluation results is analyzed specifically and newly added to the training set of the classification model.

[0120] Furthermore, this embodiment also provides a system for quickly evaluating the moisture content level of solid waste, comprising:

[0121] The data acquisition module acquires multi-angle image data and humidity data of the target solid waste, and pre-processes the multi-angle image data and humidity data to obtain pre-processed data;

[0122] The model operation module presets a solid waste moisture content identification model and inputs the preprocessed data into the solid waste moisture content identification model;

[0123] The result output module determines the level of water content of solid waste according to the output result of the solid waste water content identification model.

[0124] In summary, the present invention realizes all-round and three-dimensional monitoring of solid waste by arranging four suspension beams on the gantry, installing CCD cameras at the end of each suspension beam to obtain multi-angle images, and installing hygrometers on the suspension beam slide rails for dynamic data collection. This multi-source data collection method not only avoids the subjectivity and one-sidedness of traditional manual observation, but also can capture the characteristic information of solid waste at different angles in real time, greatly improving the accuracy and comprehensiveness of data collection.

[0125] By setting three equidistant sampling points on each cantilever beam and using a weight distribution method that decreases from the center to the outside (0.4, 0.3, 0.2, 0.1), the humidity distribution of solid waste is accurately quantified. This weighted calculation method based on spatial position not only takes into account the actual characteristics of solid waste accumulation, but also effectively eliminates the interference caused by environmental factors, making humidity assessment more accurate and reliable.

[0126] The multi-layer deep neural network structure, including multiple convolutional layers for feature extraction, pooling layers for dimensionality reduction, fully connected layers for feature fusion, and classification layers for grade determination, forms an end-to-end intelligent recognition system. The model can automatically learn the deep features of solid waste, avoid the uncertainty of traditional manual experience judgment, and significantly improve the accuracy and reliability of the judgment results.

[0127] By classifying solid waste into four levels according to moisture content (Class A <5%, Class B 5% to 20%, Class C 20% to 40%, Class D > 40%), and combining image judgment and comprehensive analysis of humidity data, a scientific evaluation system has been established. This refined classification method is directly related to the combustion efficiency of solid waste, providing a clear guiding basis for subsequent treatment and effectively improving resource utilization efficiency.

[0128] The present invention organically combines multi-sensor data collection, intelligent algorithm processing, and automated determination technologies to build a complete rapid assessment system for the moisture content of solid waste. Compared with traditional manual determination methods, this system not only shortens the assessment time from several hours to minutes, but also increases the accuracy rate to more than 95%, while significantly reducing labor costs and realizing the intelligentization and standardization of solid waste treatment.

[0129] Example 2

[0130] Reference Figure 2 - Figure 7 , which is the second embodiment of the present invention, and this embodiment provides a method for quickly evaluating the moisture content level of solid waste. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0131] This experiment was conducted in a large waste incineration power plant, and a total of 1,000 batches of solid waste samples collected from the plant from January to March 2024 were selected as experimental objects. The experimental site was set up on the garbage unloading platform, covering an area of ​​about 200 square meters, and a complete intelligent evaluation system was installed.

[0132] The system hardware configuration is as follows:

[0133] It includes a gantry 101, a suspension beam 102, a slide rail 103, a CCD camera 104, a hygrometer 105, and a batch solid waste storage area 106. The suspension beam 102 is clamped with the gantry 101. There are two suspension beams 102, and there are four slide rails 103. Each group of two slide rails is respectively sleeved on the suspension beam 102. There are also four groups of hygrometers 105, which are respectively clamped under each slide rail 103. There are four CCD cameras 104, which are respectively installed at the two ends of each slide rail 103. The two slide rails 103 are installed to form a "cross". The batch solid waste storage area 106 is vertically arranged at the bottom of the slide rail 103.

[0134] Gantry: 4.5 meters in height, 6 meters in span, made of Q235B steel;

[0135] CCD cameras: 4 high-definition industrial cameras (resolution 3840×2160), frame rate 60fps;

[0136] Humidity sensor: 4 DHT22 high-precision hygrometers, accuracy ±2%RH;

[0137] Computing platform: Industrial computer equipped with NVIDIA RTX 4090 graphics card;

[0138] The experiment was conducted in three phases:

[0139] Data collection phase (January 1st to January 31st):

[0140] First, 1,000 batches of samples were divided into four categories of A / B / C / D according to their moisture content and labeled, with 250 batches in each category. A fixed sampling area was set up 2.5 meters below the gantry, and the stacking volume of each batch of samples was controlled at 3±0.2 cubic meters. The CCD camera shot synchronously from four angles of 45°, 135°, 225°, and 315°, and 20 frames were continuously collected at each angle, and the 5 frames with the highest clarity were selected and saved. The hygrometer reciprocated along the slide at a speed of 1m / s, and recorded data at 12 preset sampling points (3 on each cantilever beam). The sampling time at each point was 3 seconds, and the average value was taken.

[0141] Model training phase (February 1st to February 15th):

[0142] 800 batches of data were used as training sets (200 batches per category) and 200 batches were used as test sets (50 batches per category). The image data was expanded to 4000 groups after preprocessing such as cropping, rotation, and noise addition. A deep learning network was used for training, and the network parameters were continuously optimized after 5 rounds of cross-validation. The humidity data used a center-weighted algorithm to fuse the data of the 12 sampling points with weights of 0.4, 0.3, 0.2, and 0.1.

[0143] System verification phase (February 16 - March 15):

[0144] 300 new batches of solid waste samples were selected for actual testing, and 5 manual evaluation experts with more than 10 years of work experience were invited to conduct manual ratings and conduct comparative experiments with traditional handheld moisture meters. The testing process of each batch of samples was strictly controlled to be completed within 5 minutes, and the accuracy, evaluation time, labor cost and other indicators of each evaluation method were recorded.

[0145] Table 1 Performance comparison data of different evaluation methods (based on 300 batches of test samples)

[0146]

[0147]

[0148] Table 2 Statistics of recognition results of 300 batches of test samples

[0149]

[0150] Table 3 System stability test data (24-hour operation for 7 consecutive days)

[0151] Test items Numeric System uptime ratio 99.2％ Average response time 2.8 seconds Data processing delay <0.5 seconds Sensor data collection accuracy 99.8％ Image acquisition clarity compliance rate 98.5％ System automatic alarm accuracy 99.1％

[0152] Through in-depth analysis of experimental data, the present invention shows significant advantages in the following aspects: Improved recognition accuracy:

[0153] The overall accuracy of the intelligent evaluation system of the present invention reaches 95.7%, which is 13.4 percentage points higher than that of manual expert evaluation (82.3%) and 17.1 percentage points higher than that of traditional handheld moisture meters (78.6%). In particular, in the identification of Class D high-water content samples, the accuracy of the system reaches 95.8%, while manual evaluation is only 79.5%, reflecting the stability of the system under complex conditions. This significant improvement is mainly due to the synergy of multi-angle image acquisition and deep learning models, which effectively overcomes the defects of traditional methods that are easily affected by environmental and human factors.

[0154] Efficiency and cost optimization:

[0155] In terms of processing efficiency, the system takes only 3.5 minutes to evaluate each batch on average, which is 86.3% higher than manual evaluation (25.6 minutes) and 81.0% higher than the traditional moisture meter method (18.4 minutes). More importantly, the system only requires one operator per shift, while the traditional method requires 2-4 professionals, which greatly reduces labor costs. Based on processing 100 batches per day, the system can save about 75% of labor costs, with an annualized cost savings of more than 500,000 yuan.

[0156] System stability verification:

[0157] Through 24 / 7 continuous operation tests, the system has shown excellent stability: the normal operation time ratio is 99.2%, the average response time is only 2.8 seconds, and the data processing delay is controlled within 0.5 seconds. Especially in terms of sensor data acquisition and image acquisition, the accuracy rates reached 99.8% and 98.5% respectively, far exceeding the expected target (95%). These data show that this system fully meets the requirements of industrial-grade applications.

[0158] Breaking capacity analysis:

[0159] In terms of the recognition ability of samples of different categories, the system showed a strong balance, with the recognition accuracy of the four categories of samples A / B / C / D being 97.3%, 94.7%, 96.0% and 94.7% respectively, with the maximum fluctuation being only 2.6 percentage points. This stable classification ability is due to the innovative multi-source data fusion algorithm and optimized deep learning model, which effectively avoids the problem of significant decline in accuracy of traditional methods in certain categories.

[0160] Practical application value:

[0161] The verification results based on 300 batches of test samples show that this system has significant application value in actual industrial environments. The misjudgment rate is only 4.3%, which is much lower than manual evaluation (17.7%) and traditional moisture meters (21.4%). Considering the dual requirements of accuracy and efficiency in the solid waste treatment industry, this performance indicator has important economic significance. A low misjudgment rate means more accurate classification and treatment, thereby improving incineration efficiency and reducing environmental pollution risks.

[0162] These data fully demonstrate the innovative value of the present invention in improving the efficiency, accuracy and economy of solid waste moisture content assessment.

[0163] Example 3

[0164] This embodiment also provides a computer device, which is suitable for a method for quickly evaluating the moisture content level of solid waste, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement a forced oscillation detection and positioning method for a distribution network as proposed in the above embodiment.

[0165] This embodiment further provides a storage medium on which a computer program is stored. When the program is executed by a processor, a forced oscillation detection and positioning method for a distribution network is implemented as proposed in the above embodiment.

[0166] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0167] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0168] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0169] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0170] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0171] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for quickly evaluating the moisture content of solid waste, characterized in that: include, Acquire multi-angle image data and humidity data of target solid waste, and pre-process the multi-angle image data and humidity data to obtain pre-processed data; Presetting a solid waste moisture content identification model, and inputting the preprocessed data into the solid waste moisture content identification model; The moisture content of the solid waste is graded according to the output result of the solid waste moisture content identification model.

2. The method for rapidly evaluating the moisture content of solid waste according to claim 1, characterized in that: The multi-angle image data includes image data acquired by CCD cameras at four different angles; the humidity data includes humidity information collected by four hygrometers moving on a preset path; The preprocessing includes data acquisition preprocessing and data enhancement preprocessing; The data acquisition preprocessing includes image data processing and humidity data processing; The data enhancement preprocessing is used to expand the data set after the data acquisition preprocessing.

3. The method for rapidly evaluating the moisture content of solid waste according to claim 2, characterized in that: The image data processing comprises: The images acquired at four different angles are judged separately to obtain the moisture content grade judgment results at each angle; The judgment results from the four angles are integrated and the lowest level is selected as the image judgment level of this batch of solid waste.

4. The method for rapidly evaluating the moisture content of solid waste according to claim 3, characterized in that: The humidity data processing includes: Three sampling points are selected at equal distances on each suspension beam, and the hygrometer moves back and forth on the slide rail to collect humidity data at each sampling point; The collected humidity data is weighted averaged, where the weights from the center to the outside are 0.4, 0.3, 0.2, and 0.1; Calculate the average of the weighted average humidity values ​​on the four cantilever beams to obtain the final humidity value.

5. The method for rapidly evaluating the moisture content level of solid waste according to claim 4, characterized in that: The solid waste moisture content identification model includes: The input layer is used to receive the preprocessed image data and humidity data; At least one feature extraction layer, used to extract feature information of input data; The output layer is used to output the determination results of the moisture content level of solid waste.

6. The method for rapidly evaluating the moisture content level of solid waste according to claim 5, characterized in that: The solid waste moisture content identification model is a deep neural network model, including: Multiple convolutional layers to extract image features; Multiple pooling layers to reduce feature dimensionality; Fully connected layer, used for feature fusion; Classification layer, used for level determination.

7. The method for rapidly evaluating the moisture content of solid waste according to claim 6, characterized in that: The determination of the moisture content level of solid waste includes: Solid waste is divided into four grades: A, B, C, and D according to the moisture content; Among them, the moisture content of Class A is less than 5%, the moisture content of Class B is 5% to 20%, the moisture content of Class C is 20% to 40%, and the moisture content of Class D is greater than 40%.

8. A system for rapidly evaluating the moisture content level of solid waste, based on the method for rapidly evaluating the moisture content level of solid waste as claimed in any one of claims 1 to 7, characterized in that: Also includes, A data acquisition module, which acquires multi-angle image data and humidity data of target solid waste, and pre-processes the multi-angle image data and humidity data to obtain pre-processed data; A model operation module, presetting a solid waste moisture content identification model, and inputting the preprocessed data into the solid waste moisture content identification model; The result output module determines the level of the water content of the solid waste according to the output result of the solid waste water content identification model.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for quickly evaluating the moisture content level of solid waste as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for quickly evaluating the moisture content level of solid waste as described in any one of claims 1 to 7 are implemented.

Images