A catering store energy consumption control and energy-saving safety management system

Abstract

The application discloses a kind of catering store energy consumption control and energy-saving safety management system, belong to catering technical field, including: energy consumption acquisition module, video acquisition module, cloud server, execution module;The application sets energy consumption acquisition module, video acquisition module, cloud server and execution module, through the cooperation and cooperation between modules, the unified collection of fluid parameters and electric parameters such as energy consumption data of catering store can be realized, can be fused with the scene characteristics such as flow / non-flow, light on / light off, flame / extinguish and personnel / no personnel of catering store simultaneously, based on the correlation of energy consumption data and scene characteristics, the energy consumption anomaly of catering store is dynamically analyzed, and timely control intervention can be based on analysis result, compared with prior art, the workload of catering store personnel is reduced, the acquisition cycle is reduced, the situation that acquisition error is avoided is generated, and the timeliness of energy consumption control intervention is improved simultaneously.

Description

Technical Field

[0001] This invention belongs to the field of catering technology, specifically relating to an energy consumption control and energy-saving safety management system for catering establishments. Background Technology

[0002] Energy consumption in catering establishments refers to the total amount of various energy and resources consumed by catering establishments in their daily operations to complete activities such as cooking, serving, storage, and cleaning. It is a core component of the operating costs of catering establishments and a primary target for energy conservation and safety management.

[0003] Current methods for energy consumption and energy-saving safety management in catering establishments largely rely on manual reading of traditional instruments such as water meters, electricity meters, and gas meters. This method has the following drawbacks: 1. Fluid parameters such as water pressure and gas pressure, as well as electrical parameters such as current and voltage, cannot be collected uniformly. The collection workload is large, the collection cycle is long, and there are problems such as collection errors. 2. Unable to identify abnormal operating conditions of energy-related equipment in a timely manner, i.e., unable to intervene in energy consumption control in a timely manner; In view of this, an energy consumption control and energy-saving safety management system for catering establishments is designed to solve the above problems. Summary of the Invention

[0004] To address the problems mentioned in the background section, this invention provides an energy consumption control and energy-saving safety management system for catering establishments. This system reduces the workload of catering staff, shortens data collection cycles, avoids data collection errors, and improves the timeliness of energy consumption control intervention.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an energy consumption control and energy-saving safety management system for catering establishments, comprising: The energy consumption data acquisition module collects energy consumption data from restaurants. The video capture module collects video data from restaurants. The cloud server receives and analyzes the collected energy consumption data and video data from restaurants to determine if there are any anomalies in the restaurant's energy consumption and video data. The execution module executes relevant instructions based on the judgment results of the restaurant's energy consumption and whether there are any anomalies in the video.

[0006] Furthermore, the energy consumption data of the energy consumption acquisition module includes water energy consumption data, electricity energy consumption data, and gas energy consumption data.

[0007] Furthermore, the cloud server includes: The video preprocessing module receives the collected video data from the restaurants and decomposes the video data into continuous video frames. The image recognition module takes a video frame as input and identifies scene features in the video frame based on the built-in YOLOv5 model. The energy consumption preprocessing module aligns the collected energy consumption data of catering stores according to the timestamp and normalizes it; The feature preprocessing module encodes the scene features identified in the video frames; The data fusion module splices together normalized energy consumption data and encoded scene features according to timestamps; The fusion analysis module constructs a fusion analysis model based on an LSTM network and analyzes the fused data based on the fusion analysis model.

[0008] Furthermore, the YOLOv5 model built into the image recognition module has been pre-trained using historical data.

[0009] Furthermore, the encoding step in the feature preprocessing module includes converting the scene feature result into a binary numerical vector, including 1 and 0.

[0010] Furthermore, the fusion analysis model in the fusion analysis module includes an input layer, an embedding layer, a bidirectional LSTM layer, a fully connected layer, and an output layer.

[0011] Compared with the prior art, the beneficial effects of the present invention are: This invention comprises an energy consumption acquisition module, a video acquisition module, a cloud server, and an execution module. Through the collaboration of these modules, it enables the unified acquisition of energy consumption data, such as fluid and electrical parameters, in restaurants. Simultaneously, it integrates this data with scene characteristics of the restaurant, including water flow / no flow, light on / off status, flame / extinguished status, and presence / absence of personnel. Based on the correlation between energy consumption data and scene characteristics, it dynamically analyzes abnormal energy consumption in restaurants and provides timely control and intervention based on the analysis results. Compared to existing technologies, this reduces the workload of restaurant staff, shortens the data acquisition cycle, avoids data acquisition errors, and improves the timeliness of energy consumption control and intervention. Attached Figure Description

[0012] Figure 1 This is a system structure diagram of the present invention; Figure 2 This is a diagram of the cloud server structure of the present invention; In the diagram: 1. Energy consumption acquisition module; 2. Video acquisition module; 3. Cloud server; 4. Execution module; 301. Video preprocessing module; 302. Image recognition module; 303. Energy consumption preprocessing module; 304. Feature preprocessing module; 305. Data fusion module; 306. Fusion analysis module. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] A restaurant energy consumption control and energy-saving safety management system includes: Energy consumption data acquisition module 1 collects energy consumption data from catering establishments; The MPX5700 series water pressure sensor was installed at the end of the water inlet pipe of the restaurant to collect the static water pressure of the water inlet pipe in real time. The Hall current sensor of model ACS712 series is connected in series with the main circuit of the power distribution box of the catering store and the branch circuits of lighting equipment and air conditioning equipment to collect the current value of the power distribution box and equipment in real time. A voltage transformer of model ZMPT101B is connected in parallel to the output terminal of the main circuit of the distribution box of the catering store to collect the AC voltage value of the output terminal of the distribution box in real time. An ultrasonic gas meter of model G1.6 was installed in the main gas pipeline of the restaurant to collect the instantaneous flow and cumulative flow of the main gas pipeline. Data collected by the water pressure sensor, Hall current sensor, voltage transformer, and ultrasonic gas meter are transmitted to the cloud server via the LoRa / NB-IoT wireless communication module. Video capture module 2 collects video data from restaurant establishments; Install network cameras inside restaurants to collect video data from inside the restaurants. Video data captured by network cameras inside restaurants is transmitted to a cloud server via Ethernet. Cloud server 3 receives the collected energy consumption data and video data from catering stores, integrates and analyzes them to determine whether there are any anomalies in the energy consumption and video data of catering stores; Cloud server 3 includes: The video preprocessing module 301 receives the collected video data from the restaurant and decomposes the video data into continuous video frames. Image recognition module 302 takes a video frame as input and identifies scene features in the video frame based on the built-in YOLOv5 model. The built-in YOLOv5 model is pre-trained using historical data, and the steps include: Collect and label scene images of catering establishments. Scene images include water usage scene images, electricity usage scene images, gas usage scene images, and self-scene images. Among them, water usage scene images include water tap running scene images and water tap not running scene images; electricity usage scene images include light on scene images and light off scene images; gas usage scene images include flame scene images and flame off scene images; and self-scene images include scene images with people staying and empty store scene images without people staying. The labeled scene images are divided into training and testing sets. The built-in YOLOv5 model is trained using the training set, and the trained built-in YOLOv5 model is tested using the testing set. The built-in YOLOv5 model, after testing, can perform scene feature recognition in video frames; The energy consumption preprocessing module 303 aligns the collected energy consumption data of catering stores according to the timestamp and normalizes it. Normalization refers to mapping the collected energy consumption data of catering establishments to the interval [0, 1]. The feature preprocessing module 304 encodes the scene features identified in the video frames; The encoding steps for scene features are as follows: The results of scene feature analysis are converted into binary numerical vectors, i.e.: If the scene feature of the water usage scene image is a water tap running scene, then it is encoded as 1; otherwise, it is encoded as 0. If the scene feature of an image of an electricity consumption scene is that the light is on, then the code is 1; otherwise, the code is 0. If the scene features of the gas scene image are used as the flame scene features, then the code is 1; otherwise, the code is 0. If the scene features of the scene image itself are those of people standing in the scene, then it is encoded as 1; otherwise, it is encoded as 0. Data fusion module 305 splices normalized energy consumption data and encoded scene features according to timestamps; Fusion analysis module 306 constructs a fusion analysis model based on an LSTM network and analyzes the fused data based on the fusion analysis model; The fusion analysis model includes an input layer, an embedding layer, a bidirectional LSTM layer, a fully connected layer, and an output layer; The input layer includes 8 input dimensions, namely 4 energy consumption data and 4 scene features. The 4 energy consumption data are water pressure, current, voltage and gas flow. The 4 scene features are running water / no running water, light on / light off, flame / extinguished and people present / no people present. The embedding layer includes 8 input dimensions, contains 32 neurons and a ReLU activation function, and 32 output dimensions map uncertain scene features into identifiable scene features; The bidirectional LSTM layer consists of two bidirectional LSTM layers. The first bidirectional LSTM layer has 32 input dimensions, 64 neurons, and dropout=0.2. The 64 output dimensions capture the temporal correlation between short-term energy consumption data and scene features. The second bidirectional LSTM layer has 128 input dimensions, 32 neurons, and dropout=0.2. The 64 output dimensions capture the temporal correlation between long-term energy consumption data and scene features. The fully connected layer has 64 input dimensions, 16 neurons and LeakyReLU activation function, and 16 output dimensions. It will fuse the temporal correlation between captured short-term and long-term energy consumption data and scene features to extract discriminative features. The output layer has 16 input dimensions, 3 neurons and a Softmax activation function. The 3 output dimensions map the discriminative features of the fully connected layer to the probability distribution of three operating states: normal, abnormal energy consumption and abnormal scene. The maximum value of the class probability is taken as the final output result. Execution module 4 executes relevant instructions based on the judgment results of the restaurant's energy consumption and whether there are any anomalies in the video. The specific steps for executing the instruction include: If the output result is an energy consumption anomaly and the associated scene characteristics are lights off / no personnel, it is determined that non-essential equipment is not turned off. Based on the energy consumption anomaly data, the equipment with abnormal energy consumption is identified, and a shutdown command for the equipment with abnormal energy consumption is sent to control the shutdown of the equipment with abnormal energy consumption, thereby completing energy consumption control and energy-saving safety management.

[0015] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A restaurant energy consumption control and energy-saving safety management system, characterized in that, include: Energy consumption acquisition module (1) collects energy consumption data of catering stores; Video acquisition module (2) collects video data from catering establishments; The cloud server (3) receives the collected energy consumption data and video data of the catering stores, integrates and analyzes them to determine whether there are any abnormalities in the energy consumption and video of the catering stores. The execution module (4) executes relevant instructions based on the judgment results of whether there are any abnormalities in the energy consumption and video of the catering store.

2. The energy consumption control and energy-saving safety management system for catering establishments according to claim 1, characterized in that: The energy consumption data of the energy consumption acquisition module (1) includes water energy consumption data, electricity energy consumption data and gas energy consumption data.

3. The energy consumption control and energy-saving safety management system for catering establishments according to claim 2, characterized in that: The cloud server (3) includes: The video preprocessing module (301) receives the collected video data from the restaurant and decomposes the video data into continuous video frames. The image recognition module (302) takes a video frame as input and identifies scene features in the video frame based on the built-in YOLOv5 model. The energy consumption preprocessing module (303) aligns the collected energy consumption data of catering stores according to the timestamp and normalizes it; The feature preprocessing module (304) encodes the scene features identified in the video frames; The data fusion module (305) splices together the normalized energy consumption data and the encoded scene features according to the timestamp; The fusion analysis module (306) constructs a fusion analysis model based on the LSTM network and analyzes the fused data based on the fusion analysis model.

4. The energy consumption control and energy-saving safety management system for catering establishments according to claim 3, characterized in that: The YOLOv5 model built into the image recognition module (302) has been pre-trained using historical data.

5. The energy consumption control and energy-saving safety management system for catering establishments according to claim 4, characterized in that: The encoding step in the feature preprocessing module (304) includes: converting the result of the scene features into a binary numerical vector, including 1 and 0.

6. The energy consumption control and energy-saving safety management system for catering establishments according to claim 5, characterized in that: The fusion analysis model in the fusion analysis module (306) includes an input layer, an embedding layer, a bidirectional LSTM layer, a fully connected layer, and an output layer.

Images