Cold storage load prediction method and system based on Internet of Things and artificial intelligence

Abstract

The invention relates to the technical field of cold storage load prediction, in particular to a cold storage load prediction method and system based on the Internet of Things and artificial intelligence, and the method comprises the steps: collecting indoor people flow data and equipment operation data, and carrying out the people flow intensity evaluation and equipment operation stability evaluation; carrying out clustering analysis on the evaluation results of the people flow density and the equipment operation stability according to the indoor space use mode, and calculating the actual probability distribution of the indoor space use mode; obtaining expected probability distribution of the indoor space use mode, and performing deviation value calculation on the expected probability distribution and actual probability distribution of the indoor space use mode; and when the deviation value is smaller than a preset deviation value threshold value, cold storage load prediction is carried out based on the actual probability distribution of the indoor space use mode.

Description

Background Art

[0002] The emergence of cooling load prediction methods and systems based on the Internet of Things and artificial intelligence is based on the continuous improvement of modern buildings' demand for energy efficiency management and comfort. Traditional air conditioning load prediction methods mainly rely on simplified historical data and static models. These methods often fail to achieve ideal prediction results when faced with the complex indoor environment and usage patterns of modern buildings. With the rapid development of the Internet of Things technology and the widespread application of artificial intelligence algorithms, people have begun to explore the application of these technologies in cooling load prediction to improve prediction accuracy and energy efficiency management.

[0003] Disadvantages of existing technologies: Traditional prediction methods are mainly based on historical load data and are highly dependent on data, while historical data often cannot fully reflect the volatility and uncertainty of actual loads; static models are difficult to adapt to the rapid changes in the indoor environment and usage patterns of modern buildings, resulting in a large deviation between the prediction results and the actual load; existing prediction tools and methods have limitations in handling complex variables and responding quickly to environmental changes, and are difficult to meet the flexibility and real-time requirements of modern buildings for energy efficiency management. Summary of the invention

[0004] The main purpose of the present invention is to provide a cold storage load prediction method based on the Internet of Things and artificial intelligence, and further to provide a cold storage load prediction system based on the Internet of Things and artificial intelligence that can run and implement the above method, so as to effectively solve the above problems mentioned in the background technology.

[0005] The technical solution of the present invention is as follows: In the first aspect, a cold storage load prediction method based on the Internet of Things and artificial intelligence is proposed, and the method includes the following steps: S1. Collect indoor human flow data and equipment operation data to evaluate human flow density and equipment operation stability; S2, cluster analysis is performed on the evaluation results of human traffic density and equipment operation stability according to the indoor space usage pattern, and the actual probability distribution of the indoor space usage pattern is calculated; S3, obtaining the expected probability distribution of the indoor space usage pattern, and calculating the deviation value with the actual probability distribution of the indoor space usage pattern; S4. When the deviation value is less than the preset deviation value threshold, a cooling load prediction is performed based on the actual probability distribution of the indoor space usage pattern.

[0006] A further improvement of the present invention is that the crowd density assessment in S1 includes the following specific steps: S11, preset the collection cycle duration, divide the collection cycle duration equally into n time intervals, and divide the indoor area into m sub-areas according to the spatial layout and function, and record the area of ​​each sub-area; S12. In each time interval i, for each sub-area j, a weighted value of the flow of people is calculated. The calculation formula of the weighted value of the flow of people is: ;in, is the weighted value of the flow of people in the jth sub-area in the i-th time interval, k represents the kth time node in the i-th time interval, s is the number of time nodes in the i-th time interval, is the passenger flow weight factor of the jth sub-area at the kth time node in the i-th time interval, is the flow of people in the jth sub-area at the kth time node in the i-th time interval; S13, calculating the density of human traffic, the calculation formula of which is: ;in, is the flow density of people in the jth sub-area during the i-th time interval, is the area of ​​the jth sub-region, the value of i is 1-n, and the value of j is 1-m.

[0007] A further improvement of the present invention is that the equipment operation stability evaluation in S1 includes the following specific steps: S14. Within each time interval i, for each sub-area j, count the number of equipment failures and the total equipment operating time within the sub-area; S15. Calculate the mean stable operation time of the equipment to evaluate the equipment operation stability. The calculation formula for the mean stable operation time of the equipment is: ;in, is the mean stable operation time of the equipment in the jth sub-area in the i-th time interval, is the total operation time of the equipment in the jth sub-area during the i-th time interval, is the number of equipment failures in the jth sub-area during the i-th time interval.

[0008] A further improvement of the present invention is that S2 comprises the following specific steps: S21, extract the mean of the human flow density and the stable operation time of the equipment in the m sub-areas within the n time interval, and convert the human flow density of the sub-area j within the time interval i into And the average stable operation time of the equipment Combine into feature vector ; S22, randomly select 4 eigenvectors from the set of all eigenvectors as the initial cluster centers of the cluster analysis algorithm, for each eigenvector , calculate its Euclidean distance to each initial cluster center, and assign it to the cluster corresponding to the initial cluster center with the smallest distance; S23, calculate the mean of all eigenvectors in a single cluster, update the initial cluster center and The reallocation is performed until a preset number of iterations is reached and then the operation is stopped to complete the cluster analysis. Each cluster represents an indoor space usage pattern, and the indoor space usage pattern includes a high-load-stable device class, a high-load-unstable device class, a low-load-stable device class, and a low-load-unstable device class.

[0009] A further improvement of the present invention is that S2 further includes: S24. In each time interval i, for each sub-area j, the actual probabilities of the four types of indoor space usage patterns are calculated respectively, and the actual probability distribution vector for the j-th sub-area in the time interval i is obtained. ,in, is the actual probability of the high-load-stable equipment class in the jth sub-area in the i-th time interval, is the actual probability of the high-load-unstable device class in the jth sub-area in the i-th time interval, is the actual probability of the low-load-stable equipment class in the jth sub-area in the i-th time interval, is the actual probability of the low-load-unstable equipment class in the jth sub-area in the i-th time interval; S25. Repeat the above steps until m sub-areas of n time intervals are traversed.

[0010] A further improvement of the present invention is that the cluster analysis algorithm in S22 is a K-Means clustering algorithm, and K=4.

[0011] A further improvement of the present invention is that the specific content of obtaining the expected probability distribution of indoor space usage patterns in S3 is: for m sub-areas, obtain the pedestrian flow density, the average of the stable operation time of the equipment and the indoor space usage pattern in 3n time intervals within the first three collection cycles, combine the pedestrian flow density and the average of the stable operation time of the equipment in the 3n time intervals within the first three collection cycles into the form of a feature vector, and construct a machine learning model. The machine learning model takes the set of all groups of feature vectors as input, takes the expected probability distribution of the predicted indoor space usage pattern corresponding to each group of feature vectors as output, takes the expected probability distribution of the actual indoor space usage pattern corresponding to each group of feature vectors as the prediction target, takes minimizing the sum of the squares of the difference between the predicted expected probability distribution and the actual expected probability distribution as the training target, trains the machine learning model until the sum of the squares of the difference between the predicted expected probability distribution and the actual expected probability distribution reaches convergence, stops the training, and outputs the expected probability distribution vector of the predicted indoor space usage pattern. , the machine learning model is any one of a random forest model, a support vector machine or a neural network model.

[0012] A further improvement of the present invention is that the specific formula for calculating the deviation value in S3 is: ; in, is the deviation between the expected probability distribution and the actual probability distribution of the indoor space usage pattern of the jth sub-area in the i-th time interval.

[0013] A further improvement of the present invention is that S4 comprises the following specific steps: S41, preset deviation value threshold T, when When , the historical average cooling load data of the j-th sub-area under the high load-stable equipment type, high load-unstable equipment type, low load-stable equipment type and low load-unstable equipment type modes are obtained respectively; S42, predicting the cold storage load of the j-th sub-area during the time interval i, the calculation formula for predicting the cold storage load is: ;in, is the predicted cooling load of the jth sub-area in the i-th time interval, is the historical average cooling load of the j-th sub-area under the high load-stable equipment mode, is the historical average cooling load of the j-th sub-area under the high load-unstable equipment type mode, is the historical average cooling load of the j-th sub-area under the low-load-stable equipment mode, is the historical average cooling load of the j-th sub-area under the low load-unstable equipment type mode.

[0014] Secondly, a cold storage load prediction system based on the Internet of Things and artificial intelligence is proposed, which includes: data evaluation module, cluster analysis module, deviation calculation module, and load prediction module; The data evaluation module is used to collect indoor human flow data and equipment operation data, and to evaluate human flow density and equipment operation stability; The cluster analysis module is used to perform cluster analysis on the crowd density and equipment operation stability evaluation results according to the indoor space usage pattern, and calculate the actual probability distribution of the indoor space usage pattern; The deviation calculation module is used to obtain the expected probability distribution of the indoor space usage pattern and calculate the deviation value with the actual probability distribution of the indoor space usage pattern; The load prediction module is used to predict the cooling load based on the actual probability distribution of the indoor space usage pattern when the deviation value is less than a preset deviation value threshold.

[0015] The technical effects of the present invention are as follows: A cold storage load prediction method based on the Internet of Things and artificial intelligence was constructed. This method collects various information such as indoor human flow data and equipment operation data, and evaluates human flow density and equipment operation stability, so as to more accurately reflect changes in indoor environment and usage patterns and improve prediction accuracy. The cluster analysis algorithm is used to divide indoor space usage patterns, and predictions are made based on the deviation between the actual probability distribution and the expected probability distribution, so that the method can adapt to load changes under different indoor environments and usage patterns. The method can monitor changes in the indoor environment in real time, and adjust the prediction model based on real-time data to achieve real-time prediction and optimization. At the same time, the method can also handle more complex variables, such as personnel flow, equipment failure, etc., which improves the flexibility and accuracy of prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings: Figure 1 This is a flow chart of a cold storage load prediction method based on the Internet of Things and artificial intelligence according to Embodiment 1 of the present invention; Figure 2 This is a structural schematic diagram of a cold storage load prediction system based on the Internet of Things and artificial intelligence according to Example 2 of the present invention. DETAILED DESCRIPTION

[0017] Example 1 This embodiment constructs a cold storage load prediction method based on the Internet of Things and artificial intelligence. The method collects various information such as indoor human flow data and equipment operation data, and evaluates human flow density and equipment operation stability, so as to more accurately reflect the changes in indoor environment and usage patterns and improve prediction accuracy. The cluster analysis algorithm is used to divide the indoor space usage pattern, and prediction is made based on the deviation value between the actual probability distribution and the expected probability distribution, so that the method can adapt to the load changes under different indoor environments and usage patterns. The method can monitor the changes in the indoor environment in real time, and adjust the prediction model according to the real-time data to achieve real-time prediction and optimization. At the same time, the method can also handle more complex variables, such as personnel flow, equipment failure, etc., which improves the flexibility and accuracy of prediction.

[0018] Cold storage load prediction method based on Internet of Things and artificial intelligence, such as Figure 1 As shown, the specific steps include: S1. Collect indoor human flow data and equipment operation data to evaluate human flow density and equipment operation stability; S2, cluster analysis is performed on the evaluation results of human traffic density and equipment operation stability according to the indoor space usage pattern, and the actual probability distribution of the indoor space usage pattern is calculated; S3, obtaining the expected probability distribution of the indoor space usage pattern, and calculating the deviation value with the actual probability distribution of the indoor space usage pattern; S4. When the deviation value is less than the preset deviation value threshold, a cooling load prediction is performed based on the actual probability distribution of the indoor space usage pattern.

[0019] In this embodiment, the crowd density assessment in S1 includes the following specific steps: S11, preset the collection cycle duration, divide the collection cycle duration equally into n time intervals, and divide the indoor area into m sub-areas according to the spatial layout and function, and record the area of ​​each sub-area; S12. In each time interval i, for each sub-area j, a weighted value of the flow of people is calculated. The calculation formula of the weighted value of the flow of people is: ;in, is the weighted value of the flow of people in the jth sub-area in the i-th time interval, k represents the kth time node in the i-th time interval, s is the number of time nodes in the i-th time interval, is the passenger flow weight factor of the jth sub-area at the kth time node in the i-th time interval, is the flow of people in the jth sub-area at the kth time node in the i-th time interval; S13, calculating the density of human traffic, the calculation formula of which is: ;in, is the flow density of people in the jth sub-area during the i-th time interval, is the area of ​​the jth sub-region, the value of i is 1-n, and the value of j is 1-m.

[0020] In this embodiment, the equipment operation stability evaluation in S1 includes the following specific steps: S14. Within each time interval i, for each sub-area j, count the number of equipment failures and the total equipment operating time within the sub-area; S15. Calculate the mean stable operation time of the equipment to evaluate the equipment operation stability. The calculation formula for the mean stable operation time of the equipment is: ;in, is the mean stable operation time of the equipment in the jth sub-area in the i-th time interval, is the total operation time of the equipment in the jth sub-area during the i-th time interval, is the number of equipment failures in the jth sub-area during the i-th time interval.

[0021] In this embodiment, S2 includes the following specific steps: S21, extract the mean of the human flow density and the stable operation time of the equipment in the m sub-areas within the n time interval, and convert the human flow density of the sub-area j within the time interval i into And the average stable operation time of the equipment Combine into feature vector ; S22, randomly select 4 eigenvectors from the set of all eigenvectors as the initial cluster centers of the cluster analysis algorithm, for each eigenvector , calculate its Euclidean distance to each initial cluster center, and assign it to the cluster corresponding to the initial cluster center with the smallest distance; S23, calculate the mean of all eigenvectors in a single cluster, update the initial cluster center and The reallocation is performed until a preset number of iterations is reached and then the operation is stopped to complete the cluster analysis. Each cluster represents an indoor space usage pattern, and the indoor space usage pattern includes a high-load-stable device class, a high-load-unstable device class, a low-load-stable device class, and a low-load-unstable device class.

[0022] In this embodiment, S2 further includes: S24. In each time interval i, for each sub-area j, the actual probabilities of the four types of indoor space usage patterns are calculated respectively, and the actual probability distribution vector for the j-th sub-area in the time interval i is obtained. ,in, is the actual probability of the high-load-stable equipment class in the jth sub-area in the i-th time interval, is the actual probability of the high-load-unstable device class in the jth sub-area in the i-th time interval, is the actual probability of the low-load-stable equipment class in the jth sub-area in the i-th time interval, is the actual probability of the low-load-unstable equipment class in the jth sub-area in the i-th time interval; S25. Repeat the above steps until m sub-areas of n time intervals are traversed.

[0023] In this embodiment, the clustering analysis algorithm in S22 is a K-Means clustering algorithm, and K=4.

[0024] In this embodiment, the specific content of obtaining the expected probability distribution of the indoor space usage pattern in S3 is: for m sub-areas, obtain the pedestrian flow density, the average of the stable operation time of the equipment and the indoor space usage pattern in 3n time intervals within the first three collection cycles, combine the pedestrian flow density and the average of the stable operation time of the equipment in the 3n time intervals within the first three collection cycles into the form of a feature vector, and construct a machine learning model. The machine learning model takes the set of all groups of feature vectors as input, takes the expected probability distribution of the predicted indoor space usage pattern corresponding to each group of feature vectors as output, takes the expected probability distribution of the actual indoor space usage pattern corresponding to each group of feature vectors as the prediction target, takes minimizing the sum of the squares of the difference between the predicted expected probability distribution and the actual expected probability distribution as the training target, trains the machine learning model until the sum of the squares of the difference between the predicted expected probability distribution and the actual expected probability distribution reaches convergence, stops the training, and outputs the expected probability distribution vector of the predicted indoor space usage pattern. , the machine learning model is any one of a random forest model, a support vector machine or a neural network model.

[0025] In this embodiment, the specific formula for calculating the deviation value in S3 is: ; in, is the deviation between the expected probability distribution and the actual probability distribution of the indoor space usage pattern of the jth sub-area in the i-th time interval.

[0026] In this embodiment, S4 includes the following specific steps: S41, preset deviation value threshold T, when When , the historical average cooling load data of the j-th sub-area under the high load-stable equipment type, high load-unstable equipment type, low load-stable equipment type and low load-unstable equipment type modes are obtained respectively; S42, predicting the cold storage load of the j-th sub-area during the time interval i, the calculation formula for predicting the cold storage load is: ;in, is the predicted cooling load of the jth sub-area in the i-th time interval, is the historical average cooling load of the j-th sub-area under the high load-stable equipment mode, is the historical average cooling load of the j-th sub-area under the high load-unstable equipment type mode, is the historical average cooling load of the j-th sub-area under the low-load-stable equipment mode, is the historical average cooling load of the j-th sub-area under the low load-unstable equipment type mode.

[0027] Example 2 This embodiment proposes a cold storage load prediction system based on the Internet of Things and artificial intelligence. Figure 2 As shown, it includes: data evaluation module, cluster analysis module, deviation calculation module, load forecasting module; The data evaluation module is used to collect indoor human flow data and equipment operation data, and to perform human flow density evaluation and equipment operation stability evaluation.

[0028] The cluster analysis module is used to perform cluster analysis on the evaluation results of the density of human traffic and the stability of equipment operation according to the indoor space use pattern, and calculate the actual probability distribution of the indoor space use pattern.

[0029] The deviation calculation module is used to obtain the expected probability distribution of the indoor space usage pattern and calculate the deviation value with the actual probability distribution of the indoor space usage pattern.

[0030] The load prediction module is used to predict the cooling load based on the actual probability distribution of the indoor space usage pattern when the deviation value is less than a preset deviation value threshold.

[0031] In this embodiment, the crowd density assessment includes the following specific steps: first, the collection cycle duration is preset, and the collection cycle duration is evenly divided into n time intervals. At the same time, the indoor area is divided into m sub-areas according to the spatial layout and function, and the area of ​​a single sub-area is recorded; then, within each time interval i, for each sub-area j, the weighted value of the crowd flow is calculated, and the calculation formula of the weighted value of the crowd flow is: ;in, is the weighted value of the flow of people in the jth sub-area in the i-th time interval, k represents the kth time node in the i-th time interval, s is the number of time nodes in the i-th time interval, is the passenger flow weight factor of the jth sub-area at the kth time node in the i-th time interval, is the flow of people in the jth sub-area at the kth time node in the i-th time interval; finally, the density of the flow of people is calculated, and the calculation formula of the density of the flow of people is: ;in, is the flow density of people in the jth sub-area during the i-th time interval, is the area of ​​the jth sub-region, the value of i is 1-n, and the value of j is 1-m.

[0032] In this embodiment, the equipment operation stability evaluation includes the following specific steps: first, within each time interval i, for each sub-area j, the number of equipment failures and the total equipment operation time in the sub-area are counted; then, the mean value of the equipment stable operation time is calculated to evaluate the equipment operation stability, and the calculation formula of the mean value of the equipment stable operation time is: ;in, is the mean stable operation time of the equipment in the jth sub-area in the i-th time interval, is the total operation time of the equipment in the jth sub-area during the i-th time interval, is the number of equipment failures in the jth sub-area during the i-th time interval.

[0033] In this embodiment, the implementation of the cluster analysis module includes the following specific steps: first, extract the mean of the human flow density and the stable operation time of the equipment in the m sub-areas within the n time intervals, and extract the human flow density of the sub-area j within the time interval i. And the average stable operation time of the equipment Combine into feature vector ; Then randomly select 4 eigenvectors from the set of all eigenvectors as the initial cluster centers of the cluster analysis algorithm. For each eigenvector , calculate its Euclidean distance to each initial cluster center, and assign it to the cluster corresponding to the initial cluster center with the smallest distance; further, calculate the mean of all feature vectors in a single cluster, update the initial cluster center and assign each feature vector The reallocation is performed until the preset number of iterations is reached and the operation is stopped to complete the cluster analysis. Each cluster represents an indoor space usage mode, which includes high-load-stable equipment class, high-load-unstable equipment class, low-load-stable equipment class and low-load-unstable equipment class. Finally, in each time interval i, for each sub-area j, the actual probabilities of the four types of indoor space usage modes are calculated respectively, and the actual probability distribution vector for the j-th sub-area in the time interval i is obtained. ,in, is the actual probability of the high-load-stable equipment class in the jth sub-area in the i-th time interval, is the actual probability of the high-load-unstable device class in the jth sub-area in the i-th time interval, is the actual probability of the low-load-stable equipment class in the jth sub-area in the i-th time interval, is the actual probability of the low-load-unstable device class in the jth sub-area in the ith time interval; repeat the above steps until m sub-areas in n time intervals are traversed.

[0034] In this embodiment, the cluster analysis algorithm is a K-Means clustering algorithm, and K=4.

[0035] In this embodiment, the specific content of obtaining the expected probability distribution of the indoor space usage pattern is as follows: for m sub-areas, the pedestrian flow density, the average of the stable operation time of the equipment and the indoor space usage pattern in 3n time intervals within the first three acquisition cycles are obtained, and the pedestrian flow density and the average of the stable operation time of the equipment in the 3n time intervals within the first three acquisition cycles are combined into the form of feature vectors to construct a machine learning model, wherein the machine learning model takes the set of all groups of feature vectors as input, takes the expected probability distribution of the predicted indoor space usage pattern corresponding to each group of feature vectors as output, takes the expected probability distribution of the actual indoor space usage pattern corresponding to each group of feature vectors as the prediction target, takes minimizing the sum of the squares of the difference between the predicted expected probability distribution and the actual expected probability distribution as the training target, trains the machine learning model until the sum of the squares of the difference between the predicted expected probability distribution and the actual expected probability distribution reaches convergence, stops the training, and outputs the expected probability distribution vector of the predicted indoor space usage pattern. , the machine learning model is any one of a random forest model, a support vector machine or a neural network model.

[0036] In this embodiment, the specific formula for calculating the deviation value is: ;in, is the deviation between the expected probability distribution and the actual probability distribution of the indoor space usage pattern of the jth sub-area in the i-th time interval.

[0037] In this embodiment, the implementation of the load forecasting module includes the following specific steps: first, a deviation value threshold T is preset. When , the historical average cooling load data of the j-th sub-area in the high load-stable equipment type, high load-unstable equipment type, low load-stable equipment type and low load-unstable equipment type modes are obtained respectively; then the cooling load of the j-th sub-area is predicted for the time interval i, and the calculation formula for the predicted cooling load is: ;in, is the predicted cooling load of the jth sub-area in the i-th time interval, is the historical average cooling load of the j-th sub-area under the high load-stable equipment mode, is the historical average cooling load of the j-th sub-area under the high load-unstable equipment type mode, is the historical average cooling load of the j-th sub-area under the low-load-stable equipment mode, is the historical average cooling load of the j-th sub-area under the low load-unstable equipment type mode.

[0038] The above-mentioned parameters and steps for each unit module to realize the corresponding functions in the cold storage load prediction system based on the Internet of Things and artificial intelligence of the present invention can refer to the parameters and steps in the embodiment of the cold storage load prediction method based on the Internet of Things and artificial intelligence in Example 1 above.

[0039] Example 3 This embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes the above-mentioned cold storage load prediction method based on the Internet of Things and artificial intelligence by calling the computer program stored in the memory.

[0040] The electronic device may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) and one or more memories, wherein the memory stores at least one computer program, and the computer program is loaded and executed by the processor to implement the cold storage load prediction method based on the Internet of Things and artificial intelligence provided by the above method embodiment. The electronic device may also include other components for realizing the functions of the device, for example, the electronic device may also have components such as a wired or wireless network interface and an input and output interface to input and output data. This embodiment will not be described in detail here.

[0041] Those skilled in the art know that the present invention can be implemented as a system, method or computer program product. Therefore, the present disclosure can be specifically implemented in the following forms, namely: it can be complete hardware, it can be complete software (including firmware, resident software, microcode, etc.), or it can be a combination of hardware and software, which is generally referred to as a "circuit", "module" or "system" herein. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable medium contains computer-readable program code.

[0042] Any combination of one or more computer-readable media may be used. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or device.

[0043] The present invention is described with reference to flowcharts and block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process or block in the flowchart and block diagram, as well as the combination of processes and blocks in the flowchart or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts. Figure 1 Process or multiple processes and boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0044] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 Process or multiple processes and boxes Figure 1 The steps for the functions specified in one or more boxes.

[0045] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A cold storage load prediction method based on the Internet of Things and artificial intelligence, characterized by: The specific steps include: S1. Collect indoor human flow data and equipment operation data to evaluate human flow density and equipment operation stability; S2, cluster analysis is performed on the evaluation results of human traffic density and equipment operation stability according to the indoor space usage pattern, and the actual probability distribution of the indoor space usage pattern is calculated; S3, obtaining the expected probability distribution of the indoor space usage pattern, and calculating the deviation value with the actual probability distribution of the indoor space usage pattern; S4. When the deviation value is less than the preset deviation value threshold, a cooling load prediction is performed based on the actual probability distribution of the indoor space usage pattern.

2. The cold storage load prediction method based on the Internet of Things and artificial intelligence according to claim 1 is characterized in that: The crowd density assessment in S1 includes the following specific steps: S11, preset the collection cycle duration, divide the collection cycle duration equally into n time intervals, and divide the indoor area into m sub-areas according to the spatial layout and function, and record the area of ​​each sub-area; S12. In each time interval i, for each sub-area j, a weighted value of the flow of people is calculated. The calculation formula of the weighted value of the flow of people is: ;in, is the weighted value of the flow of people in the jth sub-area in the i-th time interval, k represents the kth time node in the i-th time interval, s is the number of time nodes in the i-th time interval, is the passenger flow weight factor of the jth sub-area at the kth time node in the i-th time interval, is the flow of people in the jth sub-area at the kth time node in the i-th time interval; S13, calculating the density of human traffic, the calculation formula of which is: ;in, is the flow density of people in the jth sub-area during the i-th time interval, is the area of ​​the jth sub-region, the value of i is 1-n, and the value of j is 1-m.

3. The cold storage load prediction method based on the Internet of Things and artificial intelligence according to claim 2 is characterized in that: The equipment operation stability assessment in S1 includes the following specific steps: S14. Within each time interval i, for each sub-area j, count the number of equipment failures and the total equipment operating time within the sub-area; S15. Calculate the mean stable operation time of the equipment to evaluate the equipment operation stability. The calculation formula for the mean stable operation time of the equipment is: ;in, is the mean stable operation time of the equipment in the jth sub-area in the i-th time interval, is the total operation time of the equipment in the jth sub-area during the i-th time interval, is the number of equipment failures in the jth sub-area during the i-th time interval.

4. The cold storage load prediction method based on the Internet of Things and artificial intelligence according to claim 3 is characterized in that: The S2 comprises the following specific steps: S21, extract the mean of the human flow density and the stable operation time of the equipment in the m sub-areas within the n time interval, and convert the human flow density of the sub-area j within the time interval i into And the average stable operation time of the equipment Combine into feature vector ; S22, randomly select 4 eigenvectors from the set of all eigenvectors as the initial cluster centers of the cluster analysis algorithm, for each eigenvector , calculate its Euclidean distance to each initial cluster center, and assign it to the cluster corresponding to the initial cluster center with the smallest distance; S23, calculate the mean of all eigenvectors in a single cluster, update the initial cluster center and The reallocation is performed until a preset number of iterations is reached and then the operation is stopped to complete the cluster analysis. Each cluster represents an indoor space usage pattern, and the indoor space usage pattern includes a high-load-stable device class, a high-load-unstable device class, a low-load-stable device class, and a low-load-unstable device class.

5. The cold storage load prediction method based on the Internet of Things and artificial intelligence according to claim 4 is characterized in that: The S2 further includes: S24. In each time interval i, for each sub-area j, the actual probabilities of the four types of indoor space usage patterns are calculated respectively, and the actual probability distribution vector for the j-th sub-area in the time interval i is obtained. ,in, is the actual probability of the high-load-stable equipment class in the jth sub-area in the i-th time interval, is the actual probability of the high-load-unstable device class in the jth sub-area in the i-th time interval, is the actual probability of the low-load-stable equipment class in the jth sub-area in the i-th time interval, is the actual probability of the low-load-unstable equipment class in the jth sub-area in the i-th time interval; S25. Repeat the above steps until m sub-areas of n time intervals are traversed.

6. The cold storage load prediction method based on the Internet of Things and artificial intelligence according to claim 5 is characterized in that: The clustering analysis algorithm in S22 is a K-Means clustering algorithm, and K=4.

7. The cold storage load prediction method based on the Internet of Things and artificial intelligence according to claim 6 is characterized in that: The specific content of obtaining the expected probability distribution of indoor space usage pattern in S3 is as follows: for m sub-areas, obtain the pedestrian flow density, the average of the stable operation time of the equipment and the indoor space usage pattern in 3n time intervals within the first three collection cycles, combine the pedestrian flow density and the average of the stable operation time of the equipment in the 3n time intervals within the first three collection cycles into the form of feature vectors, and construct a machine learning model. The machine learning model takes the set of all groups of feature vectors as input, takes the expected probability distribution of the predicted indoor space usage pattern corresponding to each group of feature vectors as output, takes the expected probability distribution of the actual indoor space usage pattern corresponding to each group of feature vectors as the prediction target, takes minimizing the sum of the squares of the difference between the predicted expected probability distribution and the actual expected probability distribution as the training target, trains the machine learning model until the sum of the squares of the difference between the predicted expected probability distribution and the actual expected probability distribution reaches convergence, stops the training, and outputs the expected probability distribution vector of the predicted indoor space usage pattern. , the machine learning model is any one of a random forest model, a support vector machine or a neural network model.

8. The cold storage load prediction method based on the Internet of Things and artificial intelligence according to claim 7 is characterized in that: The specific formula for calculating the deviation value in S3 is: ； in, is the deviation between the expected probability distribution and the actual probability distribution of the indoor space usage pattern of the jth sub-area in the i-th time interval.

9. The cold storage load prediction method based on the Internet of Things and artificial intelligence according to claim 8 is characterized in that: The S4 comprises the following specific steps: S41, preset deviation value threshold T, when When , the historical average cooling load data of the j-th sub-area under the high load-stable equipment type, high load-unstable equipment type, low load-stable equipment type and low load-unstable equipment type modes are obtained respectively; S42, predicting the cold storage load of the j-th sub-area during the time interval i, the calculation formula for predicting the cold storage load is: ;in, is the predicted cooling load of the jth sub-area in the i-th time interval, is the historical average cooling load of the j-th sub-area under the high load-stable equipment mode, is the historical average cooling load of the j-th sub-area under the high load-unstable equipment type mode, is the historical average cooling load of the j-th sub-area under the low-load-stable equipment mode, is the historical average cooling load of the j-th sub-area under the low load-unstable equipment type mode.

10. A cold storage load prediction system based on the Internet of Things and artificial intelligence, which is implemented based on the cold storage load prediction method based on the Internet of Things and artificial intelligence described in any one of claims 1 to 9, characterized in that: The system includes: a data evaluation module, a cluster analysis module, a deviation calculation module, and a load prediction module; The data evaluation module is used to collect indoor human flow data and equipment operation data, and to evaluate human flow density and equipment operation stability; The cluster analysis module is used to perform cluster analysis on the crowd density and equipment operation stability evaluation results according to the indoor space usage pattern, and calculate the actual probability distribution of the indoor space usage pattern; The deviation calculation module is used to obtain the expected probability distribution of the indoor space usage pattern and calculate the deviation value with the actual probability distribution of the indoor space usage pattern; The load prediction module is used to predict the cooling load based on the actual probability distribution of the indoor space usage pattern when the deviation value is less than a preset deviation value threshold.

Images