Kitchen garbage crushing process safety monitoring method and system and electronic equipment

Abstract

The invention relates to the technical field of kitchen garbage crushing process safety monitoring scheme design, in particular to a kitchen garbage crushing process safety monitoring method and system and electronic equipment. A hydrogen fluoride sensor, a phosphorus pentafluoride sensor, a cyanide sensor, a methane sensor, a hydrogen sulfide sensor and a temperature sensor which are arranged in a kitchen garbage crusher are used for collecting related data in the kitchen garbage crushing process, and then a garbage crushing process abnormal state recognition model is trained; after the training of the garbage crushing process abnormal state recognition model is completed, inputting related data into the garbage crushing process abnormal state recognition model, outputting a recognition result, and further judging whether an abnormal state exists in the garbage crushing process or not; the overall state data of the kitchen garbage crushing process in a period of time is evaluated and judged integrally, and the potential safety hazards in the kitchen garbage crushing process are greatly reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of safety monitoring scheme design for the crushing process of kitchen waste, and specifically relates to a safety monitoring method and system for the crushing process of kitchen waste, and an electronic device. Background Art

[0002] The classification of existing kitchen waste may be carried out manually, and there is a situation where supervision is not in place during manual classification, which may lead to the possibility that waste such as waste batteries enters the kitchen waste crushing device. When the battery is crushed, it will short-circuit, discharge and generate high temperature, which will catch fire. The generated gas will contain hydrogen fluoride, phosphorus pentafluoride, cyanide, etc., which are extremely harmful to the human body. In addition, there is heavy metal pollution such as iron and nickel. In addition, when the battery discharges, it will generate oxygen and hydrogen, which is also prone to explosion, and there are great potential safety hazards. Moreover, during the crushing process of kitchen waste, if it is not properly processed, it may generate flammable gases such as methane, which may cause an explosion when encountering a relatively high temperature. In the prior art, there is an urgent need for a safety monitoring scheme for the crushing process of kitchen waste.

[0003] Therefore, the prior art still needs to be further developed. Summary of the Invention

[0004] The purpose of the present invention is to overcome the above technical deficiencies, and provide a safety monitoring method and system for the crushing process of kitchen waste, and an electronic device, so as to solve the problems existing in the prior art.

[0005] To achieve the above technical purpose, according to the first aspect of the present invention, the present invention provides a safety monitoring method for the crushing process of kitchen waste, including: S100. During the crushing process of kitchen waste, use a hydrogen fluoride sensor, a phosphorus pentafluoride sensor, a cyanide sensor, a methane sensor, a hydrogen sulfide sensor, and a temperature sensor arranged in the kitchen waste crusher to collect the hydrogen fluoride concentration, phosphorus pentafluoride concentration, cyanide concentration, methane concentration, hydrogen sulfide concentration, and temperature inside the crusher at preset time intervals, and then calculate the hydrogen fluoride concentration change rate, phosphorus pentafluoride concentration change rate, cyanide concentration change rate, methane concentration change rate, hydrogen sulfide concentration change rate, and temperature change rate between two adjacent time points. Record the data obtained from the above calculations as the first coupled data set. Collect a preset number of first coupled data sets, and divide each of the multiple first coupled data sets into a training data set and a validation data set according to a first preset ratio. Use the training data set to train an abnormal state recognition model for the garbage crushing process; use the validation data set to optimize the trained abnormal state recognition model for the garbage crushing process, optimize the model parameters, and complete the training of the abnormal state recognition model for the garbage crushing process; S200. After the abnormal state recognition model for the garbage crushing process is trained, relevant data is obtained again, and then the change rates of hydrogen fluoride concentration, phosphorus pentafluoride concentration, cyanide concentration, methane concentration, hydrogen sulfide concentration, and temperature change rate between two adjacent time points are calculated, which is recorded as the target coupled data set; S300. Input the target coupled data set into the abnormal state recognition model for the garbage crushing process, output the recognition result, and then determine whether there is an abnormal state in the garbage crushing process. According to the judgment result, determine whether to output an alarm signal regarding the abnormal state of the garbage crushing process.

[0006] Specifically, the method further includes: Before training the abnormal state recognition model for the garbage crushing process using the training data set, relevant managers need to confirm that the data in the first coupled data set collected is normal.

[0007] Specifically, training the abnormal state recognition model for the garbage crushing process using the training data set; optimizing the trained abnormal state recognition model for the garbage crushing process using the validation data set, optimizing the model parameters, and completing the training of the abnormal state recognition model for the garbage crushing process, including: Input the training data set in batches into a preset network layer for training. The preset network layer includes a Transformer network layer, and the Transformer network layer is used to predict the change rates of hydrogen fluoride concentration, phosphorus pentafluoride concentration, cyanide concentration, methane concentration, hydrogen sulfide concentration, and temperature change rate of the next two adjacent time points through forward propagation based on the change rates of hydrogen fluoride concentration, phosphorus pentafluoride concentration, cyanide concentration, methane concentration, hydrogen sulfide concentration, and temperature change rate between two adjacent time points, and then obtain the predicted loss value; calculate the loss value of the preset network layer and input it into an optimizer for optimization to determine the direction in which the parameter gradient of the abnormal state recognition model for the garbage crushing process decreases fastest; the abnormal state recognition model for the garbage crushing process performs backpropagation based on the loss value and the parameter gradient of the model to optimize the parameters of the abnormal state recognition model for the garbage crushing process.

[0008] Specifically, training the abnormal state recognition model for the garbage crushing process using the training data set; optimizing the trained abnormal state recognition model for the garbage crushing process using the validation data set, optimizing the model parameters, and completing the training of the abnormal state recognition model for the garbage crushing process, further includes: After each training, the validation data set is input into the preset network layer of the previous training in batches for model parameter verification, and cyclic training is performed. The total number of training rounds is set to the first preset number of rounds; record the loss value of the preset network layer, determine whether the loss value meets the first preset condition, and determine whether to end the training and output the model parameters according to the determination result.

[0009] Specifically, determining whether the loss value meets the first preset condition, and determining whether to end the training and output the model parameters according to the determination result includes: If the loss value meets the first preset condition, end the training and output the parameters of the current abnormal state recognition model for the garbage crushing process; if the loss value does not meet the first preset condition, continue the training.

[0010] Specifically, the first preset condition is: After the second preset number of rounds of training, the loss values obtained in the next round of training are all greater than or equal to the loss values that appeared during the second preset number of rounds of training.

[0011] Specifically, the recognition result is that there is an abnormal state in the garbage crushing process or there is no abnormal state in the garbage crushing process. Inputting multiple second first coupled data sets into the abnormal state recognition model of the garbage crushing process, outputting the recognition result, and then determining whether there is an abnormal state in the garbage crushing process. According to the determination result, it is determined whether to output an alarm signal regarding the existence of an abnormal state in the garbage crushing process, including: If the output result is that there is an abnormal state in the garbage crushing process, it is determined to output an alarm signal regarding the existence of an abnormal state in the garbage crushing process; If the output result is that there is no abnormal state in the garbage crushing process, it is determined not to output an alarm signal regarding the existence of an abnormality in the garbage crushing process.

[0012] Specifically, determining to output an alarm signal regarding the existence of an abnormal state in the garbage crushing process further includes: Controlling the kitchen waste crusher to stop working.

[0013] According to the second aspect of the present invention, there is provided a safety monitoring system for the kitchen waste crushing process, including: An acquisition module, including a hydrogen fluoride sensor, a phosphorus pentafluoride sensor, a cyanide sensor, a methane sensor, a hydrogen sulfide sensor, and a temperature sensor arranged in the kitchen waste crusher, for collecting the hydrogen fluoride concentration, phosphorus pentafluoride concentration, cyanide concentration, methane concentration, hydrogen sulfide concentration, and temperature inside the crusher at preset time intervals during the kitchen waste crushing process; A control module is configured to calculate the change rates of hydrogen fluoride concentration, phosphorus pentafluoride concentration, cyanide concentration, methane concentration, hydrogen sulfide concentration, and temperature at two adjacent time points. The calculated data is recorded as the first coupled data set. A preset number of first coupled data sets are collected and divided into a training data set and a validation data set according to a first preset ratio. The training data set is used to train an abnormal state recognition model for the garbage crushing process. The validation data set is used to optimize the trained abnormal state recognition model for the garbage crushing process, optimize the model parameters, and complete the training of the abnormal state recognition model for the garbage crushing process. After the training of the abnormal state recognition model for the garbage crushing process is completed, relevant data is obtained again, and then the change rates of hydrogen fluoride concentration, phosphorus pentafluoride concentration, cyanide concentration, methane concentration, hydrogen sulfide concentration, and temperature at two adjacent time points are calculated, which is recorded as the target coupled data set. The target coupled data set is input into the abnormal state recognition model for the garbage crushing process, and the recognition result is output, and then it is determined whether there is an abnormal state in the garbage crushing process. According to the determination result, it is determined whether to output an alarm signal regarding the abnormal state of the garbage crushing process.

[0014] According to a third aspect of the present invention, an electronic device is provided, including: a memory; and a processor, where computer-readable instructions are stored on the memory, and when the computer-readable instructions are executed by the processor, the above-mentioned safety monitoring method for the kitchen waste crushing process is implemented.

[0015] Beneficial effects: The present invention uses hydrogen fluoride sensors, phosphorus pentafluoride sensors, cyanide sensors, methane sensors, hydrogen sulfide sensors, and temperature sensors provided in the kitchen waste crusher to collect relevant data during the kitchen waste crushing process, and then trains an abnormal state recognition model for the garbage crushing process. After the training of the abnormal state recognition model for the garbage crushing process is completed, the relevant data is input into the abnormal state recognition model for the garbage crushing process, and the recognition result is output, and then it is determined whether there is an abnormal state in the garbage crushing process. It realizes the overall evaluation and judgment of the overall state data of the kitchen waste crushing process for a period of time, and can accurately judge whether the overall state data of the kitchen waste crushing process is in an abnormal state for a period of time, greatly improving the safety, reliability, and intelligence level of the safety monitoring of the kitchen waste crushing process, and greatly reducing the safety hazards of the kitchen waste crushing process. Description of the Drawings

[0016] Figure 1 It is a schematic flowchart of the safety monitoring method for the kitchen waste crushing process provided in a specific embodiment of the present invention; Figure 2It is a schematic diagram of the system composition of the kitchen waste crushing process safety monitoring system provided in the specific embodiment of the present invention; Figure 3 It is a schematic diagram of the structure of the abnormal state recognition model for the garbage crushing process provided in the specific embodiment of the present invention. Specific implementation manners

[0017] In order to enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the present invention. Based on the embodiments in this application, other similar embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this application. In addition, the directional terms mentioned in the following embodiments, such as "upper", "lower", "left", "right", etc., are only references to the directions in the accompanying drawings. Therefore, the directional terms used are for illustration rather than limitation of the present invention.

[0018] The present invention will be further described below in conjunction with the accompanying drawings and preferred embodiments.

[0019] Please refer to Figure 1 , the present invention provides a method for safety monitoring of the kitchen waste crushing process, including: S100. During the kitchen waste crushing process, use the hydrogen fluoride sensor, phosphorus pentafluoride sensor, cyanide sensor, methane sensor, hydrogen sulfide sensor, and temperature sensor set in the kitchen waste crusher to collect the hydrogen fluoride concentration, phosphorus pentafluoride concentration, cyanide concentration, methane concentration, hydrogen sulfide concentration, and temperature inside the crusher at preset time intervals, and then calculate the hydrogen fluoride concentration change rate, phosphorus pentafluoride concentration change rate, cyanide concentration change rate, methane concentration change rate, hydrogen sulfide concentration change rate, and temperature change rate between two adjacent time points. Record the data calculated above as the first coupled data set. Collect a preset number of first coupled data sets, and divide each of the multiple first coupled data sets into a training data set and a validation data set according to a first preset ratio. Use the training data set to train the abnormal state recognition model for the garbage crushing process; use the validation data set to optimize the trained abnormal state recognition model for the garbage crushing process, optimize the model parameters, and complete the training of the abnormal state recognition model for the garbage crushing process.

[0020] It should be noted here that before the step S100, it includes: Preset a preset number, a preset time interval, a first preset ratio, a first preset number of rounds, and a second preset number of rounds in the control module.

[0021] It can be understood that the preset quantity, preset time interval, first preset ratio, first preset round, and second preset round can be specifically set according to the actual needs of the users of the present invention. The present invention does not limit the specific values of the above parameters. There is no rule for setting the specific values of the above parameters. As long as it is applicable to the safety monitoring method for the kitchen waste crushing process proposed by the present invention, it is acceptable.

[0022] Preferably, the present invention sets the preset quantity to 2000, sets the preset time interval to 3 seconds, sets the first preset ratio to 8:2, sets the first preset round to 20000 times, and sets the second preset round to 200. The above settings can further improve the performance of the model and the efficiency of model training, and greatly improve the accuracy and reliability of the prediction of the abnormal state recognition model for the garbage crushing process of the present invention.

[0023] It can be understood that when there is no battery waste in the kitchen waste and the state of the kitchen waste crushing process is normal, there is a certain coupling rule for the change rates of hydrogen fluoride concentration, phosphorus pentafluoride concentration, cyanide concentration, methane concentration, hydrogen sulfide concentration, and temperature change rate at two adjacent time points. For example, there is a certain dynamic balance relationship among the change rates of hydrogen fluoride concentration, phosphorus pentafluoride concentration, cyanide concentration, methane concentration, hydrogen sulfide concentration, and temperature change rate at two adjacent time points. When there is an abnormal state in the garbage crushing process, this balance relationship will be broken. Based on this, the present invention designs and trains the abnormal state recognition model for the garbage crushing process.

[0024] S200. After the abnormal state recognition model for the garbage crushing process is trained, relevant data is obtained again, and then the change rates of hydrogen fluoride concentration, phosphorus pentafluoride concentration, cyanide concentration, methane concentration, hydrogen sulfide concentration, and temperature change rate at two adjacent time points are calculated, which is recorded as the target coupling data set.

[0025] Specifically, training the abnormal state recognition model for the garbage crushing process by using the training data set; optimizing the trained abnormal state recognition model for the garbage crushing process by using the verification data set, optimizing the model parameters, and completing the training of the abnormal state recognition model for the garbage crushing process, including: The training data set is input into a preset network layer in batches for training. The preset network layer includes a Transformer network layer, and the Transformer network layer is used to forward propagate and predict the hydrogen fluoride concentration change rate, phosphorus pentafluoride concentration change rate, cyanide concentration change rate, methane concentration change rate, hydrogen sulfide concentration change rate, and temperature change rate at the next two adjacent time points based on the hydrogen fluoride concentration change rate, phosphorus pentafluoride concentration change rate, cyanide concentration change rate, methane concentration change rate, hydrogen sulfide concentration change rate, and temperature change rate at two adjacent time points, so as to obtain a predicted loss value; calculate the loss value of the preset network layer and input it into an optimizer for optimization to determine the direction in which the parameter gradient of the abnormal state recognition model for the garbage crushing process drops fastest; the abnormal state recognition model for the garbage crushing process performs backpropagation according to the loss value and the parameter gradient of the model to optimize the parameters of the abnormal state recognition model for the garbage crushing process.

[0026] It can be understood that Transformer is a deep learning model based on the self-attention mechanism, consisting of a multi-head attention mechanism and a feed-forward neural network, capable of processing sequential data and capturing long-range dependencies. The Transformer model does not contain traditional convolutional layers, but processes each element in the input sequence through attention layers. The typical structure of Transformer includes an encoder and a decoder. The encoder is stacked by multiple identical layers, and the decoder adds a self-attention layer on the basis of the encoder to process sequence generation tasks. In the present invention, Transformer is used to supplement global context information and strengthen the model's ability in feature extraction and global understanding, greatly improving the model's performance and significantly reducing the computational cost.

[0027] Specifically, the preset network layer further includes a ResNet network layer, and the ResNet network layer is used to classify and predict the output result according to the collected first coupled data set. The classification result includes that there is an abnormal state in the garbage crushing process or there is no abnormal state in the garbage crushing process.

[0028] It should be noted here that ResNet is a deep convolutional neural network. It solves the problem of gradient disappearance in the training of deep networks by introducing residual connections. The core component of ResNet is the residual block, which usually contains two or three convolutional layers and a skip connection that allows the gradient to directly bypass these convolutional layers. The network structure of ResNet can be very deep. These networks stack multiple residual blocks to extract the features of pressure data and classify and predict the output result through global average pooling and fully connected layers. The classification result includes that there is an abnormal state in the garbage crushing process or there is no abnormal state in the garbage crushing process.

[0029] Please refer to Figure 3 , in the present invention, the ResNet network layer and the Transformer network layer are combined. ResNet is good at extracting local features of pressure data in the training dataset, while Transformer can supplement global context information. This combination can strengthen the model's ability in feature extraction and global understanding, greatly improving the model's performance and significantly reducing the computational cost.

[0030] It can be understood that the ResNet network layer is used to classify and predict the output results through the change rates of hydrogen fluoride concentration, phosphorus pentafluoride concentration, cyanide concentration, methane concentration, hydrogen sulfide concentration, and temperature change rate between two adjacent time points in the training dataset and the validation dataset. The specific steps are as follows: (1) First, put the training dataset into the Backbones network layer of the model. Backbone refers to a series of convolutional layers that form the backbone of a neural network. The main function of these layers is to extract features of the input data. The Backbone network usually consists of multiple convolutional layers, pooling layers, and activation functions, and can extract meaningful feature representations from the original data. Backbones are composed of multiple Backbones, and the core network layer in Backbone is the Resnet network layer. After the dataset is input into the model, the first coupled data set will be input into Backbones for training; (2) When constructing the Backbone, sensor name encoding and the Resnet network layer are added. Resnet starts layer-by-layer training after the training dataset is input, extracts the features in the first coupled data set, and assists the Transformer in predicting whether there is an abnormal state in the garbage crushing process. The extracted features obtain the name information of the sensor with the help of sensor name encoding. The name information of the sensor includes hydrogen fluoride sensor, phosphorus pentafluoride sensor, cyanide sensor, methane sensor, hydrogen sulfide sensor, and temperature sensor, which facilitates relevant personnel to quickly locate the abnormal sensor and further improves the safety of the kitchen waste crushing process described in the present invention.

[0031] Specifically, training the abnormal state recognition model of the garbage crushing process using the training dataset; optimizing the trained abnormal state recognition model of the garbage crushing process using the validation dataset, optimizing the model parameters, and completing the training of the abnormal state recognition model of the garbage crushing process, further including: After each training, the validation data set is input into the preset network layer of the previous training in batches for model parameter verification, and cyclic training is performed. The total number of training rounds is set to the first preset round; record the loss value of the preset network layer, determine whether the loss value meets the first preset condition, and determine whether to end the training and output the model parameters according to the judgment result.

[0032] Specifically, after each training, the validation data set is input into the preset network layer of the previous training in batches for model parameter verification, and cyclic training is performed, including the following methods: When the number of training times is insufficient and model optimization is desired, originally it had to be trained from scratch. Therefore, the present invention develops a method to continue the following training based on the model parameter file that has been previously trained. (1) Provide custom parameters, which can be selected and set according to user needs to continue training based on the training parameters of any previous training. (2) Load the parameters of the custom model file. (3) Put the validation data set into the custom parameter network layer for verification, obtain the loss value, and record the model parameters of the model corresponding to the minimum loss value during the training process. (4) Based on this set of model parameters, perform forward propagation training, backpropagation, and optimizer parameter optimization to obtain a new set of data. (5) Put the validation data into the new data network layer for inference to obtain the loss value. (6) Perform cyclic training to obtain the best model and complete model optimization.

[0033] It can be understood that the training method adopted by the present invention is batch training, which means that when updating the model parameters each time, only a part of the samples in the validation data set are used, which is called a batch. The advantage of batch training is that it can reduce memory consumption, speed up the training speed, increase randomness, and is beneficial to the generalization of the model.

[0034] Here it should be noted that in the present invention, batch_size = 2, and the batch_size is the size of the divided batch. For example, when batch_size = 2, it means that 2 first coupling data sets are selected from the validation data set each time and put into the model for verification.

[0035] Specifically, the determination of whether the loss value meets the first preset condition and the determination of whether to end the training and output the model parameters according to the judgment result include: If the loss value meets the first preset condition, end the training and output the parameters of the current abnormal state recognition model for garbage crushing; if the loss value does not meet the first preset condition, continue the training.

[0036] Specifically, the first preset condition is as follows: After the second preset round of training, the loss values obtained in the next round of training are all greater than or equal to the loss values that occurred during the second preset round of training.

[0037] It can be understood that the maximum number of rounds of cyclic training in the present invention is preferably 20,000 times, which can effectively ensure the accuracy of the model prediction of the present invention and record the loss value of the model training. The second preset round in the present invention is set to 200, that is, when the loss value of the training no longer decreases within 200 rounds of training, the round with the smallest loss value in the 200 rounds of training is saved as the parameters of the best verification round, and the best model file is generated. The above settings make the number of training rounds no longer rely on manual judgment, but use a deep learning model for automatic training, improving the accuracy of the model prediction, and automatically ending the training when the accuracy reaches the requirement, greatly saving the training time, and can also effectively prevent overfitting, greatly improving the intelligence level of the present invention and the efficiency of model training.

[0038] Specifically, the recognition result is that there is an abnormal state in the garbage crushing process or there is no abnormal state in the garbage crushing process. Inputting multiple target coupled data sets into the abnormal state recognition model of the garbage crushing process, outputting the recognition result, and then judging whether there is an abnormal state in the garbage crushing process, and determining whether to output an alarm signal regarding the existence of an abnormal state in the garbage crushing process according to the judgment result, including: If the output result is that there is an abnormal state in the garbage crushing process, it is determined to output an alarm signal regarding the existence of an abnormal state in the garbage crushing process; If the output result is that there is no abnormal state in the garbage crushing process, it is determined not to output an alarm signal regarding the existence of an abnormality in the garbage crushing process.

[0039] Furthermore, the specific process of the abnormal state recognition model of the garbage crushing process in the present invention is as follows: (1) Obtain the mean mu and logarithmic variance logvar of the latent vector during the forward propagation process of the abnormal state recognition model of the garbage crushing process; (2) Calculate the reasonable loss value loss representing the training effect: Calculate the divergence: k1 = -0.5 * (1 + logvar - mu^2 - (e^logvar)); Calculate the loss value loss between the hydrogen fluoride concentration change rate, phosphorus pentafluoride concentration change rate, cyanide concentration change rate, methane concentration change rate, hydrogen sulfide concentration change rate, temperature change rate and predicted value at two adjacent time points in the validation data set: loss = L1 + k1 * k1_weight; Among them, loss is the loss value, k1 is the divergence, k1_weight is the proportion of k1, with a value of 10, L1 is the absolute difference between the predicted value and the true value, and the calculation formula of L1 is as follows: ; Among them, A is the L1 loss in pytorch, which calculates the average or sum of the absolute differences between two tensors; A has a shape of N*C*H*W, which indicates that it is a four-dimensional tensor. N represents the number of samples (Batch Size), indicating that there are N data points. C represents the number of channels (Channels), and H and W represent the height and width respectively, which are usually the two-dimensional spatial dimensions of the input features; P represents the padding value, which is a boolean type. P has a shape of c*h*w*1, and the dimensions of P are the same as those of A in terms of channels, height, and width, but without the number of samples N, which means that P is the mask or padding information for a certain feature map of each sample. The boolean value of P may be used to mark whether the elements in A are filled (True) or valid (False).

[0040] (3) The loss value when using the validation dataset to verify the current model network; (4) If the validation loss in this round is less than the previous round, the best model parameters are replaced with the current training model; (5) Put the round number and the loss value loss into a dictionary one by one; (6) Compare the loss value of the current round with the loss values of the previous 200 rounds (parameters are provided and can be set according to actual needs, with a default of 200) of this round, that is, traverse the loss values from epoch - 200 to epoch - 1 (epoch is the training round, and the comparison is made when epoch is greater than 200). If the loss value in the current round of epoch is greater than the loss values from the (epoch - 200)th to the (epoch - 1)th, stop training.

[0041] (7) Save the best model parameters in the cyclic validation as a ckpt model file.

[0042] Specifically, the determination of whether the loss value meets the first preset condition and the determination of whether to end training and output the model parameters according to the determination result include: If the loss value meets the first preset condition, end training and output the parameters of the current abnormal state recognition model for the garbage crushing process; if the loss value does not meet the first preset condition, continue training.

[0043] Specifically, the first preset condition is: After the training of the second preset round, the loss values obtained in the next round of training are all greater than or equal to the loss values that occurred during the training of the second preset round.

[0044] Specifically, the method further includes: Before using the training data set to train the abnormal state recognition model for the garbage crushing process, relevant managers need to confirm that the data in the collected first coupled data set is normal.

[0045] It can be understood that for how to check that the data in the collected first coupled data set is normal, relevant staff can check according to industry standards, and the present invention will not elaborate too much here.

[0046] It should be noted here that the prediction results of the abnormal state recognition model for the garbage crushing process described in the present invention only include the existence of an abnormal state or the non-existence of an abnormal state. When there is an abnormal state, the corresponding predicted value and true value are 1, and when there is no abnormal state, the corresponding predicted value and true value are 0.

[0047] S300: Input the target coupled data set into the abnormal state recognition model for the garbage crushing process, output the recognition result, and then determine whether there is an abnormal state in the garbage crushing process. According to the determination result, determine whether to output an alarm signal regarding the existence of an abnormal state in the garbage crushing process.

[0048] Specifically, the recognition result is that there is an abnormal state in the garbage crushing process or there is no abnormal state in the garbage crushing process. The step of inputting multiple second first coupled data sets into the abnormal state recognition model for the garbage crushing process, outputting the recognition result, and then determining whether there is an abnormal state in the garbage crushing process, and determining whether to output an alarm signal regarding the existence of an abnormal state in the garbage crushing process according to the determination result includes: If the output result is that there is an abnormal state in the garbage crushing process, it is determined to output an alarm signal regarding the existence of an abnormal state in the garbage crushing process; If the output result is that there is no abnormal state in the garbage crushing process, it is determined not to output an alarm signal regarding the existence of an abnormal state in the garbage crushing process.

[0049] Specifically, the determination of outputting an alarm signal regarding the existence of an abnormal state in the garbage crushing process further includes: Controlling the kitchen waste crusher to stop working.

[0050] It should be noted here that the present invention utilizes hydrogen fluoride sensors, phosphorus pentafluoride sensors, cyanide sensors, methane sensors, hydrogen sulfide sensors, and temperature sensors disposed in the food waste grinder to collect relevant data during the food waste crushing process, and then trains an abnormal state recognition model for the garbage crushing process. After the abnormal state recognition model for the garbage crushing process is trained, the relevant data is input into the abnormal state recognition model for the garbage crushing process, and the recognition result is output, thereby determining whether there is an abnormal state in the garbage crushing process. It realizes the overall evaluation and judgment of the overall state data of the food waste crushing process over a period of time, and accurately judges whether the overall state data of the food waste crushing process is in an abnormal state over a period of time, greatly improving the safety, reliability, and intelligence level of the safety monitoring of the food waste crushing process, and greatly reducing the safety hazards in the food waste crushing process.

[0051] Please refer to Figure 2 , the present invention provides another embodiment. This embodiment provides a safety monitoring system for the food waste crushing process. The safety monitoring system for the food waste crushing process includes: An acquisition module 100, including a hydrogen fluoride sensor, a phosphorus pentafluoride sensor, a cyanide sensor, a methane sensor, a hydrogen sulfide sensor, and a temperature sensor disposed in the food waste grinder, is used to collect the hydrogen fluoride concentration, phosphorus pentafluoride concentration, cyanide concentration, methane concentration, hydrogen sulfide concentration, and temperature inside the grinder at preset time intervals during the food waste crushing process; A control module 200 is used to calculate the change rates of hydrogen fluoride concentration, phosphorus pentafluoride concentration, cyanide concentration, methane concentration, hydrogen sulfide concentration, and temperature at two adjacent time points, and record the calculated data as the first coupled data set. A preset number of first coupled data sets are collected, and each of the multiple first coupled data sets is divided into a training data set and a validation data set according to a first preset ratio, and the abnormal state recognition model for the garbage crushing process is trained using the training data set; the trained abnormal state recognition model for the garbage crushing process is optimized using the validation data set, and the model parameters are optimized to complete the training of the abnormal state recognition model for the garbage crushing process; it is used to obtain relevant data again after the abnormal state recognition model for the garbage crushing process is trained, and then calculate the change rates of hydrogen fluoride concentration, phosphorus pentafluoride concentration, cyanide concentration, methane concentration, hydrogen sulfide concentration, and temperature at two adjacent time points, which is recorded as the target coupled data set; it is used to input the target coupled data set into the abnormal state recognition model for the garbage crushing process, output the recognition result, and then determine whether there is an abnormal state in the garbage crushing process, and determine whether to output an alarm signal regarding the existence of an abnormal state in the garbage crushing process according to the judgment result.

[0052] It can be understood that the present invention utilizes hydrogen fluoride sensors, phosphorus pentafluoride sensors, cyanide sensors, methane sensors, hydrogen sulfide sensors, and temperature sensors provided in the food waste grinder to collect relevant data during the food waste crushing process, and then trains an abnormal state recognition model for the garbage crushing process. After the abnormal state recognition model for the garbage crushing process is trained, the relevant data is input into the abnormal state recognition model for the garbage crushing process, and the recognition result is output, thereby determining whether there is an abnormal state in the garbage crushing process. It realizes the overall evaluation and judgment of the overall state data of the food waste crushing process over a period of time, and can relatively accurately judge whether the overall state data of the food waste crushing process is in an abnormal state over a period of time, greatly improving the safety, reliability, and intelligence of the safety monitoring of the food waste crushing process, and greatly reducing the safety hazards in the food waste crushing process.

[0053] In a preferred embodiment, the present application further provides an electronic device, which includes: A memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the safety monitoring method for the food waste crushing process as described above is implemented. This computer device can generally be a server, a terminal, or any other electronic device with the necessary computing and / or processing capabilities. In one embodiment, this computer device may include a processor, a memory, a network interface, a communication interface, etc. connected through a system bus. The processor of this computer device can be used to provide the necessary computing, processing, and / or control capabilities. The memory of this computer device may include a non-volatile storage medium and an internal memory. An operating system, a computer program, etc. may be stored in or on the non-volatile storage medium. The internal memory can provide an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface and communication interface of this computer device can be used to connect and communicate with external devices through a network. When the computer program is executed by the processor, it executes the steps of the method of the present invention.

[0054] The present invention can be implemented as a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, causes the steps of the method of the embodiments of the present invention to be executed. In one embodiment, the computer program is distributed across a plurality of network-coupled computer devices or processors such that the computer program is stored, accessed, and executed by one or more computer devices or processors in a distributed manner. A single method step / operation, or two or more method steps / operations, can be executed by a single computer device or processor or by two or more computer devices or processors. One or more method steps / operations can be executed by one or more computer devices or processors, and one or more other method steps / operations can be executed by one or more other computer devices or processors. One or more computer devices or processors can execute a single method step / operation or execute two or more method steps / operations.

[0055] Those of ordinary skill in the art can understand that the method steps of the present invention can be completed by a computer program instructing related hardware such as computer devices or processors. The computer program can be stored in a non-transitory computer-readable storage medium, and when the computer program is executed, the steps of the present invention are caused to be executed. Depending on the situation, any reference herein to a memory, storage, database, or other medium may include non-volatile and / or volatile memory. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk, etc. Examples of volatile memory include random access memory (RAM), external cache memory, etc.

[0056] It can be understood that the present invention utilizes hydrogen fluoride sensors, phosphorus pentafluoride sensors, cyanide sensors, methane sensors, hydrogen sulfide sensors, and temperature sensors provided in the food waste grinder to collect relevant data during the food waste grinding process, and then trains an abnormal state recognition model for the food waste grinding process. After the abnormal state recognition model for the food waste grinding process is trained, the relevant data is input into the abnormal state recognition model for the food waste grinding process, and the recognition result is output, thereby determining whether there is an abnormal state in the food waste grinding process. It realizes the overall evaluation and judgment of the overall state data of the food waste grinding process over a period of time, and can more accurately judge whether the overall state data of the food waste grinding process is in an abnormal state over a period of time, greatly improving the safety, reliability, and intelligence of the safety monitoring of the food waste grinding process, and greatly reducing the safety hazards in the food waste grinding process.

[0057] The technical features described above can be combined arbitrarily. Although not all possible combinations of these technical features are described, any combination of these technical features should be considered to be covered by this specification, as long as there is no contradiction in such a combination.

[0058] The specific embodiments of the present invention described above do not constitute a limitation on the protection scope of the present invention. Any other corresponding changes and deformations made according to the technical concept of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A method for safety monitoring of kitchen waste pulverization process, characterized in that: The method comprises: S100, during the kitchen waste pulverizing process, using a hydrogen fluoride sensor, a phosphorus pentafluoride sensor, a cyanide sensor, a methane sensor, a hydrogen sulfide sensor, and a temperature sensor disposed in the kitchen waste pulverizer, the concentration of hydrogen fluoride, the concentration of phosphorus pentafluoride, the concentration of cyanide, the concentration of methane, the concentration of hydrogen sulfide, and the temperature in the pulverizer are collected at preset time intervals, and then the change rate of the hydrogen fluoride concentration, the change rate of the phosphorus pentafluoride concentration, the change rate of the cyanide concentration, the change rate of the methane concentration, the change rate of the hydrogen sulfide concentration, and the change rate of the temperature at two adjacent time points are calculated, and the data obtained by the above calculations are recorded as a first coupled data set, a preset number of first coupled data sets are collected, and a plurality of the first coupled data sets are respectively divided into a training data set and a verification data set according to a first preset ratio, and the training data set is used to train a garbage pulverizing process abnormal state recognition model; the verification data set is used to optimize the trained garbage pulverizing process abnormal state recognition model, optimize the model parameters, and complete the garbage pulverizing process abnormal state recognition model training; S200, after the training of the abnormal state recognition model of the garbage shredding process is completed, relevant data is obtained again, and then the change rate of hydrogen fluoride concentration, the change rate of phosphorus pentafluoride concentration, the change rate of cyanide concentration, the change rate of methane concentration, the change rate of hydrogen sulfide concentration, and the change rate of temperature at two adjacent time points are calculated, and recorded as the target coupling data set; S300, input the target coupling data set into the garbage shredding process abnormal state recognition model, output the recognition result, and then judge whether there is an abnormal state in the garbage shredding process, and determine whether to output an alarm signal about the abnormal state of the garbage shredding process according to the judgment result.

2. The method for safety monitoring of kitchen waste pulverization process according to claim 1, characterized in that: The method further comprises: Before using the training data set to train the garbage shredding process abnormal state recognition model, relevant management personnel are required to confirm that the data in the collected first coupling data set are not abnormal.

3. The method for safety monitoring of kitchen waste pulverization process according to claim 1, characterized in that: The method of using the training data set to train the garbage shredding process abnormal state recognition model; using the verification data set to optimize the trained garbage shredding process abnormal state recognition model, optimizing model parameters, and completing the garbage shredding process abnormal state recognition model training includes: The training data set is input into the preset network layer in batches for training, and the preset network layer includes a Transformer network layer, and the Transformer network layer is used to predict the hydrogen fluoride concentration change rate, phosphorus pentafluoride concentration change rate, cyanide concentration change rate, methane concentration change rate, hydrogen sulfide concentration change rate, and temperature change rate at the next two adjacent time points according to the hydrogen fluoride concentration change rate, phosphorus pentafluoride concentration change rate, cyanide concentration change rate, methane concentration change rate, hydrogen sulfide concentration change rate, and temperature change rate at two adjacent time points by forward propagation, and then obtain the predicted loss value; the loss value of the preset network layer is calculated and input into the optimizer for optimization, and the direction in which the parameter gradient of the garbage shredding process abnormal state recognition model decreases fastest is determined; the garbage shredding process abnormal state recognition model performs back propagation according to the loss value and the parameter gradient of the model to optimize the parameters of the garbage shredding process abnormal state recognition model.

4. The method for safety monitoring of kitchen waste pulverization process according to claim 1, characterized in that: The method of using the training data set to train the garbage shredding process abnormal state recognition model; using the verification data set to optimize the trained garbage shredding process abnormal state recognition model, optimizing the model parameters, and completing the garbage shredding process abnormal state recognition model training also includes: After each training, the verification data set is input into the preset network layer of the previous training in batches for model parameter verification, and cyclic training is performed, with the total number of training rounds set to the first preset round; the loss value of the preset network layer is recorded, and it is determined whether the loss value meets the first preset condition, and whether to end the training and output the model parameters based on the judgment result.

5. The method for safety monitoring of kitchen waste pulverization process according to claim 4, characterized in that: The determining whether the loss value satisfies the first preset condition, and determining whether to end the training and output the model parameters according to the determination result, includes: If the loss value meets the first preset condition, the training is terminated and the parameters of the abnormal state recognition model of the current garbage crushing process are output; if the loss value does not meet the first preset condition, the training is continued.

6. The method for safety monitoring of kitchen waste pulverization process according to claim 5, characterized in that: The first preset condition is: After the second preset round of training, the loss values ​​obtained in the next round of training are greater than or equal to the loss values ​​that occurred during the second preset round of training.

7. The method for safety monitoring of kitchen waste pulverization process according to claim 1, characterized in that: The recognition result is that the garbage shredding process has an abnormal state or the garbage shredding process does not have an abnormal state, the plurality of second and first coupled data sets are input into the garbage shredding process abnormal state recognition model, the recognition result is output, and then it is determined whether the garbage shredding process has an abnormal state, and whether to output an alarm signal related to the existence of an abnormal state in the garbage shredding process according to the judgment result, including: If the output result is that the garbage crushing process is in an abnormal state, it is determined that an alarm signal related to the abnormal state of the garbage crushing process is output; If the output result is that there is no abnormal state in the garbage crushing process, it is determined that no alarm signal regarding the abnormality in the garbage crushing process is output.

8. The method for safety monitoring of kitchen waste pulverization process according to claim 7, characterized in that: The determining and outputting of an alarm signal regarding an abnormal state in the garbage crushing process also includes: Control the food waste disposer to stop working.

9. A kitchen waste crushing process safety monitoring system, characterized in that: include: The acquisition module includes a hydrogen fluoride sensor, a phosphorus pentafluoride sensor, a cyanide sensor, a methane sensor, a hydrogen sulfide sensor, and a temperature sensor disposed in the kitchen waste crusher, and is used to collect the concentration of hydrogen fluoride, phosphorus pentafluoride, cyanide, methane, hydrogen sulfide, and temperature inside the crusher at preset time intervals during the kitchen waste crushing process; The control module is used to calculate the change rate of hydrogen fluoride concentration, the change rate of phosphorus pentafluoride concentration, the change rate of cyanide concentration, the change rate of methane concentration, the change rate of hydrogen sulfide concentration, and the change rate of temperature at two adjacent time points, record the data obtained by the above calculations as a first coupled data set, collect a preset number of first coupled data sets, divide the plurality of first coupled data sets into training data sets and verification data sets according to a first preset ratio, use the training data sets to train the garbage shredding process abnormal state recognition model; use the verification data set to optimize the trained garbage shredding process abnormal state recognition model, optimize the model parameters, and complete the garbage shredding process abnormal state recognition model. Training of an abnormal state recognition model for a garbage pulverizing process; after the training of the abnormal state recognition model for a garbage pulverizing process is completed, obtaining relevant data again, and then calculating the change rate of hydrogen fluoride concentration, the change rate of phosphorus pentafluoride concentration, the change rate of cyanide concentration, the change rate of methane concentration, the change rate of hydrogen sulfide concentration, and the change rate of temperature at two adjacent time points, which are recorded as a target coupled data set; inputting the target coupled data set into the abnormal state recognition model for a garbage pulverizing process, outputting a recognition result, and then judging whether there is an abnormal state in the garbage pulverizing process, and judging whether to output an alarm signal regarding the existence of an abnormal state in the garbage pulverizing process according to the judgment result.

10. An electronic device, characterized in that: include: Memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the method for safely monitoring the kitchen waste shredding process according to any one of claims 1 to 8 is implemented.

Images